{
  "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": 0.04491599,
      "auxiliary_loss_mlp": 0.02221579,
      "balance_loss_clip": 2.47289801,
      "balance_loss_mlp": 1.82561398,
      "epoch": 0.00012024289063909097,
      "flos": 24932483919360.0,
      "grad_norm": 40.163321662020806,
      "language_loss": 2.57927752,
      "learning_rate": 0.0,
      "loss": 1.89822352,
      "num_input_tokens_seen": 20375,
      "step": 1,
      "time_per_iteration": 16.226963996887207
    },
    {
      "auxiliary_loss_clip": 0.0300904,
      "auxiliary_loss_mlp": 0.01468188,
      "balance_loss_clip": 1.63999248,
      "balance_loss_mlp": 1.19372022,
      "epoch": 0.00024048578127818193,
      "flos": 30664624377600.0,
      "grad_norm": 54.95453609262513,
      "language_loss": 1.88825572,
      "learning_rate": 5.021476677069823e-07,
      "loss": 1.93302798,
      "num_input_tokens_seen": 39035,
      "step": 2,
      "time_per_iteration": 2.5960869789123535
    },
    {
      "auxiliary_loss_clip": 0.03015296,
      "auxiliary_loss_mlp": 0.01474571,
      "balance_loss_clip": 1.64831674,
      "balance_loss_mlp": 1.19075799,
      "epoch": 0.0003607286719172729,
      "flos": 19026227969280.0,
      "grad_norm": 40.1721642573067,
      "language_loss": 1.61505961,
      "learning_rate": 7.958852231401551e-07,
      "loss": 1.65995836,
      "num_input_tokens_seen": 57600,
      "step": 3,
      "time_per_iteration": 2.535675287246704
    },
    {
      "auxiliary_loss_clip": 0.02993917,
      "auxiliary_loss_mlp": 0.01471694,
      "balance_loss_clip": 1.64810443,
      "balance_loss_mlp": 1.21610916,
      "epoch": 0.00048097156255636386,
      "flos": 19316314206720.0,
      "grad_norm": 36.83340058166358,
      "language_loss": 1.64525616,
      "learning_rate": 1.0042953354139647e-06,
      "loss": 1.68991232,
      "num_input_tokens_seen": 76465,
      "step": 4,
      "time_per_iteration": 2.6027774810791016
    },
    {
      "auxiliary_loss_clip": 0.02987625,
      "auxiliary_loss_mlp": 0.0146254,
      "balance_loss_clip": 1.6403805,
      "balance_loss_mlp": 1.19169617,
      "epoch": 0.0006012144531954548,
      "flos": 13991264893440.0,
      "grad_norm": 54.983568980586625,
      "language_loss": 1.93322444,
      "learning_rate": 1.1659507774310057e-06,
      "loss": 1.97772598,
      "num_input_tokens_seen": 94350,
      "step": 5,
      "time_per_iteration": 2.77620267868042
    },
    {
      "auxiliary_loss_clip": 0.02991995,
      "auxiliary_loss_mlp": 0.01455131,
      "balance_loss_clip": 1.64246082,
      "balance_loss_mlp": 1.18085408,
      "epoch": 0.0007214573438345458,
      "flos": 23148988225920.0,
      "grad_norm": 44.92862293676176,
      "language_loss": 1.60884368,
      "learning_rate": 1.2980328908471373e-06,
      "loss": 1.65331507,
      "num_input_tokens_seen": 114595,
      "step": 6,
      "time_per_iteration": 2.878549814224243
    },
    {
      "auxiliary_loss_clip": 0.02899924,
      "auxiliary_loss_mlp": 0.0152134,
      "balance_loss_clip": 1.74713707,
      "balance_loss_mlp": 1.17191362,
      "epoch": 0.0008417002344736367,
      "flos": 67663246170240.0,
      "grad_norm": 4.5917703107897285,
      "language_loss": 0.81486475,
      "learning_rate": 1.4097067265369432e-06,
      "loss": 0.85907733,
      "num_input_tokens_seen": 179590,
      "step": 7,
      "time_per_iteration": 3.304527521133423
    },
    {
      "auxiliary_loss_clip": 0.02973682,
      "auxiliary_loss_mlp": 0.01441301,
      "balance_loss_clip": 1.64579487,
      "balance_loss_mlp": 1.17904043,
      "epoch": 0.0009619431251127277,
      "flos": 21281381504640.0,
      "grad_norm": 41.01975025988405,
      "language_loss": 1.58570433,
      "learning_rate": 1.506443003120947e-06,
      "loss": 1.6298542,
      "num_input_tokens_seen": 195090,
      "step": 8,
      "time_per_iteration": 2.810551404953003
    },
    {
      "auxiliary_loss_clip": 0.02993168,
      "auxiliary_loss_mlp": 0.01467015,
      "balance_loss_clip": 1.64896607,
      "balance_loss_mlp": 1.19788778,
      "epoch": 0.0010821860157518186,
      "flos": 23331342597120.0,
      "grad_norm": 17.493215014814783,
      "language_loss": 1.47886181,
      "learning_rate": 1.5917704462803102e-06,
      "loss": 1.52346373,
      "num_input_tokens_seen": 211635,
      "step": 9,
      "time_per_iteration": 2.7871103286743164
    },
    {
      "auxiliary_loss_clip": 0.02977469,
      "auxiliary_loss_mlp": 0.01457293,
      "balance_loss_clip": 1.64932227,
      "balance_loss_mlp": 1.19140863,
      "epoch": 0.0012024289063909096,
      "flos": 17010166337280.0,
      "grad_norm": 13.220119208854346,
      "language_loss": 1.52825212,
      "learning_rate": 1.6680984451379884e-06,
      "loss": 1.57259977,
      "num_input_tokens_seen": 224705,
      "step": 10,
      "time_per_iteration": 2.7486579418182373
    },
    {
      "auxiliary_loss_clip": 0.02995587,
      "auxiliary_loss_mlp": 0.01466967,
      "balance_loss_clip": 1.65137935,
      "balance_loss_mlp": 1.18849444,
      "epoch": 0.0013226717970300007,
      "flos": 21288133261440.0,
      "grad_norm": 19.089142408113418,
      "language_loss": 1.32471442,
      "learning_rate": 1.7371455188905097e-06,
      "loss": 1.36933994,
      "num_input_tokens_seen": 244635,
      "step": 11,
      "time_per_iteration": 2.8070144653320312
    },
    {
      "auxiliary_loss_clip": 0.03003385,
      "auxiliary_loss_mlp": 0.01460643,
      "balance_loss_clip": 1.64764857,
      "balance_loss_mlp": 1.17778349,
      "epoch": 0.0014429146876690916,
      "flos": 27237884935680.0,
      "grad_norm": 10.54493964200517,
      "language_loss": 1.2536633,
      "learning_rate": 1.8001805585541196e-06,
      "loss": 1.2983036,
      "num_input_tokens_seen": 265765,
      "step": 12,
      "time_per_iteration": 2.863173246383667
    },
    {
      "auxiliary_loss_clip": 0.02981205,
      "auxiliary_loss_mlp": 0.01472424,
      "balance_loss_clip": 1.65053058,
      "balance_loss_mlp": 1.19871926,
      "epoch": 0.0015631575783081825,
      "flos": 19062174504960.0,
      "grad_norm": 6.646512759096275,
      "language_loss": 1.29329753,
      "learning_rate": 1.8581671739548328e-06,
      "loss": 1.337834,
      "num_input_tokens_seen": 283500,
      "step": 13,
      "time_per_iteration": 2.8205015659332275
    },
    {
      "auxiliary_loss_clip": 0.02961987,
      "auxiliary_loss_mlp": 0.01475863,
      "balance_loss_clip": 1.64167118,
      "balance_loss_mlp": 1.19815302,
      "epoch": 0.0016834004689472734,
      "flos": 48139473985920.0,
      "grad_norm": 6.278591479558552,
      "language_loss": 1.13597274,
      "learning_rate": 1.9118543942439254e-06,
      "loss": 1.18035126,
      "num_input_tokens_seen": 305685,
      "step": 14,
      "time_per_iteration": 3.964249849319458
    },
    {
      "auxiliary_loss_clip": 0.02981122,
      "auxiliary_loss_mlp": 0.01449155,
      "balance_loss_clip": 1.64573061,
      "balance_loss_mlp": 1.17754877,
      "epoch": 0.0018036433595863645,
      "flos": 34970026314240.0,
      "grad_norm": 5.484670188989572,
      "language_loss": 1.12807953,
      "learning_rate": 1.961836000571161e-06,
      "loss": 1.17238235,
      "num_input_tokens_seen": 327340,
      "step": 15,
      "time_per_iteration": 3.9112637042999268
    },
    {
      "auxiliary_loss_clip": 0.0280488,
      "auxiliary_loss_mlp": 0.0147059,
      "balance_loss_clip": 1.72419047,
      "balance_loss_mlp": 1.13337076,
      "epoch": 0.0019238862502254555,
      "flos": 59768284440960.0,
      "grad_norm": 3.7866858979891935,
      "language_loss": 0.64658284,
      "learning_rate": 2.0085906708279293e-06,
      "loss": 0.68933755,
      "num_input_tokens_seen": 382710,
      "step": 16,
      "time_per_iteration": 3.3153250217437744
    },
    {
      "auxiliary_loss_clip": 0.02950809,
      "auxiliary_loss_mlp": 0.0143603,
      "balance_loss_clip": 1.64240313,
      "balance_loss_mlp": 1.16957331,
      "epoch": 0.0020441291408645466,
      "flos": 20814543417600.0,
      "grad_norm": 4.3541656018488855,
      "language_loss": 1.15949893,
      "learning_rate": 2.0525099325728135e-06,
      "loss": 1.20336747,
      "num_input_tokens_seen": 400890,
      "step": 17,
      "time_per_iteration": 2.841297149658203
    },
    {
      "auxiliary_loss_clip": 0.02776008,
      "auxiliary_loss_mlp": 0.0145669,
      "balance_loss_clip": 1.71820652,
      "balance_loss_mlp": 1.12099707,
      "epoch": 0.0021643720315036373,
      "flos": 63857001582720.0,
      "grad_norm": 3.5308769842636045,
      "language_loss": 0.7215389,
      "learning_rate": 2.0939181139872922e-06,
      "loss": 0.76386595,
      "num_input_tokens_seen": 462605,
      "step": 18,
      "time_per_iteration": 3.239359140396118
    },
    {
      "auxiliary_loss_clip": 0.02947094,
      "auxiliary_loss_mlp": 0.01426851,
      "balance_loss_clip": 1.64437175,
      "balance_loss_mlp": 1.17450881,
      "epoch": 0.0022846149221427284,
      "flos": 31284981192960.0,
      "grad_norm": 5.057843935804984,
      "language_loss": 1.01627076,
      "learning_rate": 2.1330868934640175e-06,
      "loss": 1.06001019,
      "num_input_tokens_seen": 483280,
      "step": 19,
      "time_per_iteration": 2.8798208236694336
    },
    {
      "auxiliary_loss_clip": 0.02738828,
      "auxiliary_loss_mlp": 0.01431733,
      "balance_loss_clip": 1.7070657,
      "balance_loss_mlp": 1.10214293,
      "epoch": 0.002404857812781819,
      "flos": 51083648161920.0,
      "grad_norm": 3.8932250024616897,
      "language_loss": 0.76434153,
      "learning_rate": 2.170246112844971e-06,
      "loss": 0.80604714,
      "num_input_tokens_seen": 537620,
      "step": 20,
      "time_per_iteration": 3.01880145072937
    },
    {
      "auxiliary_loss_clip": 0.0287993,
      "auxiliary_loss_mlp": 0.01429534,
      "balance_loss_clip": 1.62626266,
      "balance_loss_mlp": 1.17108786,
      "epoch": 0.0025251007034209102,
      "flos": 15815347309440.0,
      "grad_norm": 4.528375664338746,
      "language_loss": 1.01726222,
      "learning_rate": 2.2055919496770983e-06,
      "loss": 1.06035674,
      "num_input_tokens_seen": 555760,
      "step": 21,
      "time_per_iteration": 2.7465481758117676
    },
    {
      "auxiliary_loss_clip": 0.02882308,
      "auxiliary_loss_mlp": 0.0142223,
      "balance_loss_clip": 1.62947464,
      "balance_loss_mlp": 1.16492879,
      "epoch": 0.0026453435940600014,
      "flos": 37851857458560.0,
      "grad_norm": 3.6890059988262593,
      "language_loss": 0.89403272,
      "learning_rate": 2.2392931865974923e-06,
      "loss": 0.93707812,
      "num_input_tokens_seen": 578450,
      "step": 22,
      "time_per_iteration": 3.0133705139160156
    },
    {
      "auxiliary_loss_clip": 0.02834745,
      "auxiliary_loss_mlp": 0.01427875,
      "balance_loss_clip": 1.61556959,
      "balance_loss_mlp": 1.16961956,
      "epoch": 0.002765586484699092,
      "flos": 21141976821120.0,
      "grad_norm": 4.089971394814634,
      "language_loss": 1.01706481,
      "learning_rate": 2.271496085962064e-06,
      "loss": 1.05969119,
      "num_input_tokens_seen": 596145,
      "step": 23,
      "time_per_iteration": 2.8292930126190186
    },
    {
      "auxiliary_loss_clip": 0.02820214,
      "auxiliary_loss_mlp": 0.01408314,
      "balance_loss_clip": 1.61054301,
      "balance_loss_mlp": 1.16074002,
      "epoch": 0.002885829375338183,
      "flos": 20667381396480.0,
      "grad_norm": 4.491538893383789,
      "language_loss": 1.02586675,
      "learning_rate": 2.3023282262611022e-06,
      "loss": 1.06815195,
      "num_input_tokens_seen": 614920,
      "step": 24,
      "time_per_iteration": 2.7971348762512207
    },
    {
      "auxiliary_loss_clip": 0.02805424,
      "auxiliary_loss_mlp": 0.01400227,
      "balance_loss_clip": 1.60972285,
      "balance_loss_mlp": 1.16495526,
      "epoch": 0.003006072265977274,
      "flos": 34823869873920.0,
      "grad_norm": 3.331871990777231,
      "language_loss": 0.92589688,
      "learning_rate": 2.3319015548620114e-06,
      "loss": 0.96795332,
      "num_input_tokens_seen": 636060,
      "step": 25,
      "time_per_iteration": 2.8684446811676025
    },
    {
      "auxiliary_loss_clip": 0.02781421,
      "auxiliary_loss_mlp": 0.01396556,
      "balance_loss_clip": 1.60625172,
      "balance_loss_mlp": 1.16614866,
      "epoch": 0.003126315156616365,
      "flos": 24422021118720.0,
      "grad_norm": 2.8577445218707056,
      "language_loss": 0.92953634,
      "learning_rate": 2.3603148416618152e-06,
      "loss": 0.9713161,
      "num_input_tokens_seen": 655575,
      "step": 26,
      "time_per_iteration": 2.885298728942871
    },
    {
      "auxiliary_loss_clip": 0.02802371,
      "auxiliary_loss_mlp": 0.0138588,
      "balance_loss_clip": 1.6114316,
      "balance_loss_mlp": 1.1516571,
      "epoch": 0.003246558047255456,
      "flos": 23622326674560.0,
      "grad_norm": 2.3972637729499553,
      "language_loss": 1.00903463,
      "learning_rate": 2.3876556694204647e-06,
      "loss": 1.05091715,
      "num_input_tokens_seen": 675730,
      "step": 27,
      "time_per_iteration": 2.774613618850708
    },
    {
      "auxiliary_loss_clip": 0.02760736,
      "auxiliary_loss_mlp": 0.0139506,
      "balance_loss_clip": 1.59588456,
      "balance_loss_mlp": 1.15530634,
      "epoch": 0.003366800937894547,
      "flos": 17820275725440.0,
      "grad_norm": 2.656257689039716,
      "language_loss": 0.90690333,
      "learning_rate": 2.414002061950908e-06,
      "loss": 0.94846123,
      "num_input_tokens_seen": 694605,
      "step": 28,
      "time_per_iteration": 2.8128139972686768
    },
    {
      "auxiliary_loss_clip": 0.02749347,
      "auxiliary_loss_mlp": 0.01396387,
      "balance_loss_clip": 1.59300232,
      "balance_loss_mlp": 1.16617036,
      "epoch": 0.003487043828533638,
      "flos": 24426115269120.0,
      "grad_norm": 2.4008841500620517,
      "language_loss": 0.9993763,
      "learning_rate": 2.4394238264681557e-06,
      "loss": 1.04083359,
      "num_input_tokens_seen": 714340,
      "step": 29,
      "time_per_iteration": 2.8445684909820557
    },
    {
      "auxiliary_loss_clip": 0.02710701,
      "auxiliary_loss_mlp": 0.01342423,
      "balance_loss_clip": 1.58956349,
      "balance_loss_mlp": 1.10953557,
      "epoch": 0.003607286719172729,
      "flos": 26140311002880.0,
      "grad_norm": 2.0678962121924775,
      "language_loss": 0.9955495,
      "learning_rate": 2.4639836682781433e-06,
      "loss": 1.03608084,
      "num_input_tokens_seen": 734470,
      "step": 30,
      "time_per_iteration": 2.8105156421661377
    },
    {
      "auxiliary_loss_clip": 0.02720673,
      "auxiliary_loss_mlp": 0.01348615,
      "balance_loss_clip": 1.59222877,
      "balance_loss_mlp": 1.10218537,
      "epoch": 0.00372752960981182,
      "flos": 20593082113920.0,
      "grad_norm": 2.7295349189738256,
      "language_loss": 1.0017463,
      "learning_rate": 2.487738122623307e-06,
      "loss": 1.04243922,
      "num_input_tokens_seen": 753380,
      "step": 31,
      "time_per_iteration": 2.8798696994781494
    },
    {
      "auxiliary_loss_clip": 0.02662504,
      "auxiliary_loss_mlp": 0.01343031,
      "balance_loss_clip": 1.57569802,
      "balance_loss_mlp": 1.10814071,
      "epoch": 0.003847772500450911,
      "flos": 22674608282880.0,
      "grad_norm": 2.513649326653245,
      "language_loss": 0.98945636,
      "learning_rate": 2.510738338534912e-06,
      "loss": 1.02951169,
      "num_input_tokens_seen": 772105,
      "step": 32,
      "time_per_iteration": 2.8228604793548584
    },
    {
      "auxiliary_loss_clip": 0.02588818,
      "auxiliary_loss_mlp": 0.01371655,
      "balance_loss_clip": 1.55665135,
      "balance_loss_mlp": 1.13800478,
      "epoch": 0.003968015391090002,
      "flos": 17967796882560.0,
      "grad_norm": 2.70018168897321,
      "language_loss": 1.02677512,
      "learning_rate": 2.5330307420306648e-06,
      "loss": 1.06637979,
      "num_input_tokens_seen": 788955,
      "step": 33,
      "time_per_iteration": 2.7979047298431396
    },
    {
      "auxiliary_loss_clip": 0.02564916,
      "auxiliary_loss_mlp": 0.01320494,
      "balance_loss_clip": 1.55383253,
      "balance_loss_mlp": 1.10687065,
      "epoch": 0.004088258281729093,
      "flos": 27304103658240.0,
      "grad_norm": 3.738118320304094,
      "language_loss": 0.87965846,
      "learning_rate": 2.554657600279796e-06,
      "loss": 0.91851258,
      "num_input_tokens_seen": 810230,
      "step": 34,
      "time_per_iteration": 2.8355112075805664
    },
    {
      "auxiliary_loss_clip": 0.02527101,
      "auxiliary_loss_mlp": 0.01326616,
      "balance_loss_clip": 1.54346848,
      "balance_loss_mlp": 1.09983253,
      "epoch": 0.004208501172368184,
      "flos": 23258587599360.0,
      "grad_norm": 2.1664910393617194,
      "language_loss": 1.03335226,
      "learning_rate": 2.5756575039679493e-06,
      "loss": 1.0718894,
      "num_input_tokens_seen": 829780,
      "step": 35,
      "time_per_iteration": 2.7929413318634033
    },
    {
      "auxiliary_loss_clip": 0.02498675,
      "auxiliary_loss_mlp": 0.01322131,
      "balance_loss_clip": 1.53181314,
      "balance_loss_mlp": 1.1127044,
      "epoch": 0.0043287440630072746,
      "flos": 17312104062720.0,
      "grad_norm": 1.96712159981555,
      "language_loss": 0.95004499,
      "learning_rate": 2.5960657816942747e-06,
      "loss": 0.98825306,
      "num_input_tokens_seen": 848695,
      "step": 36,
      "time_per_iteration": 2.852844476699829
    },
    {
      "auxiliary_loss_clip": 0.02189552,
      "auxiliary_loss_mlp": 0.01334769,
      "balance_loss_clip": 1.52825308,
      "balance_loss_mlp": 1.0662148,
      "epoch": 0.004448986953646365,
      "flos": 53092491160320.0,
      "grad_norm": 1.387905975389154,
      "language_loss": 0.61020124,
      "learning_rate": 2.6159148575788668e-06,
      "loss": 0.64544445,
      "num_input_tokens_seen": 906730,
      "step": 37,
      "time_per_iteration": 3.325557231903076
    },
    {
      "auxiliary_loss_clip": 0.02461854,
      "auxiliary_loss_mlp": 0.01286017,
      "balance_loss_clip": 1.51970005,
      "balance_loss_mlp": 1.07496929,
      "epoch": 0.004569229844285457,
      "flos": 13444165866240.0,
      "grad_norm": 2.4302642910780907,
      "language_loss": 0.98947513,
      "learning_rate": 2.635234561171e-06,
      "loss": 1.02695382,
      "num_input_tokens_seen": 925125,
      "step": 38,
      "time_per_iteration": 2.8126745223999023
    },
    {
      "auxiliary_loss_clip": 0.02429918,
      "auxiliary_loss_mlp": 0.01287115,
      "balance_loss_clip": 1.51529503,
      "balance_loss_mlp": 1.09523535,
      "epoch": 0.0046894727349245475,
      "flos": 16209609966720.0,
      "grad_norm": 3.7219891565306384,
      "language_loss": 0.9411552,
      "learning_rate": 2.6540523970949877e-06,
      "loss": 0.97832555,
      "num_input_tokens_seen": 939970,
      "step": 39,
      "time_per_iteration": 2.777009963989258
    },
    {
      "auxiliary_loss_clip": 0.02430649,
      "auxiliary_loss_mlp": 0.01286289,
      "balance_loss_clip": 1.51683307,
      "balance_loss_mlp": 1.08239388,
      "epoch": 0.004809715625563638,
      "flos": 23914244505600.0,
      "grad_norm": 3.145172711394561,
      "language_loss": 0.92502004,
      "learning_rate": 2.6723937805519533e-06,
      "loss": 0.96218938,
      "num_input_tokens_seen": 957470,
      "step": 40,
      "time_per_iteration": 3.8056347370147705
    },
    {
      "auxiliary_loss_clip": 0.02380222,
      "auxiliary_loss_mlp": 0.01288794,
      "balance_loss_clip": 1.50057077,
      "balance_loss_mlp": 1.09376717,
      "epoch": 0.00492995851620273,
      "flos": 20773030273920.0,
      "grad_norm": 2.3267289374857247,
      "language_loss": 0.93226171,
      "learning_rate": 2.690282243737839e-06,
      "loss": 0.96895188,
      "num_input_tokens_seen": 976405,
      "step": 41,
      "time_per_iteration": 3.7384727001190186
    },
    {
      "auxiliary_loss_clip": 0.0234594,
      "auxiliary_loss_mlp": 0.01262382,
      "balance_loss_clip": 1.48543429,
      "balance_loss_mlp": 1.07202816,
      "epoch": 0.0050502014068418205,
      "flos": 20338655103360.0,
      "grad_norm": 2.848675288888808,
      "language_loss": 0.99460036,
      "learning_rate": 2.7077396173840807e-06,
      "loss": 1.03068352,
      "num_input_tokens_seen": 994690,
      "step": 42,
      "time_per_iteration": 2.7416727542877197
    },
    {
      "auxiliary_loss_clip": 0.02346066,
      "auxiliary_loss_mlp": 0.01275769,
      "balance_loss_clip": 1.48910332,
      "balance_loss_mlp": 1.07902563,
      "epoch": 0.005170444297480911,
      "flos": 25994872834560.0,
      "grad_norm": 2.575373489594992,
      "language_loss": 0.92822099,
      "learning_rate": 2.7247861909342594e-06,
      "loss": 0.96443933,
      "num_input_tokens_seen": 1015615,
      "step": 43,
      "time_per_iteration": 2.872300148010254
    },
    {
      "auxiliary_loss_clip": 0.0232423,
      "auxiliary_loss_mlp": 0.01282572,
      "balance_loss_clip": 1.48540139,
      "balance_loss_mlp": 1.08830881,
      "epoch": 0.005290687188120003,
      "flos": 20954055841920.0,
      "grad_norm": 2.244574759184508,
      "language_loss": 0.83192968,
      "learning_rate": 2.7414408543044743e-06,
      "loss": 0.86799777,
      "num_input_tokens_seen": 1031255,
      "step": 44,
      "time_per_iteration": 2.755058765411377
    },
    {
      "auxiliary_loss_clip": 0.02304633,
      "auxiliary_loss_mlp": 0.01242318,
      "balance_loss_clip": 1.47259796,
      "balance_loss_mlp": 1.05291784,
      "epoch": 0.005410930078759093,
      "flos": 15851401585920.0,
      "grad_norm": 4.630753036823466,
      "language_loss": 0.79513079,
      "learning_rate": 2.7577212237113157e-06,
      "loss": 0.83060026,
      "num_input_tokens_seen": 1048295,
      "step": 45,
      "time_per_iteration": 2.8247509002685547
    },
    {
      "auxiliary_loss_clip": 0.02281564,
      "auxiliary_loss_mlp": 0.01249999,
      "balance_loss_clip": 1.46995521,
      "balance_loss_mlp": 1.06603527,
      "epoch": 0.005531172969398184,
      "flos": 21104988791040.0,
      "grad_norm": 1.9719248686857345,
      "language_loss": 1.04239273,
      "learning_rate": 2.7736437536690466e-06,
      "loss": 1.07770824,
      "num_input_tokens_seen": 1067925,
      "step": 46,
      "time_per_iteration": 2.80381178855896
    },
    {
      "auxiliary_loss_clip": 0.02263206,
      "auxiliary_loss_mlp": 0.0123503,
      "balance_loss_clip": 1.46415186,
      "balance_loss_mlp": 1.06117487,
      "epoch": 0.005651415860037276,
      "flos": 20844887431680.0,
      "grad_norm": 1.9275815132256588,
      "language_loss": 1.07867312,
      "learning_rate": 2.789223836941131e-06,
      "loss": 1.11365545,
      "num_input_tokens_seen": 1088060,
      "step": 47,
      "time_per_iteration": 2.785576105117798
    },
    {
      "auxiliary_loss_clip": 0.02229983,
      "auxiliary_loss_mlp": 0.01246862,
      "balance_loss_clip": 1.4529804,
      "balance_loss_mlp": 1.07224441,
      "epoch": 0.005771658750676366,
      "flos": 13260195383040.0,
      "grad_norm": 2.6716556385447974,
      "language_loss": 1.08693254,
      "learning_rate": 2.8044758939680847e-06,
      "loss": 1.121701,
      "num_input_tokens_seen": 1104130,
      "step": 48,
      "time_per_iteration": 2.769165515899658
    },
    {
      "auxiliary_loss_clip": 0.02213242,
      "auxiliary_loss_mlp": 0.01251453,
      "balance_loss_clip": 1.45413268,
      "balance_loss_mlp": 1.07492805,
      "epoch": 0.005891901641315457,
      "flos": 24425396997120.0,
      "grad_norm": 3.313589062545653,
      "language_loss": 1.02070785,
      "learning_rate": 2.8194134530738863e-06,
      "loss": 1.05535483,
      "num_input_tokens_seen": 1122900,
      "step": 49,
      "time_per_iteration": 2.8311607837677
    },
    {
      "auxiliary_loss_clip": 0.02203759,
      "auxiliary_loss_mlp": 0.01247635,
      "balance_loss_clip": 1.44937098,
      "balance_loss_mlp": 1.08465219,
      "epoch": 0.006012144531954548,
      "flos": 23076197314560.0,
      "grad_norm": 5.683830445648056,
      "language_loss": 0.90051711,
      "learning_rate": 2.834049222568994e-06,
      "loss": 0.93503106,
      "num_input_tokens_seen": 1140250,
      "step": 50,
      "time_per_iteration": 2.8686678409576416
    },
    {
      "auxiliary_loss_clip": 0.02188846,
      "auxiliary_loss_mlp": 0.01235823,
      "balance_loss_clip": 1.44044685,
      "balance_loss_mlp": 1.07551026,
      "epoch": 0.006132387422593639,
      "flos": 22528775064960.0,
      "grad_norm": 2.426369706199979,
      "language_loss": 0.92602342,
      "learning_rate": 2.848395155712969e-06,
      "loss": 0.96027017,
      "num_input_tokens_seen": 1160470,
      "step": 51,
      "time_per_iteration": 2.837876796722412
    },
    {
      "auxiliary_loss_clip": 0.02185804,
      "auxiliary_loss_mlp": 0.01217172,
      "balance_loss_clip": 1.4457972,
      "balance_loss_mlp": 1.05142355,
      "epoch": 0.00625263031323273,
      "flos": 27628340751360.0,
      "grad_norm": 2.296299873823433,
      "language_loss": 0.97654855,
      "learning_rate": 2.8624625093687977e-06,
      "loss": 1.01057827,
      "num_input_tokens_seen": 1177605,
      "step": 52,
      "time_per_iteration": 2.8265416622161865
    },
    {
      "auxiliary_loss_clip": 0.02169193,
      "auxiliary_loss_mlp": 0.01228689,
      "balance_loss_clip": 1.4359324,
      "balance_loss_mlp": 1.07667327,
      "epoch": 0.006372873203871821,
      "flos": 23110671392640.0,
      "grad_norm": 2.6102562872732027,
      "language_loss": 0.89007723,
      "learning_rate": 2.876261897070029e-06,
      "loss": 0.92405605,
      "num_input_tokens_seen": 1197735,
      "step": 53,
      "time_per_iteration": 2.861121892929077
    },
    {
      "auxiliary_loss_clip": 0.02158876,
      "auxiliary_loss_mlp": 0.01232933,
      "balance_loss_clip": 1.43290758,
      "balance_loss_mlp": 1.07900953,
      "epoch": 0.006493116094510912,
      "flos": 22856028900480.0,
      "grad_norm": 2.504731415245237,
      "language_loss": 0.92582208,
      "learning_rate": 2.889803337127447e-06,
      "loss": 0.95974016,
      "num_input_tokens_seen": 1216335,
      "step": 54,
      "time_per_iteration": 2.8140838146209717
    },
    {
      "auxiliary_loss_clip": 0.02150208,
      "auxiliary_loss_mlp": 0.0123312,
      "balance_loss_clip": 1.4285928,
      "balance_loss_mlp": 1.06822944,
      "epoch": 0.006613358985150003,
      "flos": 23071708114560.0,
      "grad_norm": 2.42290319628069,
      "language_loss": 0.84639096,
      "learning_rate": 2.903096296321516e-06,
      "loss": 0.88022423,
      "num_input_tokens_seen": 1234480,
      "step": 55,
      "time_per_iteration": 2.856668472290039
    },
    {
      "auxiliary_loss_clip": 0.02123154,
      "auxiliary_loss_mlp": 0.0122991,
      "balance_loss_clip": 1.42408085,
      "balance_loss_mlp": 1.08113694,
      "epoch": 0.006733601875789094,
      "flos": 26537662229760.0,
      "grad_norm": 2.0248832025360497,
      "language_loss": 0.91771901,
      "learning_rate": 2.9161497296578907e-06,
      "loss": 0.95124966,
      "num_input_tokens_seen": 1253870,
      "step": 56,
      "time_per_iteration": 2.8442752361297607
    },
    {
      "auxiliary_loss_clip": 0.02121915,
      "auxiliary_loss_mlp": 0.01212995,
      "balance_loss_clip": 1.42322659,
      "balance_loss_mlp": 1.0646987,
      "epoch": 0.006853844766428185,
      "flos": 15523178083200.0,
      "grad_norm": 3.5954849599607535,
      "language_loss": 0.85932827,
      "learning_rate": 2.928972116604173e-06,
      "loss": 0.89267743,
      "num_input_tokens_seen": 1270145,
      "step": 57,
      "time_per_iteration": 2.8010828495025635
    },
    {
      "auxiliary_loss_clip": 0.02097864,
      "auxiliary_loss_mlp": 0.01217619,
      "balance_loss_clip": 1.41159272,
      "balance_loss_mlp": 1.0836277,
      "epoch": 0.006974087657067276,
      "flos": 24243760897920.0,
      "grad_norm": 2.6591121281541072,
      "language_loss": 1.02064562,
      "learning_rate": 2.9415714941751377e-06,
      "loss": 1.05380046,
      "num_input_tokens_seen": 1291365,
      "step": 58,
      "time_per_iteration": 2.80631947517395
    },
    {
      "auxiliary_loss_clip": 0.02089877,
      "auxiliary_loss_mlp": 0.01228649,
      "balance_loss_clip": 1.41032386,
      "balance_loss_mlp": 1.08016181,
      "epoch": 0.007094330547706367,
      "flos": 25772513690880.0,
      "grad_norm": 2.5126778207248255,
      "language_loss": 0.93590796,
      "learning_rate": 2.9539554871897396e-06,
      "loss": 0.96909326,
      "num_input_tokens_seen": 1311535,
      "step": 59,
      "time_per_iteration": 2.8197238445281982
    },
    {
      "auxiliary_loss_clip": 0.02074519,
      "auxiliary_loss_mlp": 0.01213149,
      "balance_loss_clip": 1.4071002,
      "balance_loss_mlp": 1.07181454,
      "epoch": 0.007214573438345458,
      "flos": 21319015979520.0,
      "grad_norm": 2.1657645678408883,
      "language_loss": 0.9750675,
      "learning_rate": 2.9661313359851253e-06,
      "loss": 1.00794411,
      "num_input_tokens_seen": 1329420,
      "step": 60,
      "time_per_iteration": 2.7548983097076416
    },
    {
      "auxiliary_loss_clip": 0.0206097,
      "auxiliary_loss_mlp": 0.01208096,
      "balance_loss_clip": 1.40042186,
      "balance_loss_mlp": 1.06666601,
      "epoch": 0.007334816328984549,
      "flos": 24937088192640.0,
      "grad_norm": 2.447107206499474,
      "language_loss": 0.93989247,
      "learning_rate": 2.978105921839922e-06,
      "loss": 0.97258312,
      "num_input_tokens_seen": 1349965,
      "step": 61,
      "time_per_iteration": 2.8515472412109375
    },
    {
      "auxiliary_loss_clip": 0.02045748,
      "auxiliary_loss_mlp": 0.01193601,
      "balance_loss_clip": 1.39928341,
      "balance_loss_mlp": 1.06361508,
      "epoch": 0.00745505921962364,
      "flos": 18510586277760.0,
      "grad_norm": 2.2628650062593993,
      "language_loss": 0.72203124,
      "learning_rate": 2.9898857903302893e-06,
      "loss": 0.75442475,
      "num_input_tokens_seen": 1368915,
      "step": 62,
      "time_per_iteration": 2.8011655807495117
    },
    {
      "auxiliary_loss_clip": 0.02036683,
      "auxiliary_loss_mlp": 0.01200397,
      "balance_loss_clip": 1.39645278,
      "balance_loss_mlp": 1.06335354,
      "epoch": 0.007575302110262731,
      "flos": 18477656484480.0,
      "grad_norm": 3.1766018259275857,
      "language_loss": 0.88179195,
      "learning_rate": 3.001477172817253e-06,
      "loss": 0.91416276,
      "num_input_tokens_seen": 1386805,
      "step": 63,
      "time_per_iteration": 2.7530806064605713
    },
    {
      "auxiliary_loss_clip": 0.02040345,
      "auxiliary_loss_mlp": 0.01202791,
      "balance_loss_clip": 1.39826393,
      "balance_loss_mlp": 1.07423592,
      "epoch": 0.007695545000901822,
      "flos": 24973178382720.0,
      "grad_norm": 2.4657284504653796,
      "language_loss": 0.96154374,
      "learning_rate": 3.012886006241894e-06,
      "loss": 0.99397498,
      "num_input_tokens_seen": 1406190,
      "step": 64,
      "time_per_iteration": 2.8475453853607178
    },
    {
      "auxiliary_loss_clip": 0.02012405,
      "auxiliary_loss_mlp": 0.0119366,
      "balance_loss_clip": 1.38691652,
      "balance_loss_mlp": 1.07035041,
      "epoch": 0.007815787891540913,
      "flos": 21324223451520.0,
      "grad_norm": 1.9991685820894902,
      "language_loss": 0.88282537,
      "learning_rate": 3.0241179513858383e-06,
      "loss": 0.914886,
      "num_input_tokens_seen": 1425500,
      "step": 65,
      "time_per_iteration": 2.7386250495910645
    },
    {
      "auxiliary_loss_clip": 0.02003539,
      "auxiliary_loss_mlp": 0.01179024,
      "balance_loss_clip": 1.37830842,
      "balance_loss_mlp": 1.05647659,
      "epoch": 0.007936030782180003,
      "flos": 21575777374080.0,
      "grad_norm": 2.229825165253108,
      "language_loss": 0.87873471,
      "learning_rate": 3.035178409737647e-06,
      "loss": 0.91056037,
      "num_input_tokens_seen": 1442950,
      "step": 66,
      "time_per_iteration": 3.7674920558929443
    },
    {
      "auxiliary_loss_clip": 0.0197785,
      "auxiliary_loss_mlp": 0.01179481,
      "balance_loss_clip": 1.36962128,
      "balance_loss_mlp": 1.06475437,
      "epoch": 0.008056273672819095,
      "flos": 20120785159680.0,
      "grad_norm": 2.0235569430546168,
      "language_loss": 0.88794947,
      "learning_rate": 3.046072539090907e-06,
      "loss": 0.91952282,
      "num_input_tokens_seen": 1460915,
      "step": 67,
      "time_per_iteration": 4.620994329452515
    },
    {
      "auxiliary_loss_clip": 0.01975824,
      "auxiliary_loss_mlp": 0.01176868,
      "balance_loss_clip": 1.36495888,
      "balance_loss_mlp": 1.0657177,
      "epoch": 0.008176516563458186,
      "flos": 18333116156160.0,
      "grad_norm": 2.5070186835816815,
      "language_loss": 1.04548109,
      "learning_rate": 3.056805267986779e-06,
      "loss": 1.07700801,
      "num_input_tokens_seen": 1478385,
      "step": 68,
      "time_per_iteration": 2.800161361694336
    },
    {
      "auxiliary_loss_clip": 0.01950921,
      "auxiliary_loss_mlp": 0.01173359,
      "balance_loss_clip": 1.35912144,
      "balance_loss_mlp": 1.06692863,
      "epoch": 0.008296759454097276,
      "flos": 21872076664320.0,
      "grad_norm": 2.3185109693254993,
      "language_loss": 0.95364261,
      "learning_rate": 3.0673813091022194e-06,
      "loss": 0.98488545,
      "num_input_tokens_seen": 1497605,
      "step": 69,
      "time_per_iteration": 2.8168087005615234
    },
    {
      "auxiliary_loss_clip": 0.01694084,
      "auxiliary_loss_mlp": 0.01205752,
      "balance_loss_clip": 1.29890954,
      "balance_loss_mlp": 1.06842256,
      "epoch": 0.008417002344736368,
      "flos": 63408228036480.0,
      "grad_norm": 1.3499245316351247,
      "language_loss": 0.62147617,
      "learning_rate": 3.0778051716749317e-06,
      "loss": 0.65047455,
      "num_input_tokens_seen": 1561150,
      "step": 70,
      "time_per_iteration": 3.4276390075683594
    },
    {
      "auxiliary_loss_clip": 0.01919551,
      "auxiliary_loss_mlp": 0.01180379,
      "balance_loss_clip": 1.3435421,
      "balance_loss_mlp": 1.07633328,
      "epoch": 0.008537245235375458,
      "flos": 22966454286720.0,
      "grad_norm": 2.9357889082858977,
      "language_loss": 0.90367472,
      "learning_rate": 3.0880811730470094e-06,
      "loss": 0.93467402,
      "num_input_tokens_seen": 1580605,
      "step": 71,
      "time_per_iteration": 2.791699171066284
    },
    {
      "auxiliary_loss_clip": 0.01664234,
      "auxiliary_loss_mlp": 0.01175268,
      "balance_loss_clip": 1.28190291,
      "balance_loss_mlp": 1.04709399,
      "epoch": 0.008657488126014549,
      "flos": 61984046712960.0,
      "grad_norm": 1.1528580608414556,
      "language_loss": 0.58654857,
      "learning_rate": 3.098213449401257e-06,
      "loss": 0.6149435,
      "num_input_tokens_seen": 1647535,
      "step": 72,
      "time_per_iteration": 3.2722995281219482
    },
    {
      "auxiliary_loss_clip": 0.01893758,
      "auxiliary_loss_mlp": 0.0115481,
      "balance_loss_clip": 1.33117318,
      "balance_loss_mlp": 1.05014443,
      "epoch": 0.00877773101665364,
      "flos": 30296791152000.0,
      "grad_norm": 2.097757480071856,
      "language_loss": 0.98849636,
      "learning_rate": 3.1082059657570015e-06,
      "loss": 1.01898205,
      "num_input_tokens_seen": 1666770,
      "step": 73,
      "time_per_iteration": 2.9573159217834473
    },
    {
      "auxiliary_loss_clip": 0.01887045,
      "auxiliary_loss_mlp": 0.01157442,
      "balance_loss_clip": 1.33067703,
      "balance_loss_mlp": 1.06078672,
      "epoch": 0.00889797390729273,
      "flos": 23514056104320.0,
      "grad_norm": 2.3820051095163195,
      "language_loss": 0.96637833,
      "learning_rate": 3.1180625252858496e-06,
      "loss": 0.99682319,
      "num_input_tokens_seen": 1685200,
      "step": 74,
      "time_per_iteration": 2.7658281326293945
    },
    {
      "auxiliary_loss_clip": 0.0186253,
      "auxiliary_loss_mlp": 0.01168921,
      "balance_loss_clip": 1.31867814,
      "balance_loss_mlp": 1.06950092,
      "epoch": 0.009018216797931822,
      "flos": 23075838178560.0,
      "grad_norm": 3.5958232747645495,
      "language_loss": 0.80363262,
      "learning_rate": 3.1277867780021663e-06,
      "loss": 0.83394718,
      "num_input_tokens_seen": 1701835,
      "step": 75,
      "time_per_iteration": 2.811174154281616
    },
    {
      "auxiliary_loss_clip": 0.01841367,
      "auxiliary_loss_mlp": 0.01145317,
      "balance_loss_clip": 1.31616116,
      "balance_loss_mlp": 1.05943847,
      "epoch": 0.009138459688570914,
      "flos": 15918877284480.0,
      "grad_norm": 1.8322208760358931,
      "language_loss": 0.95819902,
      "learning_rate": 3.1373822288779824e-06,
      "loss": 0.98806578,
      "num_input_tokens_seen": 1718415,
      "step": 76,
      "time_per_iteration": 2.7934348583221436
    },
    {
      "auxiliary_loss_clip": 0.01849599,
      "auxiliary_loss_mlp": 0.01170548,
      "balance_loss_clip": 1.3156352,
      "balance_loss_mlp": 1.07408416,
      "epoch": 0.009258702579210003,
      "flos": 27016531372800.0,
      "grad_norm": 1.9056880690729283,
      "language_loss": 0.79426813,
      "learning_rate": 3.1468522454274533e-06,
      "loss": 0.82446957,
      "num_input_tokens_seen": 1738770,
      "step": 77,
      "time_per_iteration": 2.8460984230041504
    },
    {
      "auxiliary_loss_clip": 0.01822034,
      "auxiliary_loss_mlp": 0.01167028,
      "balance_loss_clip": 1.30575919,
      "balance_loss_mlp": 1.07304299,
      "epoch": 0.009378945469849095,
      "flos": 26903196984960.0,
      "grad_norm": 2.1665553481433077,
      "language_loss": 0.91903865,
      "learning_rate": 3.15620006480197e-06,
      "loss": 0.94892925,
      "num_input_tokens_seen": 1758040,
      "step": 78,
      "time_per_iteration": 2.782259941101074
    },
    {
      "auxiliary_loss_clip": 0.01825896,
      "auxiliary_loss_mlp": 0.01148958,
      "balance_loss_clip": 1.30329621,
      "balance_loss_mlp": 1.0603137,
      "epoch": 0.009499188360488187,
      "flos": 35694236327040.0,
      "grad_norm": 3.296738972822618,
      "language_loss": 0.74674737,
      "learning_rate": 3.1654288004333087e-06,
      "loss": 0.77649587,
      "num_input_tokens_seen": 1776705,
      "step": 79,
      "time_per_iteration": 2.8932721614837646
    },
    {
      "auxiliary_loss_clip": 0.01805992,
      "auxiliary_loss_mlp": 0.01146265,
      "balance_loss_clip": 1.29915977,
      "balance_loss_mlp": 1.06653786,
      "epoch": 0.009619431251127276,
      "flos": 21503201944320.0,
      "grad_norm": 3.2681294437199204,
      "language_loss": 0.76094091,
      "learning_rate": 3.1745414482589353e-06,
      "loss": 0.79046339,
      "num_input_tokens_seen": 1795915,
      "step": 80,
      "time_per_iteration": 2.7435810565948486
    },
    {
      "auxiliary_loss_clip": 0.01784892,
      "auxiliary_loss_mlp": 0.01153992,
      "balance_loss_clip": 1.29052162,
      "balance_loss_mlp": 1.07302499,
      "epoch": 0.009739674141766368,
      "flos": 17421056991360.0,
      "grad_norm": 3.9944259323804734,
      "language_loss": 0.87182081,
      "learning_rate": 3.1835408925606204e-06,
      "loss": 0.90120971,
      "num_input_tokens_seen": 1814055,
      "step": 81,
      "time_per_iteration": 2.8428235054016113
    },
    {
      "auxiliary_loss_clip": 0.01775799,
      "auxiliary_loss_mlp": 0.01145835,
      "balance_loss_clip": 1.28615808,
      "balance_loss_mlp": 1.06868315,
      "epoch": 0.00985991703240546,
      "flos": 27527109246720.0,
      "grad_norm": 2.119452305500667,
      "language_loss": 0.89250362,
      "learning_rate": 3.1924299114448214e-06,
      "loss": 0.92171991,
      "num_input_tokens_seen": 1834535,
      "step": 82,
      "time_per_iteration": 2.8540806770324707
    },
    {
      "auxiliary_loss_clip": 0.01778879,
      "auxiliary_loss_mlp": 0.01159706,
      "balance_loss_clip": 1.28671062,
      "balance_loss_mlp": 1.07778549,
      "epoch": 0.00998015992304455,
      "flos": 13808084509440.0,
      "grad_norm": 2.0821045700117815,
      "language_loss": 0.83279049,
      "learning_rate": 3.2012111819909055e-06,
      "loss": 0.86217642,
      "num_input_tokens_seen": 1851865,
      "step": 83,
      "time_per_iteration": 2.7830352783203125
    },
    {
      "auxiliary_loss_clip": 0.01767898,
      "auxiliary_loss_mlp": 0.01147947,
      "balance_loss_clip": 1.28383541,
      "balance_loss_mlp": 1.07098591,
      "epoch": 0.010100402813683641,
      "flos": 20191385341440.0,
      "grad_norm": 2.200420330636526,
      "language_loss": 0.95237505,
      "learning_rate": 3.2098872850910627e-06,
      "loss": 0.98153353,
      "num_input_tokens_seen": 1868540,
      "step": 84,
      "time_per_iteration": 2.820887565612793
    },
    {
      "auxiliary_loss_clip": 0.01769197,
      "auxiliary_loss_mlp": 0.01140347,
      "balance_loss_clip": 1.28550935,
      "balance_loss_mlp": 1.06476879,
      "epoch": 0.010220645704322733,
      "flos": 17201642762880.0,
      "grad_norm": 1.8923507372060147,
      "language_loss": 0.89354932,
      "learning_rate": 3.2184607100038194e-06,
      "loss": 0.92264473,
      "num_input_tokens_seen": 1887180,
      "step": 85,
      "time_per_iteration": 2.800595998764038
    },
    {
      "auxiliary_loss_clip": 0.01746566,
      "auxiliary_loss_mlp": 0.01129378,
      "balance_loss_clip": 1.27404165,
      "balance_loss_mlp": 1.05499101,
      "epoch": 0.010340888594961822,
      "flos": 21470415805440.0,
      "grad_norm": 2.378485622949192,
      "language_loss": 0.93350601,
      "learning_rate": 3.2269338586412414e-06,
      "loss": 0.96226549,
      "num_input_tokens_seen": 1904765,
      "step": 86,
      "time_per_iteration": 2.7614243030548096
    },
    {
      "auxiliary_loss_clip": 0.01739104,
      "auxiliary_loss_mlp": 0.01142659,
      "balance_loss_clip": 1.27134991,
      "balance_loss_mlp": 1.07518649,
      "epoch": 0.010461131485600914,
      "flos": 23002831785600.0,
      "grad_norm": 2.3198102440964106,
      "language_loss": 0.9631446,
      "learning_rate": 3.2353090496083106e-06,
      "loss": 0.99196231,
      "num_input_tokens_seen": 1922600,
      "step": 87,
      "time_per_iteration": 2.806793212890625
    },
    {
      "auxiliary_loss_clip": 0.01729518,
      "auxiliary_loss_mlp": 0.0112693,
      "balance_loss_clip": 1.26786017,
      "balance_loss_mlp": 1.0581224,
      "epoch": 0.010581374376240005,
      "flos": 33546850571520.0,
      "grad_norm": 2.9160813256824896,
      "language_loss": 0.81401861,
      "learning_rate": 3.2435885220114572e-06,
      "loss": 0.84258306,
      "num_input_tokens_seen": 1943950,
      "step": 88,
      "time_per_iteration": 2.8113462924957275
    },
    {
      "auxiliary_loss_clip": 0.01728568,
      "auxiliary_loss_mlp": 0.01132877,
      "balance_loss_clip": 1.26657844,
      "balance_loss_mlp": 1.06945801,
      "epoch": 0.010701617266879095,
      "flos": 21763087822080.0,
      "grad_norm": 2.8277232742236866,
      "language_loss": 0.93954146,
      "learning_rate": 3.2517744390519113e-06,
      "loss": 0.96815592,
      "num_input_tokens_seen": 1962815,
      "step": 89,
      "time_per_iteration": 2.8294315338134766
    },
    {
      "auxiliary_loss_clip": 0.01709919,
      "auxiliary_loss_mlp": 0.01115001,
      "balance_loss_clip": 1.25421834,
      "balance_loss_mlp": 1.05453777,
      "epoch": 0.010821860157518187,
      "flos": 19060199256960.0,
      "grad_norm": 2.0253127648094074,
      "language_loss": 0.7519933,
      "learning_rate": 3.259868891418298e-06,
      "loss": 0.78024244,
      "num_input_tokens_seen": 1980580,
      "step": 90,
      "time_per_iteration": 2.7273151874542236
    },
    {
      "auxiliary_loss_clip": 0.01710881,
      "auxiliary_loss_mlp": 0.0114234,
      "balance_loss_clip": 1.25902343,
      "balance_loss_mlp": 1.07205427,
      "epoch": 0.010942103048157278,
      "flos": 25447378757760.0,
      "grad_norm": 2.198992716400347,
      "language_loss": 0.8498674,
      "learning_rate": 3.2678739004917757e-06,
      "loss": 0.87839967,
      "num_input_tokens_seen": 2000315,
      "step": 91,
      "time_per_iteration": 2.851680278778076
    },
    {
      "auxiliary_loss_clip": 0.01704175,
      "auxiliary_loss_mlp": 0.01132637,
      "balance_loss_clip": 1.25302362,
      "balance_loss_mlp": 1.06979024,
      "epoch": 0.011062345938796368,
      "flos": 27493928058240.0,
      "grad_norm": 1.6217159425185765,
      "language_loss": 0.921597,
      "learning_rate": 3.275791421376029e-06,
      "loss": 0.94996512,
      "num_input_tokens_seen": 2023760,
      "step": 92,
      "time_per_iteration": 3.658412218093872
    },
    {
      "auxiliary_loss_clip": 0.016959,
      "auxiliary_loss_mlp": 0.01119514,
      "balance_loss_clip": 1.25192046,
      "balance_loss_mlp": 1.06255579,
      "epoch": 0.01118258882943546,
      "flos": 16071210864000.0,
      "grad_norm": 2.2343151093219045,
      "language_loss": 0.96135616,
      "learning_rate": 3.2836233457634622e-06,
      "loss": 0.9895103,
      "num_input_tokens_seen": 2041895,
      "step": 93,
      "time_per_iteration": 4.5933310985565186
    },
    {
      "auxiliary_loss_clip": 0.01690163,
      "auxiliary_loss_mlp": 0.0113285,
      "balance_loss_clip": 1.24838412,
      "balance_loss_mlp": 1.06389964,
      "epoch": 0.011302831720074551,
      "flos": 20668602458880.0,
      "grad_norm": 2.271387745778157,
      "language_loss": 0.85538971,
      "learning_rate": 3.2913715046481135e-06,
      "loss": 0.88361984,
      "num_input_tokens_seen": 2061640,
      "step": 94,
      "time_per_iteration": 2.7608888149261475
    },
    {
      "auxiliary_loss_clip": 0.01682073,
      "auxiliary_loss_mlp": 0.01111468,
      "balance_loss_clip": 1.24334979,
      "balance_loss_mlp": 1.05305505,
      "epoch": 0.011423074610713641,
      "flos": 13072238490240.0,
      "grad_norm": 2.2394329663049635,
      "language_loss": 0.88671839,
      "learning_rate": 3.299037670895023e-06,
      "loss": 0.91465378,
      "num_input_tokens_seen": 2078255,
      "step": 95,
      "time_per_iteration": 2.7603001594543457
    },
    {
      "auxiliary_loss_clip": 0.01685062,
      "auxiliary_loss_mlp": 0.01109305,
      "balance_loss_clip": 1.24709272,
      "balance_loss_mlp": 1.0501771,
      "epoch": 0.011543317501352733,
      "flos": 30335646689280.0,
      "grad_norm": 1.9716010671710815,
      "language_loss": 0.80162954,
      "learning_rate": 3.3066235616750667e-06,
      "loss": 0.82957315,
      "num_input_tokens_seen": 2099490,
      "step": 96,
      "time_per_iteration": 2.835517168045044
    },
    {
      "auxiliary_loss_clip": 0.01651588,
      "auxiliary_loss_mlp": 0.0110807,
      "balance_loss_clip": 1.22912633,
      "balance_loss_mlp": 1.05344832,
      "epoch": 0.011663560391991824,
      "flos": 15522962601600.0,
      "grad_norm": 2.426808710548288,
      "language_loss": 0.92517221,
      "learning_rate": 3.3141308407736276e-06,
      "loss": 0.95276886,
      "num_input_tokens_seen": 2116125,
      "step": 97,
      "time_per_iteration": 2.7461814880371094
    },
    {
      "auxiliary_loss_clip": 0.01649204,
      "auxiliary_loss_mlp": 0.01114609,
      "balance_loss_clip": 1.22166777,
      "balance_loss_mlp": 1.05996323,
      "epoch": 0.011783803282630914,
      "flos": 19902125116800.0,
      "grad_norm": 4.315403232198553,
      "language_loss": 0.86542857,
      "learning_rate": 3.321561120780869e-06,
      "loss": 0.89306664,
      "num_input_tokens_seen": 2134835,
      "step": 98,
      "time_per_iteration": 2.7719523906707764
    },
    {
      "auxiliary_loss_clip": 0.01652237,
      "auxiliary_loss_mlp": 0.01123995,
      "balance_loss_clip": 1.23327065,
      "balance_loss_mlp": 1.07018375,
      "epoch": 0.011904046173270006,
      "flos": 22340674517760.0,
      "grad_norm": 2.3121610371157866,
      "language_loss": 1.01338661,
      "learning_rate": 3.3289159651708192e-06,
      "loss": 1.0411489,
      "num_input_tokens_seen": 2152410,
      "step": 99,
      "time_per_iteration": 2.7645716667175293
    },
    {
      "auxiliary_loss_clip": 0.01643335,
      "auxiliary_loss_mlp": 0.01112099,
      "balance_loss_clip": 1.22376597,
      "balance_loss_mlp": 1.0565474,
      "epoch": 0.012024289063909096,
      "flos": 19100060375040.0,
      "grad_norm": 1.9825972759369996,
      "language_loss": 0.97755855,
      "learning_rate": 3.3361968902759768e-06,
      "loss": 1.00511289,
      "num_input_tokens_seen": 2172090,
      "step": 100,
      "time_per_iteration": 2.7129034996032715
    },
    {
      "auxiliary_loss_clip": 0.01625094,
      "auxiliary_loss_mlp": 0.01095759,
      "balance_loss_clip": 1.21462393,
      "balance_loss_mlp": 1.04707408,
      "epoch": 0.012144531954548187,
      "flos": 15012205159680.0,
      "grad_norm": 3.20056968164523,
      "language_loss": 0.93713617,
      "learning_rate": 3.343405367163663e-06,
      "loss": 0.96434474,
      "num_input_tokens_seen": 2189020,
      "step": 101,
      "time_per_iteration": 2.7386598587036133
    },
    {
      "auxiliary_loss_clip": 0.01636917,
      "auxiliary_loss_mlp": 0.01112425,
      "balance_loss_clip": 1.22053385,
      "balance_loss_mlp": 1.0616653,
      "epoch": 0.012264774845187279,
      "flos": 15122020014720.0,
      "grad_norm": 2.350256743794026,
      "language_loss": 0.81376719,
      "learning_rate": 3.350542823419951e-06,
      "loss": 0.84126055,
      "num_input_tokens_seen": 2205620,
      "step": 102,
      "time_per_iteration": 2.718799114227295
    },
    {
      "auxiliary_loss_clip": 0.01636278,
      "auxiliary_loss_mlp": 0.01100069,
      "balance_loss_clip": 1.22151434,
      "balance_loss_mlp": 1.04663968,
      "epoch": 0.012385017735826368,
      "flos": 13949248959360.0,
      "grad_norm": 3.617264296968206,
      "language_loss": 0.87479031,
      "learning_rate": 3.3576106448465615e-06,
      "loss": 0.90215373,
      "num_input_tokens_seen": 2219000,
      "step": 103,
      "time_per_iteration": 2.695237874984741
    },
    {
      "auxiliary_loss_clip": 0.01615814,
      "auxiliary_loss_mlp": 0.01108381,
      "balance_loss_clip": 1.20921195,
      "balance_loss_mlp": 1.05962396,
      "epoch": 0.01250526062646546,
      "flos": 23623260428160.0,
      "grad_norm": 2.098614039988764,
      "language_loss": 0.88140273,
      "learning_rate": 3.3646101770757797e-06,
      "loss": 0.9086448,
      "num_input_tokens_seen": 2237790,
      "step": 104,
      "time_per_iteration": 2.7375972270965576
    },
    {
      "auxiliary_loss_clip": 0.0160539,
      "auxiliary_loss_mlp": 0.01103918,
      "balance_loss_clip": 1.20787764,
      "balance_loss_mlp": 1.05251515,
      "epoch": 0.012625503517104552,
      "flos": 34640078958720.0,
      "grad_norm": 1.6674427748119232,
      "language_loss": 0.85632902,
      "learning_rate": 3.371542727108104e-06,
      "loss": 0.88342214,
      "num_input_tokens_seen": 2259965,
      "step": 105,
      "time_per_iteration": 2.864018678665161
    },
    {
      "auxiliary_loss_clip": 0.01613643,
      "auxiliary_loss_mlp": 0.01115266,
      "balance_loss_clip": 1.20845723,
      "balance_loss_mlp": 1.06376719,
      "epoch": 0.012745746407743641,
      "flos": 17821891837440.0,
      "grad_norm": 3.6040229704753215,
      "language_loss": 0.90128565,
      "learning_rate": 3.3784095647770114e-06,
      "loss": 0.92857468,
      "num_input_tokens_seen": 2278610,
      "step": 106,
      "time_per_iteration": 2.6798036098480225
    },
    {
      "auxiliary_loss_clip": 0.01602969,
      "auxiliary_loss_mlp": 0.01105155,
      "balance_loss_clip": 1.20407164,
      "balance_loss_mlp": 1.05179727,
      "epoch": 0.012865989298382733,
      "flos": 20595057361920.0,
      "grad_norm": 5.844317503764526,
      "language_loss": 0.88332534,
      "learning_rate": 3.3852119241449547e-06,
      "loss": 0.91040665,
      "num_input_tokens_seen": 2297730,
      "step": 107,
      "time_per_iteration": 2.7417352199554443
    },
    {
      "auxiliary_loss_clip": 0.01595791,
      "auxiliary_loss_mlp": 0.01107104,
      "balance_loss_clip": 1.19982409,
      "balance_loss_mlp": 1.05663037,
      "epoch": 0.012986232189021825,
      "flos": 23948969978880.0,
      "grad_norm": 2.5717829705124173,
      "language_loss": 0.96081209,
      "learning_rate": 3.3919510048344295e-06,
      "loss": 0.98784101,
      "num_input_tokens_seen": 2315740,
      "step": 108,
      "time_per_iteration": 2.7866291999816895
    },
    {
      "auxiliary_loss_clip": 0.01585557,
      "auxiliary_loss_mlp": 0.01115806,
      "balance_loss_clip": 1.19923759,
      "balance_loss_mlp": 1.0680989,
      "epoch": 0.013106475079660914,
      "flos": 23725425686400.0,
      "grad_norm": 2.1352850769786533,
      "language_loss": 0.86749071,
      "learning_rate": 3.3986279732976907e-06,
      "loss": 0.89450443,
      "num_input_tokens_seen": 2334215,
      "step": 109,
      "time_per_iteration": 2.7947704792022705
    },
    {
      "auxiliary_loss_clip": 0.01578879,
      "auxiliary_loss_mlp": 0.01101873,
      "balance_loss_clip": 1.19448781,
      "balance_loss_mlp": 1.05585861,
      "epoch": 0.013226717970300006,
      "flos": 21102438925440.0,
      "grad_norm": 1.9952479998999384,
      "language_loss": 0.9541949,
      "learning_rate": 3.4052439640284983e-06,
      "loss": 0.98100239,
      "num_input_tokens_seen": 2353130,
      "step": 110,
      "time_per_iteration": 2.774613380432129
    },
    {
      "auxiliary_loss_clip": 0.01579156,
      "auxiliary_loss_mlp": 0.01089863,
      "balance_loss_clip": 1.1911397,
      "balance_loss_mlp": 1.04446864,
      "epoch": 0.013346960860939098,
      "flos": 24863902231680.0,
      "grad_norm": 1.8123564720375656,
      "language_loss": 0.81381464,
      "learning_rate": 3.4118000807190217e-06,
      "loss": 0.84050488,
      "num_input_tokens_seen": 2374010,
      "step": 111,
      "time_per_iteration": 2.770792245864868
    },
    {
      "auxiliary_loss_clip": 0.01581371,
      "auxiliary_loss_mlp": 0.01097763,
      "balance_loss_clip": 1.19124568,
      "balance_loss_mlp": 1.054919,
      "epoch": 0.013467203751578187,
      "flos": 28181940140160.0,
      "grad_norm": 1.8752645698568884,
      "language_loss": 0.75926155,
      "learning_rate": 3.4182973973648723e-06,
      "loss": 0.78605288,
      "num_input_tokens_seen": 2395220,
      "step": 112,
      "time_per_iteration": 2.8275082111358643
    },
    {
      "auxiliary_loss_clip": 0.01573474,
      "auxiliary_loss_mlp": 0.01100741,
      "balance_loss_clip": 1.18915594,
      "balance_loss_mlp": 1.05613327,
      "epoch": 0.013587446642217279,
      "flos": 18916233546240.0,
      "grad_norm": 2.569109691972843,
      "language_loss": 0.95035571,
      "learning_rate": 3.424736959321014e-06,
      "loss": 0.97709781,
      "num_input_tokens_seen": 2413025,
      "step": 113,
      "time_per_iteration": 2.711380958557129
    },
    {
      "auxiliary_loss_clip": 0.01579039,
      "auxiliary_loss_mlp": 0.011035,
      "balance_loss_clip": 1.19279599,
      "balance_loss_mlp": 1.05827212,
      "epoch": 0.01370768953285637,
      "flos": 23988615615360.0,
      "grad_norm": 2.146199732362796,
      "language_loss": 0.88576341,
      "learning_rate": 3.431119784311155e-06,
      "loss": 0.91258878,
      "num_input_tokens_seen": 2432700,
      "step": 114,
      "time_per_iteration": 2.7693426609039307
    },
    {
      "auxiliary_loss_clip": 0.01565014,
      "auxiliary_loss_mlp": 0.01101605,
      "balance_loss_clip": 1.18894136,
      "balance_loss_mlp": 1.05773604,
      "epoch": 0.01382793242349546,
      "flos": 39202565512320.0,
      "grad_norm": 2.1666806059867594,
      "language_loss": 0.77601242,
      "learning_rate": 3.43744686339307e-06,
      "loss": 0.80267859,
      "num_input_tokens_seen": 2455020,
      "step": 115,
      "time_per_iteration": 2.8842124938964844
    },
    {
      "auxiliary_loss_clip": 0.01549065,
      "auxiliary_loss_mlp": 0.01078877,
      "balance_loss_clip": 1.17411411,
      "balance_loss_mlp": 1.04165959,
      "epoch": 0.013948175314134552,
      "flos": 41353506714240.0,
      "grad_norm": 2.1122867395809837,
      "language_loss": 0.90540463,
      "learning_rate": 3.44371916188212e-06,
      "loss": 0.93168402,
      "num_input_tokens_seen": 2475775,
      "step": 116,
      "time_per_iteration": 2.9128360748291016
    },
    {
      "auxiliary_loss_clip": 0.01553238,
      "auxiliary_loss_mlp": 0.01085621,
      "balance_loss_clip": 1.17802,
      "balance_loss_mlp": 1.04931021,
      "epoch": 0.014068418204773643,
      "flos": 22453542028800.0,
      "grad_norm": 2.0338129181752675,
      "language_loss": 0.85834879,
      "learning_rate": 3.449937620235143e-06,
      "loss": 0.88473737,
      "num_input_tokens_seen": 2496370,
      "step": 117,
      "time_per_iteration": 2.7472524642944336
    },
    {
      "auxiliary_loss_clip": 0.01540398,
      "auxiliary_loss_mlp": 0.01096417,
      "balance_loss_clip": 1.1737566,
      "balance_loss_mlp": 1.05784118,
      "epoch": 0.014188661095412733,
      "flos": 23805147922560.0,
      "grad_norm": 1.8402946869425383,
      "language_loss": 0.89322424,
      "learning_rate": 3.456103154896722e-06,
      "loss": 0.91959238,
      "num_input_tokens_seen": 2517645,
      "step": 118,
      "time_per_iteration": 3.6204569339752197
    },
    {
      "auxiliary_loss_clip": 0.01534657,
      "auxiliary_loss_mlp": 0.01094205,
      "balance_loss_clip": 1.17189145,
      "balance_loss_mlp": 1.05212426,
      "epoch": 0.014308903986051825,
      "flos": 23660248458240.0,
      "grad_norm": 1.8717561131930636,
      "language_loss": 0.92343897,
      "learning_rate": 3.462216659109757e-06,
      "loss": 0.94972754,
      "num_input_tokens_seen": 2537825,
      "step": 119,
      "time_per_iteration": 4.65583872795105
    },
    {
      "auxiliary_loss_clip": 0.01544189,
      "auxiliary_loss_mlp": 0.01094502,
      "balance_loss_clip": 1.17379403,
      "balance_loss_mlp": 1.05435193,
      "epoch": 0.014429146876690916,
      "flos": 20667991927680.0,
      "grad_norm": 2.4130892512528015,
      "language_loss": 0.85072464,
      "learning_rate": 3.4682790036921077e-06,
      "loss": 0.87711149,
      "num_input_tokens_seen": 2556485,
      "step": 120,
      "time_per_iteration": 3.699442148208618
    },
    {
      "auxiliary_loss_clip": 0.01518442,
      "auxiliary_loss_mlp": 0.01082955,
      "balance_loss_clip": 1.16455662,
      "balance_loss_mlp": 1.04931378,
      "epoch": 0.014549389767330006,
      "flos": 20229199384320.0,
      "grad_norm": 1.919712002858095,
      "language_loss": 0.83186501,
      "learning_rate": 3.4742910377810193e-06,
      "loss": 0.85787892,
      "num_input_tokens_seen": 2573945,
      "step": 121,
      "time_per_iteration": 2.709707736968994
    },
    {
      "auxiliary_loss_clip": 0.01524272,
      "auxiliary_loss_mlp": 0.0110188,
      "balance_loss_clip": 1.16599607,
      "balance_loss_mlp": 1.06661797,
      "epoch": 0.014669632657969098,
      "flos": 18004174381440.0,
      "grad_norm": 3.2851498112503377,
      "language_loss": 0.88705194,
      "learning_rate": 3.4802535895469042e-06,
      "loss": 0.91331339,
      "num_input_tokens_seen": 2592695,
      "step": 122,
      "time_per_iteration": 2.670933723449707
    },
    {
      "auxiliary_loss_clip": 0.01522686,
      "auxiliary_loss_mlp": 0.01093607,
      "balance_loss_clip": 1.16536188,
      "balance_loss_mlp": 1.05636621,
      "epoch": 0.01478987554860819,
      "flos": 22741796672640.0,
      "grad_norm": 2.0979371559302504,
      "language_loss": 0.89413953,
      "learning_rate": 3.4861674668779934e-06,
      "loss": 0.92030245,
      "num_input_tokens_seen": 2610925,
      "step": 123,
      "time_per_iteration": 2.747762441635132
    },
    {
      "auxiliary_loss_clip": 0.01518753,
      "auxiliary_loss_mlp": 0.01094183,
      "balance_loss_clip": 1.16223788,
      "balance_loss_mlp": 1.05925477,
      "epoch": 0.01491011843924728,
      "flos": 17198590106880.0,
      "grad_norm": 2.125410033053841,
      "language_loss": 0.84308249,
      "learning_rate": 3.492033458037272e-06,
      "loss": 0.86921191,
      "num_input_tokens_seen": 2629495,
      "step": 124,
      "time_per_iteration": 2.7083122730255127
    },
    {
      "auxiliary_loss_clip": 0.01508444,
      "auxiliary_loss_mlp": 0.01089675,
      "balance_loss_clip": 1.15483034,
      "balance_loss_mlp": 1.05603445,
      "epoch": 0.01503036132988637,
      "flos": 17673867889920.0,
      "grad_norm": 2.4145961821687947,
      "language_loss": 0.8689959,
      "learning_rate": 3.497852332293018e-06,
      "loss": 0.89497715,
      "num_input_tokens_seen": 2645070,
      "step": 125,
      "time_per_iteration": 2.7551441192626953
    },
    {
      "auxiliary_loss_clip": 0.01509738,
      "auxiliary_loss_mlp": 0.01090909,
      "balance_loss_clip": 1.1598649,
      "balance_loss_mlp": 1.05741096,
      "epoch": 0.015150604220525462,
      "flos": 18878239935360.0,
      "grad_norm": 3.1882841277881284,
      "language_loss": 0.96611857,
      "learning_rate": 3.5036248405242356e-06,
      "loss": 0.99212503,
      "num_input_tokens_seen": 2663825,
      "step": 126,
      "time_per_iteration": 2.7205774784088135
    },
    {
      "auxiliary_loss_clip": 0.0150564,
      "auxiliary_loss_mlp": 0.01113146,
      "balance_loss_clip": 1.15599477,
      "balance_loss_mlp": 1.07325864,
      "epoch": 0.015270847111164552,
      "flos": 39420184060800.0,
      "grad_norm": 2.0125373774373183,
      "language_loss": 0.82604873,
      "learning_rate": 3.509351715802146e-06,
      "loss": 0.85223657,
      "num_input_tokens_seen": 2684710,
      "step": 127,
      "time_per_iteration": 2.903897762298584
    },
    {
      "auxiliary_loss_clip": 0.01493251,
      "auxiliary_loss_mlp": 0.01082611,
      "balance_loss_clip": 1.14927626,
      "balance_loss_mlp": 1.04711056,
      "epoch": 0.015391090001803644,
      "flos": 43762466286720.0,
      "grad_norm": 2.1210038515489344,
      "language_loss": 0.78434384,
      "learning_rate": 3.5150336739488763e-06,
      "loss": 0.81010252,
      "num_input_tokens_seen": 2706995,
      "step": 128,
      "time_per_iteration": 2.8854024410247803
    },
    {
      "auxiliary_loss_clip": 0.01495778,
      "auxiliary_loss_mlp": 0.01086586,
      "balance_loss_clip": 1.15144837,
      "balance_loss_mlp": 1.05652153,
      "epoch": 0.015511332892442733,
      "flos": 18916341287040.0,
      "grad_norm": 1.9920031885938865,
      "language_loss": 0.83905834,
      "learning_rate": 3.5206714140744143e-06,
      "loss": 0.86488199,
      "num_input_tokens_seen": 2727050,
      "step": 129,
      "time_per_iteration": 2.7462501525878906
    },
    {
      "auxiliary_loss_clip": 0.01500479,
      "auxiliary_loss_mlp": 0.01099146,
      "balance_loss_clip": 1.1551038,
      "balance_loss_mlp": 1.06297779,
      "epoch": 0.015631575783081827,
      "flos": 24535283679360.0,
      "grad_norm": 2.7927649198648914,
      "language_loss": 0.87461013,
      "learning_rate": 3.5262656190928208e-06,
      "loss": 0.90060633,
      "num_input_tokens_seen": 2745350,
      "step": 130,
      "time_per_iteration": 2.845433235168457
    },
    {
      "auxiliary_loss_clip": 0.01513118,
      "auxiliary_loss_mlp": 0.01090015,
      "balance_loss_clip": 1.22729838,
      "balance_loss_mlp": 1.06273985,
      "epoch": 0.015751818673720917,
      "flos": 62328536098560.0,
      "grad_norm": 1.053613045616893,
      "language_loss": 0.71592617,
      "learning_rate": 3.5318169562186737e-06,
      "loss": 0.74195743,
      "num_input_tokens_seen": 2814195,
      "step": 131,
      "time_per_iteration": 3.3020341396331787
    },
    {
      "auxiliary_loss_clip": 0.01490268,
      "auxiliary_loss_mlp": 0.01096562,
      "balance_loss_clip": 1.14874685,
      "balance_loss_mlp": 1.064399,
      "epoch": 0.015872061564360006,
      "flos": 23878549365120.0,
      "grad_norm": 2.3704484556815,
      "language_loss": 0.82033777,
      "learning_rate": 3.5373260774446292e-06,
      "loss": 0.84620613,
      "num_input_tokens_seen": 2834645,
      "step": 132,
      "time_per_iteration": 2.75390625
    },
    {
      "auxiliary_loss_clip": 0.01481003,
      "auxiliary_loss_mlp": 0.01094375,
      "balance_loss_clip": 1.1441102,
      "balance_loss_mlp": 1.06228364,
      "epoch": 0.0159923044549991,
      "flos": 23367899664000.0,
      "grad_norm": 1.832005124637457,
      "language_loss": 0.90313101,
      "learning_rate": 3.542793620000961e-06,
      "loss": 0.92888486,
      "num_input_tokens_seen": 2854120,
      "step": 133,
      "time_per_iteration": 2.716811418533325
    },
    {
      "auxiliary_loss_clip": 0.01486233,
      "auxiliary_loss_mlp": 0.01099182,
      "balance_loss_clip": 1.14925385,
      "balance_loss_mlp": 1.06582761,
      "epoch": 0.01611254734563819,
      "flos": 17858305249920.0,
      "grad_norm": 5.059662976380581,
      "language_loss": 0.86841142,
      "learning_rate": 3.5482202067978894e-06,
      "loss": 0.89426553,
      "num_input_tokens_seen": 2871330,
      "step": 134,
      "time_per_iteration": 2.743974447250366
    },
    {
      "auxiliary_loss_clip": 0.01484045,
      "auxiliary_loss_mlp": 0.01089784,
      "balance_loss_clip": 1.1492188,
      "balance_loss_mlp": 1.05826473,
      "epoch": 0.01623279023627728,
      "flos": 20954774113920.0,
      "grad_norm": 1.9500121656249,
      "language_loss": 0.7600261,
      "learning_rate": 3.553606446851471e-06,
      "loss": 0.78576446,
      "num_input_tokens_seen": 2888070,
      "step": 135,
      "time_per_iteration": 2.6938154697418213
    },
    {
      "auxiliary_loss_clip": 0.01479727,
      "auxiliary_loss_mlp": 0.01082928,
      "balance_loss_clip": 1.144701,
      "balance_loss_mlp": 1.05357885,
      "epoch": 0.016353033126916373,
      "flos": 15742412743680.0,
      "grad_norm": 1.9634076313710365,
      "language_loss": 0.83299983,
      "learning_rate": 3.5589529356937613e-06,
      "loss": 0.85862637,
      "num_input_tokens_seen": 2906465,
      "step": 136,
      "time_per_iteration": 2.69671630859375
    },
    {
      "auxiliary_loss_clip": 0.0147532,
      "auxiliary_loss_mlp": 0.01087005,
      "balance_loss_clip": 1.14303315,
      "balance_loss_mlp": 1.05688047,
      "epoch": 0.016473276017555463,
      "flos": 18807280617600.0,
      "grad_norm": 2.0032207055993547,
      "language_loss": 0.7676419,
      "learning_rate": 3.5642602557679627e-06,
      "loss": 0.79326516,
      "num_input_tokens_seen": 2924915,
      "step": 137,
      "time_per_iteration": 2.7246298789978027
    },
    {
      "auxiliary_loss_clip": 0.01479305,
      "auxiliary_loss_mlp": 0.01088637,
      "balance_loss_clip": 1.15059805,
      "balance_loss_mlp": 1.0598954,
      "epoch": 0.016593518908194552,
      "flos": 24352641999360.0,
      "grad_norm": 2.2814956466904905,
      "language_loss": 0.84126949,
      "learning_rate": 3.569528976809202e-06,
      "loss": 0.86694896,
      "num_input_tokens_seen": 2942130,
      "step": 138,
      "time_per_iteration": 2.9137814044952393
    },
    {
      "auxiliary_loss_clip": 0.01468537,
      "auxiliary_loss_mlp": 0.01073625,
      "balance_loss_clip": 1.14031768,
      "balance_loss_mlp": 1.04510999,
      "epoch": 0.016713761798833646,
      "flos": 22346133384960.0,
      "grad_norm": 1.7302916707947082,
      "language_loss": 0.89940566,
      "learning_rate": 3.5747596562115522e-06,
      "loss": 0.92482728,
      "num_input_tokens_seen": 2962745,
      "step": 139,
      "time_per_iteration": 2.8356661796569824
    },
    {
      "auxiliary_loss_clip": 0.01476201,
      "auxiliary_loss_mlp": 0.01072274,
      "balance_loss_clip": 1.14602876,
      "balance_loss_mlp": 1.03899121,
      "epoch": 0.016834004689472735,
      "flos": 17821820010240.0,
      "grad_norm": 11.659725848067133,
      "language_loss": 0.90954,
      "learning_rate": 3.5799528393819138e-06,
      "loss": 0.93502474,
      "num_input_tokens_seen": 2981825,
      "step": 140,
      "time_per_iteration": 2.7296414375305176
    },
    {
      "auxiliary_loss_clip": 0.01462237,
      "auxiliary_loss_mlp": 0.01093168,
      "balance_loss_clip": 1.13566709,
      "balance_loss_mlp": 1.06354427,
      "epoch": 0.016954247580111825,
      "flos": 20519501103360.0,
      "grad_norm": 1.9649385670469819,
      "language_loss": 0.88076293,
      "learning_rate": 3.585109060081286e-06,
      "loss": 0.90631694,
      "num_input_tokens_seen": 3001625,
      "step": 141,
      "time_per_iteration": 2.728119134902954
    },
    {
      "auxiliary_loss_clip": 0.01467646,
      "auxiliary_loss_mlp": 0.01091003,
      "balance_loss_clip": 1.13944304,
      "balance_loss_mlp": 1.06133199,
      "epoch": 0.017074490470750915,
      "flos": 22088869200000.0,
      "grad_norm": 1.8505606643583166,
      "language_loss": 0.78662086,
      "learning_rate": 3.590228840753992e-06,
      "loss": 0.81220734,
      "num_input_tokens_seen": 3022055,
      "step": 142,
      "time_per_iteration": 2.8031117916107178
    },
    {
      "auxiliary_loss_clip": 0.01466192,
      "auxiliary_loss_mlp": 0.01068637,
      "balance_loss_clip": 1.14059305,
      "balance_loss_mlp": 1.04185033,
      "epoch": 0.01719473336139001,
      "flos": 15997270717440.0,
      "grad_norm": 2.0564516927045267,
      "language_loss": 0.87326121,
      "learning_rate": 3.5953126928453423e-06,
      "loss": 0.89860952,
      "num_input_tokens_seen": 3039605,
      "step": 143,
      "time_per_iteration": 2.743452548980713
    },
    {
      "auxiliary_loss_clip": 0.01455996,
      "auxiliary_loss_mlp": 0.01080635,
      "balance_loss_clip": 1.13414657,
      "balance_loss_mlp": 1.05251336,
      "epoch": 0.017314976252029098,
      "flos": 22492038430080.0,
      "grad_norm": 2.0381563751313254,
      "language_loss": 0.80287719,
      "learning_rate": 3.600361117108239e-06,
      "loss": 0.82824349,
      "num_input_tokens_seen": 3059405,
      "step": 144,
      "time_per_iteration": 2.72440505027771
    },
    {
      "auxiliary_loss_clip": 0.01456845,
      "auxiliary_loss_mlp": 0.01083546,
      "balance_loss_clip": 1.13420248,
      "balance_loss_mlp": 1.05565143,
      "epoch": 0.017435219142668188,
      "flos": 22018053536640.0,
      "grad_norm": 2.084531643465103,
      "language_loss": 0.97235394,
      "learning_rate": 3.6053746038991616e-06,
      "loss": 0.99775779,
      "num_input_tokens_seen": 3078490,
      "step": 145,
      "time_per_iteration": 4.572920560836792
    },
    {
      "auxiliary_loss_clip": 0.0143102,
      "auxiliary_loss_mlp": 0.01019412,
      "balance_loss_clip": 1.17987156,
      "balance_loss_mlp": 1.00081575,
      "epoch": 0.01755546203330728,
      "flos": 72240526149120.0,
      "grad_norm": 1.0501163205754471,
      "language_loss": 0.58482635,
      "learning_rate": 3.6103536334639843e-06,
      "loss": 0.60933065,
      "num_input_tokens_seen": 3131755,
      "step": 146,
      "time_per_iteration": 4.197690486907959
    },
    {
      "auxiliary_loss_clip": 0.01445533,
      "auxiliary_loss_mlp": 0.01083313,
      "balance_loss_clip": 1.13059616,
      "balance_loss_mlp": 1.05572844,
      "epoch": 0.01767570492394637,
      "flos": 25337061112320.0,
      "grad_norm": 2.382601794250895,
      "language_loss": 0.85594136,
      "learning_rate": 3.615298676214041e-06,
      "loss": 0.88122976,
      "num_input_tokens_seen": 3152035,
      "step": 147,
      "time_per_iteration": 2.766993999481201
    },
    {
      "auxiliary_loss_clip": 0.01441292,
      "auxiliary_loss_mlp": 0.01071382,
      "balance_loss_clip": 1.12834907,
      "balance_loss_mlp": 1.04333234,
      "epoch": 0.01779594781458546,
      "flos": 20449188230400.0,
      "grad_norm": 2.961288353323809,
      "language_loss": 0.88844216,
      "learning_rate": 3.6202101929928317e-06,
      "loss": 0.91356885,
      "num_input_tokens_seen": 3170625,
      "step": 148,
      "time_per_iteration": 2.7267377376556396
    },
    {
      "auxiliary_loss_clip": 0.01447279,
      "auxiliary_loss_mlp": 0.01086578,
      "balance_loss_clip": 1.13072133,
      "balance_loss_mlp": 1.06135368,
      "epoch": 0.017916190705224554,
      "flos": 16253601148800.0,
      "grad_norm": 2.436463314393052,
      "language_loss": 0.88274872,
      "learning_rate": 3.6250886353337413e-06,
      "loss": 0.90808731,
      "num_input_tokens_seen": 3188155,
      "step": 149,
      "time_per_iteration": 2.691784620285034
    },
    {
      "auxiliary_loss_clip": 0.01444672,
      "auxiliary_loss_mlp": 0.01080986,
      "balance_loss_clip": 1.12895632,
      "balance_loss_mlp": 1.05365133,
      "epoch": 0.018036433595863644,
      "flos": 23330588411520.0,
      "grad_norm": 2.3329753000018636,
      "language_loss": 0.86464584,
      "learning_rate": 3.6299344457091488e-06,
      "loss": 0.88990235,
      "num_input_tokens_seen": 3209015,
      "step": 150,
      "time_per_iteration": 2.7465267181396484
    },
    {
      "auxiliary_loss_clip": 0.01437247,
      "auxiliary_loss_mlp": 0.01076502,
      "balance_loss_clip": 1.12654889,
      "balance_loss_mlp": 1.04945302,
      "epoch": 0.018156676486502734,
      "flos": 18588010043520.0,
      "grad_norm": 2.0579689237579877,
      "language_loss": 0.9378795,
      "learning_rate": 3.634748057771256e-06,
      "loss": 0.96301699,
      "num_input_tokens_seen": 3224955,
      "step": 151,
      "time_per_iteration": 2.7523319721221924
    },
    {
      "auxiliary_loss_clip": 0.01436746,
      "auxiliary_loss_mlp": 0.01067535,
      "balance_loss_clip": 1.12638187,
      "balance_loss_mlp": 1.04172659,
      "epoch": 0.018276919377141827,
      "flos": 25448707560960.0,
      "grad_norm": 2.482481565513713,
      "language_loss": 0.85601401,
      "learning_rate": 3.639529896584965e-06,
      "loss": 0.88105685,
      "num_input_tokens_seen": 3246330,
      "step": 152,
      "time_per_iteration": 2.812182903289795
    },
    {
      "auxiliary_loss_clip": 0.01435856,
      "auxiliary_loss_mlp": 0.01071972,
      "balance_loss_clip": 1.12612534,
      "balance_loss_mlp": 1.04574609,
      "epoch": 0.018397162267780917,
      "flos": 20047311889920.0,
      "grad_norm": 3.061464646453129,
      "language_loss": 0.88891762,
      "learning_rate": 3.6442803788531233e-06,
      "loss": 0.91399592,
      "num_input_tokens_seen": 3264290,
      "step": 153,
      "time_per_iteration": 2.7511959075927734
    },
    {
      "auxiliary_loss_clip": 0.01439962,
      "auxiliary_loss_mlp": 0.01077652,
      "balance_loss_clip": 1.12550211,
      "balance_loss_mlp": 1.0514977,
      "epoch": 0.018517405158420007,
      "flos": 27565282425600.0,
      "grad_norm": 2.473585892953188,
      "language_loss": 0.96023136,
      "learning_rate": 3.6489999131344357e-06,
      "loss": 0.98540747,
      "num_input_tokens_seen": 3287065,
      "step": 154,
      "time_per_iteration": 2.837027072906494
    },
    {
      "auxiliary_loss_clip": 0.01428734,
      "auxiliary_loss_mlp": 0.01066119,
      "balance_loss_clip": 1.12491441,
      "balance_loss_mlp": 1.04264688,
      "epoch": 0.0186376480490591,
      "flos": 19354056422400.0,
      "grad_norm": 1.8744212005900394,
      "language_loss": 0.90426761,
      "learning_rate": 3.653688900054313e-06,
      "loss": 0.92921615,
      "num_input_tokens_seen": 3305595,
      "step": 155,
      "time_per_iteration": 2.6998584270477295
    },
    {
      "auxiliary_loss_clip": 0.01430692,
      "auxiliary_loss_mlp": 0.01074298,
      "balance_loss_clip": 1.12151444,
      "balance_loss_mlp": 1.04810786,
      "epoch": 0.01875789093969819,
      "flos": 26687840993280.0,
      "grad_norm": 2.143159911420179,
      "language_loss": 0.75835758,
      "learning_rate": 3.6583477325089526e-06,
      "loss": 0.78340751,
      "num_input_tokens_seen": 3326135,
      "step": 156,
      "time_per_iteration": 2.798849105834961
    },
    {
      "auxiliary_loss_clip": 0.01429337,
      "auxiliary_loss_mlp": 0.01069369,
      "balance_loss_clip": 1.12481165,
      "balance_loss_mlp": 1.04133058,
      "epoch": 0.01887813383033728,
      "flos": 24353001135360.0,
      "grad_norm": 2.2798337698658626,
      "language_loss": 1.04201221,
      "learning_rate": 3.6629767958628916e-06,
      "loss": 1.06699944,
      "num_input_tokens_seen": 3343510,
      "step": 157,
      "time_per_iteration": 2.7249958515167236
    },
    {
      "auxiliary_loss_clip": 0.0143294,
      "auxiliary_loss_mlp": 0.01086132,
      "balance_loss_clip": 1.12559891,
      "balance_loss_mlp": 1.0616349,
      "epoch": 0.018998376720976373,
      "flos": 14647532330880.0,
      "grad_norm": 2.208392420949861,
      "language_loss": 0.85842657,
      "learning_rate": 3.667576468140291e-06,
      "loss": 0.88361734,
      "num_input_tokens_seen": 3361325,
      "step": 158,
      "time_per_iteration": 2.7555198669433594
    },
    {
      "auxiliary_loss_clip": 0.01423382,
      "auxiliary_loss_mlp": 0.01067781,
      "balance_loss_clip": 1.11920881,
      "balance_loss_mlp": 1.04265153,
      "epoch": 0.019118619611615463,
      "flos": 29305261146240.0,
      "grad_norm": 3.1108215275087976,
      "language_loss": 0.88954639,
      "learning_rate": 3.672147120210184e-06,
      "loss": 0.91445804,
      "num_input_tokens_seen": 3377925,
      "step": 159,
      "time_per_iteration": 2.744236946105957
    },
    {
      "auxiliary_loss_clip": 0.01421746,
      "auxiliary_loss_mlp": 0.01078553,
      "balance_loss_clip": 1.12347507,
      "balance_loss_mlp": 1.05454397,
      "epoch": 0.019238862502254553,
      "flos": 20886723797760.0,
      "grad_norm": 2.0268856099516204,
      "language_loss": 0.86402106,
      "learning_rate": 3.6766891159659177e-06,
      "loss": 0.88902396,
      "num_input_tokens_seen": 3396335,
      "step": 160,
      "time_per_iteration": 2.752629518508911
    },
    {
      "auxiliary_loss_clip": 0.01424919,
      "auxiliary_loss_mlp": 0.01069441,
      "balance_loss_clip": 1.1219734,
      "balance_loss_mlp": 1.0442518,
      "epoch": 0.019359105392893646,
      "flos": 21360672777600.0,
      "grad_norm": 4.069574723431663,
      "language_loss": 0.8775329,
      "learning_rate": 3.6812028124990075e-06,
      "loss": 0.90247649,
      "num_input_tokens_seen": 3413605,
      "step": 161,
      "time_per_iteration": 2.6554267406463623
    },
    {
      "auxiliary_loss_clip": 0.01413899,
      "auxiliary_loss_mlp": 0.01065284,
      "balance_loss_clip": 1.11526823,
      "balance_loss_mlp": 1.04047656,
      "epoch": 0.019479348283532736,
      "flos": 16283729681280.0,
      "grad_norm": 3.645914814587508,
      "language_loss": 0.81561714,
      "learning_rate": 3.6856885602676016e-06,
      "loss": 0.84040892,
      "num_input_tokens_seen": 3429640,
      "step": 162,
      "time_per_iteration": 2.7112302780151367
    },
    {
      "auxiliary_loss_clip": 0.0141921,
      "auxiliary_loss_mlp": 0.01070403,
      "balance_loss_clip": 1.11683297,
      "balance_loss_mlp": 1.04674006,
      "epoch": 0.019599591174171826,
      "flos": 22091239497600.0,
      "grad_norm": 2.1810945709548357,
      "language_loss": 0.94150376,
      "learning_rate": 3.6901467032597733e-06,
      "loss": 0.96639991,
      "num_input_tokens_seen": 3448125,
      "step": 163,
      "time_per_iteration": 2.733766794204712
    },
    {
      "auxiliary_loss_clip": 0.01419667,
      "auxiliary_loss_mlp": 0.01074439,
      "balance_loss_clip": 1.12046766,
      "balance_loss_mlp": 1.04805756,
      "epoch": 0.01971983406481092,
      "flos": 19609668581760.0,
      "grad_norm": 2.109186211048345,
      "language_loss": 0.87333548,
      "learning_rate": 3.694577579151804e-06,
      "loss": 0.89827651,
      "num_input_tokens_seen": 3466535,
      "step": 164,
      "time_per_iteration": 2.6854336261749268
    },
    {
      "auxiliary_loss_clip": 0.01418396,
      "auxiliary_loss_mlp": 0.01064735,
      "balance_loss_clip": 1.12055302,
      "balance_loss_mlp": 1.03986788,
      "epoch": 0.01984007695545001,
      "flos": 19099342103040.0,
      "grad_norm": 2.3304050801001117,
      "language_loss": 0.73676556,
      "learning_rate": 3.6989815194616703e-06,
      "loss": 0.76159692,
      "num_input_tokens_seen": 3483730,
      "step": 165,
      "time_per_iteration": 2.759139060974121
    },
    {
      "auxiliary_loss_clip": 0.01420701,
      "auxiliary_loss_mlp": 0.01068829,
      "balance_loss_clip": 1.11759734,
      "balance_loss_mlp": 1.04384291,
      "epoch": 0.0199603198460891,
      "flos": 20848406964480.0,
      "grad_norm": 2.2489207654864676,
      "language_loss": 0.79755867,
      "learning_rate": 3.703358849697888e-06,
      "loss": 0.82245398,
      "num_input_tokens_seen": 3503640,
      "step": 166,
      "time_per_iteration": 2.700531005859375
    },
    {
      "auxiliary_loss_clip": 0.0141159,
      "auxiliary_loss_mlp": 0.0106641,
      "balance_loss_clip": 1.11741471,
      "balance_loss_mlp": 1.04410613,
      "epoch": 0.020080562736728192,
      "flos": 21870747861120.0,
      "grad_norm": 1.7294341119100956,
      "language_loss": 0.82596946,
      "learning_rate": 3.7077098895038803e-06,
      "loss": 0.85074943,
      "num_input_tokens_seen": 3523010,
      "step": 167,
      "time_per_iteration": 2.7318639755249023
    },
    {
      "auxiliary_loss_clip": 0.01416088,
      "auxiliary_loss_mlp": 0.01066235,
      "balance_loss_clip": 1.11691654,
      "balance_loss_mlp": 1.04090285,
      "epoch": 0.020200805627367282,
      "flos": 21688788539520.0,
      "grad_norm": 3.322848819909023,
      "language_loss": 0.96673244,
      "learning_rate": 3.712034952798045e-06,
      "loss": 0.99155569,
      "num_input_tokens_seen": 3541125,
      "step": 168,
      "time_per_iteration": 2.7357845306396484
    },
    {
      "auxiliary_loss_clip": 0.01413543,
      "auxiliary_loss_mlp": 0.01069605,
      "balance_loss_clip": 1.11691141,
      "balance_loss_mlp": 1.04273558,
      "epoch": 0.02032104851800637,
      "flos": 33543043729920.0,
      "grad_norm": 2.1757137189965086,
      "language_loss": 0.84556484,
      "learning_rate": 3.7163343479096656e-06,
      "loss": 0.87039638,
      "num_input_tokens_seen": 3562700,
      "step": 169,
      "time_per_iteration": 2.817506790161133
    },
    {
      "auxiliary_loss_clip": 0.01410345,
      "auxiliary_loss_mlp": 0.01068796,
      "balance_loss_clip": 1.1162976,
      "balance_loss_mlp": 1.04572892,
      "epoch": 0.020441291408645465,
      "flos": 31686965274240.0,
      "grad_norm": 2.0076367521977145,
      "language_loss": 0.82739466,
      "learning_rate": 3.720608377710802e-06,
      "loss": 0.85218614,
      "num_input_tokens_seen": 3582790,
      "step": 170,
      "time_per_iteration": 3.701653480529785
    },
    {
      "auxiliary_loss_clip": 0.01405187,
      "auxiliary_loss_mlp": 0.01066815,
      "balance_loss_clip": 1.11157393,
      "balance_loss_mlp": 1.04135156,
      "epoch": 0.020561534299284555,
      "flos": 20886687884160.0,
      "grad_norm": 2.1410121371794224,
      "language_loss": 0.86637318,
      "learning_rate": 3.7248573397443277e-06,
      "loss": 0.89109319,
      "num_input_tokens_seen": 3601715,
      "step": 171,
      "time_per_iteration": 3.607347249984741
    },
    {
      "auxiliary_loss_clip": 0.01410584,
      "auxiliary_loss_mlp": 0.01079611,
      "balance_loss_clip": 1.11724651,
      "balance_loss_mlp": 1.0536356,
      "epoch": 0.020681777189923645,
      "flos": 20996610480000.0,
      "grad_norm": 2.013478207570808,
      "language_loss": 0.97540963,
      "learning_rate": 3.729081526348224e-06,
      "loss": 1.00031161,
      "num_input_tokens_seen": 3620245,
      "step": 172,
      "time_per_iteration": 4.59281587600708
    },
    {
      "auxiliary_loss_clip": 0.01406025,
      "auxiliary_loss_mlp": 0.01068682,
      "balance_loss_clip": 1.11373091,
      "balance_loss_mlp": 1.04472065,
      "epoch": 0.020802020080562738,
      "flos": 28257532312320.0,
      "grad_norm": 1.8447003191776574,
      "language_loss": 0.85117358,
      "learning_rate": 3.7332812247762777e-06,
      "loss": 0.87592059,
      "num_input_tokens_seen": 3641545,
      "step": 173,
      "time_per_iteration": 2.7831366062164307
    },
    {
      "auxiliary_loss_clip": 0.01403997,
      "auxiliary_loss_mlp": 0.0107497,
      "balance_loss_clip": 1.11315608,
      "balance_loss_mlp": 1.05270207,
      "epoch": 0.020922262971201828,
      "flos": 19681274344320.0,
      "grad_norm": 2.586859632048097,
      "language_loss": 0.95557338,
      "learning_rate": 3.737456717315293e-06,
      "loss": 0.98036307,
      "num_input_tokens_seen": 3660510,
      "step": 174,
      "time_per_iteration": 2.6288578510284424
    },
    {
      "auxiliary_loss_clip": 0.01401339,
      "auxiliary_loss_mlp": 0.0108346,
      "balance_loss_clip": 1.11330152,
      "balance_loss_mlp": 1.06052458,
      "epoch": 0.021042505861840918,
      "flos": 15666353694720.0,
      "grad_norm": 1.7004689553143786,
      "language_loss": 0.90733087,
      "learning_rate": 3.7416082813989552e-06,
      "loss": 0.93217891,
      "num_input_tokens_seen": 3677505,
      "step": 175,
      "time_per_iteration": 2.7079241275787354
    },
    {
      "auxiliary_loss_clip": 0.01403174,
      "auxiliary_loss_mlp": 0.01088252,
      "balance_loss_clip": 1.11100888,
      "balance_loss_mlp": 1.0629437,
      "epoch": 0.02116274875248001,
      "flos": 21142012734720.0,
      "grad_norm": 2.143233810620858,
      "language_loss": 0.89369005,
      "learning_rate": 3.745736189718439e-06,
      "loss": 0.91860431,
      "num_input_tokens_seen": 3696760,
      "step": 176,
      "time_per_iteration": 2.6605892181396484
    },
    {
      "auxiliary_loss_clip": 0.01394122,
      "auxiliary_loss_mlp": 0.01059589,
      "balance_loss_clip": 1.10825133,
      "balance_loss_mlp": 1.03721344,
      "epoch": 0.0212829916431191,
      "flos": 24715770543360.0,
      "grad_norm": 3.3079660954819037,
      "language_loss": 0.73108786,
      "learning_rate": 3.749840710329894e-06,
      "loss": 0.75562501,
      "num_input_tokens_seen": 3717465,
      "step": 177,
      "time_per_iteration": 2.7484419345855713
    },
    {
      "auxiliary_loss_clip": 0.01412669,
      "auxiliary_loss_mlp": 0.0106895,
      "balance_loss_clip": 1.11644006,
      "balance_loss_mlp": 1.04267645,
      "epoch": 0.02140323453375819,
      "flos": 16645493508480.0,
      "grad_norm": 3.3460694310099224,
      "language_loss": 0.97627079,
      "learning_rate": 3.7539221067588938e-06,
      "loss": 1.00108695,
      "num_input_tokens_seen": 3731440,
      "step": 178,
      "time_per_iteration": 2.6608803272247314
    },
    {
      "auxiliary_loss_clip": 0.01401184,
      "auxiliary_loss_mlp": 0.01076376,
      "balance_loss_clip": 1.10986352,
      "balance_loss_mlp": 1.05066228,
      "epoch": 0.021523477424397284,
      "flos": 20299332689280.0,
      "grad_norm": 4.385082777790487,
      "language_loss": 0.93709004,
      "learning_rate": 3.757980638101964e-06,
      "loss": 0.96186566,
      "num_input_tokens_seen": 3744935,
      "step": 179,
      "time_per_iteration": 2.6934523582458496
    },
    {
      "auxiliary_loss_clip": 0.0140312,
      "auxiliary_loss_mlp": 0.01082913,
      "balance_loss_clip": 1.11291373,
      "balance_loss_mlp": 1.05736637,
      "epoch": 0.021643720315036374,
      "flos": 26104005331200.0,
      "grad_norm": 2.9171192819066856,
      "language_loss": 0.89243287,
      "learning_rate": 3.7620165591252806e-06,
      "loss": 0.91729319,
      "num_input_tokens_seen": 3763035,
      "step": 180,
      "time_per_iteration": 2.785236358642578
    },
    {
      "auxiliary_loss_clip": 0.01399639,
      "auxiliary_loss_mlp": 0.01075465,
      "balance_loss_clip": 1.11308694,
      "balance_loss_mlp": 1.0543294,
      "epoch": 0.021763963205675464,
      "flos": 24787663614720.0,
      "grad_norm": 1.8630456881373627,
      "language_loss": 0.94602054,
      "learning_rate": 3.766030120360636e-06,
      "loss": 0.97077155,
      "num_input_tokens_seen": 3782665,
      "step": 181,
      "time_per_iteration": 2.7162742614746094
    },
    {
      "auxiliary_loss_clip": 0.01394915,
      "auxiliary_loss_mlp": 0.01065372,
      "balance_loss_clip": 1.10885477,
      "balance_loss_mlp": 1.04236484,
      "epoch": 0.021884206096314557,
      "flos": 25813559957760.0,
      "grad_norm": 3.2113082344119794,
      "language_loss": 0.90231967,
      "learning_rate": 3.7700215681987578e-06,
      "loss": 0.9269225,
      "num_input_tokens_seen": 3802435,
      "step": 182,
      "time_per_iteration": 2.748892307281494
    },
    {
      "auxiliary_loss_clip": 0.01397713,
      "auxiliary_loss_mlp": 0.01077639,
      "balance_loss_clip": 1.11117792,
      "balance_loss_mlp": 1.05417824,
      "epoch": 0.022004448986953647,
      "flos": 20082719721600.0,
      "grad_norm": 5.0290064404443084,
      "language_loss": 0.82320875,
      "learning_rate": 3.7739911449800767e-06,
      "loss": 0.84796226,
      "num_input_tokens_seen": 3822490,
      "step": 183,
      "time_per_iteration": 2.682551383972168
    },
    {
      "auxiliary_loss_clip": 0.0139254,
      "auxiliary_loss_mlp": 0.01081798,
      "balance_loss_clip": 1.1058991,
      "balance_loss_mlp": 1.05819452,
      "epoch": 0.022124691877592736,
      "flos": 20480609652480.0,
      "grad_norm": 2.6159138476673167,
      "language_loss": 0.80607408,
      "learning_rate": 3.7779390890830114e-06,
      "loss": 0.83081746,
      "num_input_tokens_seen": 3841140,
      "step": 184,
      "time_per_iteration": 2.71777081489563
    },
    {
      "auxiliary_loss_clip": 0.01388369,
      "auxiliary_loss_mlp": 0.01066382,
      "balance_loss_clip": 1.10517335,
      "balance_loss_mlp": 1.04205167,
      "epoch": 0.02224493476823183,
      "flos": 23586847015680.0,
      "grad_norm": 1.7441224904116959,
      "language_loss": 0.860174,
      "learning_rate": 3.7818656350098723e-06,
      "loss": 0.88472152,
      "num_input_tokens_seen": 3862090,
      "step": 185,
      "time_per_iteration": 2.7396037578582764
    },
    {
      "auxiliary_loss_clip": 0.01392149,
      "auxiliary_loss_mlp": 0.01077181,
      "balance_loss_clip": 1.10847151,
      "balance_loss_mlp": 1.05347037,
      "epoch": 0.02236517765887092,
      "flos": 16909940413440.0,
      "grad_norm": 3.28773701223614,
      "language_loss": 0.77037346,
      "learning_rate": 3.7857710134704447e-06,
      "loss": 0.79506671,
      "num_input_tokens_seen": 3881025,
      "step": 186,
      "time_per_iteration": 2.7352263927459717
    },
    {
      "auxiliary_loss_clip": 0.01384312,
      "auxiliary_loss_mlp": 0.01073251,
      "balance_loss_clip": 1.10541677,
      "balance_loss_mlp": 1.0495522,
      "epoch": 0.02248542054951001,
      "flos": 43508182930560.0,
      "grad_norm": 2.516275041734204,
      "language_loss": 0.79211593,
      "learning_rate": 3.7896554514633234e-06,
      "loss": 0.81669158,
      "num_input_tokens_seen": 3905310,
      "step": 187,
      "time_per_iteration": 2.8640170097351074
    },
    {
      "auxiliary_loss_clip": 0.01386188,
      "auxiliary_loss_mlp": 0.01058456,
      "balance_loss_clip": 1.10521626,
      "balance_loss_mlp": 1.03610396,
      "epoch": 0.022605663440149103,
      "flos": 23367648268800.0,
      "grad_norm": 1.9684343875014363,
      "language_loss": 0.84067595,
      "learning_rate": 3.7935191723550955e-06,
      "loss": 0.86512244,
      "num_input_tokens_seen": 3924265,
      "step": 188,
      "time_per_iteration": 2.7250490188598633
    },
    {
      "auxiliary_loss_clip": 0.01385011,
      "auxiliary_loss_mlp": 0.01060667,
      "balance_loss_clip": 1.10462189,
      "balance_loss_mlp": 1.03788638,
      "epoch": 0.022725906330788193,
      "flos": 29019915504000.0,
      "grad_norm": 1.9555368343417607,
      "language_loss": 0.88785356,
      "learning_rate": 3.797362395957408e-06,
      "loss": 0.91231036,
      "num_input_tokens_seen": 3944830,
      "step": 189,
      "time_per_iteration": 2.7804479598999023
    },
    {
      "auxiliary_loss_clip": 0.01388042,
      "auxiliary_loss_mlp": 0.01067998,
      "balance_loss_clip": 1.10773063,
      "balance_loss_mlp": 1.04311883,
      "epoch": 0.022846149221427282,
      "flos": 24496176746880.0,
      "grad_norm": 2.0929264196833572,
      "language_loss": 0.78068787,
      "learning_rate": 3.8011853386020055e-06,
      "loss": 0.80524826,
      "num_input_tokens_seen": 3965735,
      "step": 190,
      "time_per_iteration": 2.7155423164367676
    },
    {
      "auxiliary_loss_clip": 0.01390607,
      "auxiliary_loss_mlp": 0.01075058,
      "balance_loss_clip": 1.1090467,
      "balance_loss_mlp": 1.05036998,
      "epoch": 0.022966392112066376,
      "flos": 15523537219200.0,
      "grad_norm": 3.254710492069304,
      "language_loss": 0.89683783,
      "learning_rate": 3.804988213213804e-06,
      "loss": 0.92149448,
      "num_input_tokens_seen": 3983975,
      "step": 191,
      "time_per_iteration": 2.6694326400756836
    },
    {
      "auxiliary_loss_clip": 0.01387734,
      "auxiliary_loss_mlp": 0.01017432,
      "balance_loss_clip": 1.17144036,
      "balance_loss_mlp": 1.00532043,
      "epoch": 0.023086635002705466,
      "flos": 55650408433920.0,
      "grad_norm": 1.0196660135367834,
      "language_loss": 0.63193452,
      "learning_rate": 3.808771229382049e-06,
      "loss": 0.65598619,
      "num_input_tokens_seen": 4043440,
      "step": 192,
      "time_per_iteration": 3.1630778312683105
    },
    {
      "auxiliary_loss_clip": 0.0137717,
      "auxiliary_loss_mlp": 0.01069142,
      "balance_loss_clip": 1.10006905,
      "balance_loss_mlp": 1.04843569,
      "epoch": 0.023206877893344555,
      "flos": 19313441118720.0,
      "grad_norm": 2.055474719199043,
      "language_loss": 0.84470254,
      "learning_rate": 3.8125345934296324e-06,
      "loss": 0.86916566,
      "num_input_tokens_seen": 4061750,
      "step": 193,
      "time_per_iteration": 2.7106425762176514
    },
    {
      "auxiliary_loss_clip": 0.01384897,
      "auxiliary_loss_mlp": 0.01081176,
      "balance_loss_clip": 1.10525799,
      "balance_loss_mlp": 1.05745387,
      "epoch": 0.02332712078398365,
      "flos": 23072965090560.0,
      "grad_norm": 2.6010968155171836,
      "language_loss": 0.87956256,
      "learning_rate": 3.81627850848061e-06,
      "loss": 0.90422326,
      "num_input_tokens_seen": 4082345,
      "step": 194,
      "time_per_iteration": 2.7001044750213623
    },
    {
      "auxiliary_loss_clip": 0.01380007,
      "auxiliary_loss_mlp": 0.01071738,
      "balance_loss_clip": 1.10329127,
      "balance_loss_mlp": 1.04933858,
      "epoch": 0.02344736367462274,
      "flos": 24425971614720.0,
      "grad_norm": 2.169277772215798,
      "language_loss": 0.86128247,
      "learning_rate": 3.820003174525994e-06,
      "loss": 0.8858,
      "num_input_tokens_seen": 4101770,
      "step": 195,
      "time_per_iteration": 2.7809293270111084
    },
    {
      "auxiliary_loss_clip": 0.01379638,
      "auxiliary_loss_mlp": 0.01069782,
      "balance_loss_clip": 1.1038835,
      "balance_loss_mlp": 1.04468894,
      "epoch": 0.02356760656526183,
      "flos": 21579799697280.0,
      "grad_norm": 2.9043350723453223,
      "language_loss": 0.82842988,
      "learning_rate": 3.823708788487851e-06,
      "loss": 0.85292411,
      "num_input_tokens_seen": 4118770,
      "step": 196,
      "time_per_iteration": 2.7868335247039795
    },
    {
      "auxiliary_loss_clip": 0.01379707,
      "auxiliary_loss_mlp": 0.01073871,
      "balance_loss_clip": 1.10377216,
      "balance_loss_mlp": 1.05196023,
      "epoch": 0.02368784945590092,
      "flos": 25193598192000.0,
      "grad_norm": 1.9727864790142842,
      "language_loss": 0.84473842,
      "learning_rate": 3.827395544281781e-06,
      "loss": 0.8692742,
      "num_input_tokens_seen": 4141110,
      "step": 197,
      "time_per_iteration": 4.493316411972046
    },
    {
      "auxiliary_loss_clip": 0.01379303,
      "auxiliary_loss_mlp": 0.01089044,
      "balance_loss_clip": 1.10354948,
      "balance_loss_mlp": 1.06597662,
      "epoch": 0.02380809234654001,
      "flos": 27562481164800.0,
      "grad_norm": 1.9952628394020147,
      "language_loss": 0.79124099,
      "learning_rate": 3.831063632877802e-06,
      "loss": 0.81592441,
      "num_input_tokens_seen": 4161430,
      "step": 198,
      "time_per_iteration": 3.6180975437164307
    },
    {
      "auxiliary_loss_clip": 0.0137846,
      "auxiliary_loss_mlp": 0.01061515,
      "balance_loss_clip": 1.10738134,
      "balance_loss_mlp": 1.03958011,
      "epoch": 0.0239283352371791,
      "flos": 18259786540800.0,
      "grad_norm": 3.200289910352633,
      "language_loss": 0.7574442,
      "learning_rate": 3.834713242359712e-06,
      "loss": 0.7818439,
      "num_input_tokens_seen": 4179260,
      "step": 199,
      "time_per_iteration": 3.607980728149414
    },
    {
      "auxiliary_loss_clip": 0.0137814,
      "auxiliary_loss_mlp": 0.01066842,
      "balance_loss_clip": 1.10190654,
      "balance_loss_mlp": 1.0431428,
      "epoch": 0.02404857812781819,
      "flos": 21395110942080.0,
      "grad_norm": 2.7515519254464444,
      "language_loss": 0.87201428,
      "learning_rate": 3.838344557982959e-06,
      "loss": 0.89646411,
      "num_input_tokens_seen": 4200640,
      "step": 200,
      "time_per_iteration": 2.7081263065338135
    },
    {
      "auxiliary_loss_clip": 0.01370098,
      "auxiliary_loss_mlp": 0.01067301,
      "balance_loss_clip": 1.09836841,
      "balance_loss_mlp": 1.04568815,
      "epoch": 0.024168821018457284,
      "flos": 16654256426880.0,
      "grad_norm": 3.1781211489644114,
      "language_loss": 0.84858656,
      "learning_rate": 3.841957762231063e-06,
      "loss": 0.87296057,
      "num_input_tokens_seen": 4218170,
      "step": 201,
      "time_per_iteration": 2.659226417541504
    },
    {
      "auxiliary_loss_clip": 0.01370107,
      "auxiliary_loss_mlp": 0.01071343,
      "balance_loss_clip": 1.09710014,
      "balance_loss_mlp": 1.04964721,
      "epoch": 0.024289063909096374,
      "flos": 22820872464000.0,
      "grad_norm": 2.3406761757277157,
      "language_loss": 0.87808323,
      "learning_rate": 3.8455530348706454e-06,
      "loss": 0.90249765,
      "num_input_tokens_seen": 4237770,
      "step": 202,
      "time_per_iteration": 2.6990745067596436
    },
    {
      "auxiliary_loss_clip": 0.01372492,
      "auxiliary_loss_mlp": 0.01062201,
      "balance_loss_clip": 1.10030937,
      "balance_loss_mlp": 1.04098153,
      "epoch": 0.024409306799735464,
      "flos": 17748598135680.0,
      "grad_norm": 1.9717380135094764,
      "language_loss": 0.77401578,
      "learning_rate": 3.849130553005099e-06,
      "loss": 0.79836273,
      "num_input_tokens_seen": 4255985,
      "step": 203,
      "time_per_iteration": 2.689941167831421
    },
    {
      "auxiliary_loss_clip": 0.01368754,
      "auxiliary_loss_mlp": 0.01063101,
      "balance_loss_clip": 1.09911489,
      "balance_loss_mlp": 1.04078543,
      "epoch": 0.024529549690374557,
      "flos": 21616213109760.0,
      "grad_norm": 2.103325283558823,
      "language_loss": 0.83555341,
      "learning_rate": 3.852690491126933e-06,
      "loss": 0.85987186,
      "num_input_tokens_seen": 4276035,
      "step": 204,
      "time_per_iteration": 2.6832735538482666
    },
    {
      "auxiliary_loss_clip": 0.01374371,
      "auxiliary_loss_mlp": 0.01061436,
      "balance_loss_clip": 1.10003459,
      "balance_loss_mlp": 1.04015708,
      "epoch": 0.024649792581013647,
      "flos": 25551662918400.0,
      "grad_norm": 3.556613141061763,
      "language_loss": 0.91168368,
      "learning_rate": 3.856233021168845e-06,
      "loss": 0.93604177,
      "num_input_tokens_seen": 4295730,
      "step": 205,
      "time_per_iteration": 2.7232744693756104
    },
    {
      "auxiliary_loss_clip": 0.01361881,
      "auxiliary_loss_mlp": 0.01058844,
      "balance_loss_clip": 1.0939976,
      "balance_loss_mlp": 1.03817272,
      "epoch": 0.024770035471652737,
      "flos": 34495574544000.0,
      "grad_norm": 2.6279391336879017,
      "language_loss": 0.91106135,
      "learning_rate": 3.859758312553544e-06,
      "loss": 0.93526864,
      "num_input_tokens_seen": 4317950,
      "step": 206,
      "time_per_iteration": 2.756917715072632
    },
    {
      "auxiliary_loss_clip": 0.01368763,
      "auxiliary_loss_mlp": 0.01063592,
      "balance_loss_clip": 1.09857321,
      "balance_loss_mlp": 1.04133594,
      "epoch": 0.02489027836229183,
      "flos": 21505428587520.0,
      "grad_norm": 2.029577045432846,
      "language_loss": 0.9187212,
      "learning_rate": 3.8632665322423735e-06,
      "loss": 0.94304472,
      "num_input_tokens_seen": 4337605,
      "step": 207,
      "time_per_iteration": 2.7604188919067383
    },
    {
      "auxiliary_loss_clip": 0.01361526,
      "auxiliary_loss_mlp": 0.01066448,
      "balance_loss_clip": 1.09542036,
      "balance_loss_mlp": 1.04615879,
      "epoch": 0.02501052125293092,
      "flos": 23219013790080.0,
      "grad_norm": 6.165214836580721,
      "language_loss": 0.85940605,
      "learning_rate": 3.866757844782762e-06,
      "loss": 0.88368583,
      "num_input_tokens_seen": 4358110,
      "step": 208,
      "time_per_iteration": 2.7291061878204346
    },
    {
      "auxiliary_loss_clip": 0.01361605,
      "auxiliary_loss_mlp": 0.01075121,
      "balance_loss_clip": 1.09571183,
      "balance_loss_mlp": 1.05302,
      "epoch": 0.02513076414357001,
      "flos": 26388920010240.0,
      "grad_norm": 2.47855774911894,
      "language_loss": 0.91529131,
      "learning_rate": 3.870232412354527e-06,
      "loss": 0.93965858,
      "num_input_tokens_seen": 4374955,
      "step": 209,
      "time_per_iteration": 2.6958751678466797
    },
    {
      "auxiliary_loss_clip": 0.01365342,
      "auxiliary_loss_mlp": 0.01070656,
      "balance_loss_clip": 1.09574938,
      "balance_loss_mlp": 1.04894769,
      "epoch": 0.025251007034209103,
      "flos": 13590430047360.0,
      "grad_norm": 2.1225811415845954,
      "language_loss": 0.92692697,
      "learning_rate": 3.873690394815086e-06,
      "loss": 0.95128691,
      "num_input_tokens_seen": 4391535,
      "step": 210,
      "time_per_iteration": 2.704148054122925
    },
    {
      "auxiliary_loss_clip": 0.01366294,
      "auxiliary_loss_mlp": 0.0105962,
      "balance_loss_clip": 1.09800601,
      "balance_loss_mlp": 1.03927088,
      "epoch": 0.025371249924848193,
      "flos": 15049229103360.0,
      "grad_norm": 13.003292359778728,
      "language_loss": 0.91199493,
      "learning_rate": 3.877131949743587e-06,
      "loss": 0.93625408,
      "num_input_tokens_seen": 4408400,
      "step": 211,
      "time_per_iteration": 2.692728281021118
    },
    {
      "auxiliary_loss_clip": 0.01359485,
      "auxiliary_loss_mlp": 0.01061634,
      "balance_loss_clip": 1.09381807,
      "balance_loss_mlp": 1.04044998,
      "epoch": 0.025491492815487283,
      "flos": 25553853648000.0,
      "grad_norm": 3.817600752169996,
      "language_loss": 0.7809149,
      "learning_rate": 3.880557232483993e-06,
      "loss": 0.80512613,
      "num_input_tokens_seen": 4427840,
      "step": 212,
      "time_per_iteration": 2.775796413421631
    },
    {
      "auxiliary_loss_clip": 0.01357592,
      "auxiliary_loss_mlp": 0.01056672,
      "balance_loss_clip": 1.09125209,
      "balance_loss_mlp": 1.03631139,
      "epoch": 0.025611735706126376,
      "flos": 20630752502400.0,
      "grad_norm": 1.9348574651845132,
      "language_loss": 0.86673051,
      "learning_rate": 3.883966396187164e-06,
      "loss": 0.89087319,
      "num_input_tokens_seen": 4447110,
      "step": 213,
      "time_per_iteration": 2.773782253265381
    },
    {
      "auxiliary_loss_clip": 0.01357898,
      "auxiliary_loss_mlp": 0.01065793,
      "balance_loss_clip": 1.09486008,
      "balance_loss_mlp": 1.04490805,
      "epoch": 0.025731978596765466,
      "flos": 19062282245760.0,
      "grad_norm": 2.6232203632264395,
      "language_loss": 0.89797944,
      "learning_rate": 3.887359591851937e-06,
      "loss": 0.92221636,
      "num_input_tokens_seen": 4464715,
      "step": 214,
      "time_per_iteration": 2.668405771255493
    },
    {
      "auxiliary_loss_clip": 0.0135696,
      "auxiliary_loss_mlp": 0.01070314,
      "balance_loss_clip": 1.09464169,
      "balance_loss_mlp": 1.05023932,
      "epoch": 0.025852221487404556,
      "flos": 22163814927360.0,
      "grad_norm": 1.8396707831657761,
      "language_loss": 0.92265213,
      "learning_rate": 3.890736968365265e-06,
      "loss": 0.94692487,
      "num_input_tokens_seen": 4485030,
      "step": 215,
      "time_per_iteration": 2.6965348720550537
    },
    {
      "auxiliary_loss_clip": 0.01354457,
      "auxiliary_loss_mlp": 0.01050971,
      "balance_loss_clip": 1.09159636,
      "balance_loss_mlp": 1.02937007,
      "epoch": 0.02597246437804365,
      "flos": 26541971861760.0,
      "grad_norm": 1.939547369646986,
      "language_loss": 0.8492285,
      "learning_rate": 3.894098672541412e-06,
      "loss": 0.87328279,
      "num_input_tokens_seen": 4505935,
      "step": 216,
      "time_per_iteration": 2.718322277069092
    },
    {
      "auxiliary_loss_clip": 0.01358146,
      "auxiliary_loss_mlp": 0.01070121,
      "balance_loss_clip": 1.09356809,
      "balance_loss_mlp": 1.04902077,
      "epoch": 0.02609270726868274,
      "flos": 32671671696000.0,
      "grad_norm": 1.8225429636783088,
      "language_loss": 0.75406915,
      "learning_rate": 3.89744484916025e-06,
      "loss": 0.77835178,
      "num_input_tokens_seen": 4527045,
      "step": 217,
      "time_per_iteration": 2.829684019088745
    },
    {
      "auxiliary_loss_clip": 0.01359738,
      "auxiliary_loss_mlp": 0.01061686,
      "balance_loss_clip": 1.0942173,
      "balance_loss_mlp": 1.03965569,
      "epoch": 0.02621295015932183,
      "flos": 26243553669120.0,
      "grad_norm": 2.0058874775593725,
      "language_loss": 0.87323701,
      "learning_rate": 3.900775641004673e-06,
      "loss": 0.89745122,
      "num_input_tokens_seen": 4546360,
      "step": 218,
      "time_per_iteration": 2.7772092819213867
    },
    {
      "auxiliary_loss_clip": 0.01362935,
      "auxiliary_loss_mlp": 0.01066346,
      "balance_loss_clip": 1.09626436,
      "balance_loss_mlp": 1.04287338,
      "epoch": 0.026333193049960922,
      "flos": 42921402353280.0,
      "grad_norm": 2.960122245288127,
      "language_loss": 0.74437767,
      "learning_rate": 3.904091188897156e-06,
      "loss": 0.76867044,
      "num_input_tokens_seen": 4565495,
      "step": 219,
      "time_per_iteration": 2.9297783374786377
    },
    {
      "auxiliary_loss_clip": 0.01359912,
      "auxiliary_loss_mlp": 0.01067537,
      "balance_loss_clip": 1.09447396,
      "balance_loss_mlp": 1.04557908,
      "epoch": 0.026453435940600012,
      "flos": 17963846386560.0,
      "grad_norm": 2.0298664325234066,
      "language_loss": 0.8192997,
      "learning_rate": 3.90739163173548e-06,
      "loss": 0.84357417,
      "num_input_tokens_seen": 4583330,
      "step": 220,
      "time_per_iteration": 2.7693541049957275
    },
    {
      "auxiliary_loss_clip": 0.01356187,
      "auxiliary_loss_mlp": 0.01048305,
      "balance_loss_clip": 1.09329796,
      "balance_loss_mlp": 1.02801538,
      "epoch": 0.026573678831239102,
      "flos": 18984319776000.0,
      "grad_norm": 2.374874771320717,
      "language_loss": 0.88084781,
      "learning_rate": 3.910677106527646e-06,
      "loss": 0.90489274,
      "num_input_tokens_seen": 4600520,
      "step": 221,
      "time_per_iteration": 2.7835569381713867
    },
    {
      "auxiliary_loss_clip": 0.01351427,
      "auxiliary_loss_mlp": 0.01066193,
      "balance_loss_clip": 1.09182298,
      "balance_loss_mlp": 1.04636836,
      "epoch": 0.026693921721878195,
      "flos": 29241448634880.0,
      "grad_norm": 2.284663811863766,
      "language_loss": 0.84000146,
      "learning_rate": 3.913947748426004e-06,
      "loss": 0.8641777,
      "num_input_tokens_seen": 4617340,
      "step": 222,
      "time_per_iteration": 2.7775650024414062
    },
    {
      "auxiliary_loss_clip": 0.01355981,
      "auxiliary_loss_mlp": 0.01054237,
      "balance_loss_clip": 1.0937686,
      "balance_loss_mlp": 1.03394771,
      "epoch": 0.026814164612517285,
      "flos": 14128083797760.0,
      "grad_norm": 2.760770655916662,
      "language_loss": 0.76178062,
      "learning_rate": 3.9172036907606136e-06,
      "loss": 0.78588283,
      "num_input_tokens_seen": 4630820,
      "step": 223,
      "time_per_iteration": 3.577866315841675
    },
    {
      "auxiliary_loss_clip": 0.01351998,
      "auxiliary_loss_mlp": 0.01064684,
      "balance_loss_clip": 1.0895499,
      "balance_loss_mlp": 1.04314291,
      "epoch": 0.026934407503156375,
      "flos": 23511973115520.0,
      "grad_norm": 1.7666934401793748,
      "language_loss": 0.95082068,
      "learning_rate": 3.920445065071855e-06,
      "loss": 0.97498751,
      "num_input_tokens_seen": 4651985,
      "step": 224,
      "time_per_iteration": 3.657883644104004
    },
    {
      "auxiliary_loss_clip": 0.01355131,
      "auxiliary_loss_mlp": 0.01056447,
      "balance_loss_clip": 1.09389293,
      "balance_loss_mlp": 1.03687322,
      "epoch": 0.027054650393795468,
      "flos": 28950356816640.0,
      "grad_norm": 2.2370036531447233,
      "language_loss": 0.79889721,
      "learning_rate": 3.923672001142322e-06,
      "loss": 0.82301301,
      "num_input_tokens_seen": 4672295,
      "step": 225,
      "time_per_iteration": 3.6633412837982178
    },
    {
      "auxiliary_loss_clip": 0.01349938,
      "auxiliary_loss_mlp": 0.01066852,
      "balance_loss_clip": 1.08887017,
      "balance_loss_mlp": 1.04502439,
      "epoch": 0.027174893284434558,
      "flos": 31431568596480.0,
      "grad_norm": 1.8776618456063423,
      "language_loss": 0.84347028,
      "learning_rate": 3.926884627027996e-06,
      "loss": 0.86763817,
      "num_input_tokens_seen": 4696065,
      "step": 226,
      "time_per_iteration": 3.6602859497070312
    },
    {
      "auxiliary_loss_clip": 0.01346401,
      "auxiliary_loss_mlp": 0.01067774,
      "balance_loss_clip": 1.08773077,
      "balance_loss_mlp": 1.04916549,
      "epoch": 0.027295136175073648,
      "flos": 22054466949120.0,
      "grad_norm": 1.9544126421754122,
      "language_loss": 0.77598101,
      "learning_rate": 3.930083069088744e-06,
      "loss": 0.80012274,
      "num_input_tokens_seen": 4716065,
      "step": 227,
      "time_per_iteration": 2.6831133365631104
    },
    {
      "auxiliary_loss_clip": 0.0133468,
      "auxiliary_loss_mlp": 0.0101034,
      "balance_loss_clip": 1.13474405,
      "balance_loss_mlp": 1.00018334,
      "epoch": 0.02741537906571274,
      "flos": 60800752972800.0,
      "grad_norm": 0.9869608786419882,
      "language_loss": 0.59321207,
      "learning_rate": 3.933267452018137e-06,
      "loss": 0.61666226,
      "num_input_tokens_seen": 4775860,
      "step": 228,
      "time_per_iteration": 3.2391090393066406
    },
    {
      "auxiliary_loss_clip": 0.01348307,
      "auxiliary_loss_mlp": 0.01069996,
      "balance_loss_clip": 1.09302735,
      "balance_loss_mlp": 1.0495038,
      "epoch": 0.02753562195635183,
      "flos": 24606278910720.0,
      "grad_norm": 2.373849544444169,
      "language_loss": 0.84407276,
      "learning_rate": 3.936437898872622e-06,
      "loss": 0.86825579,
      "num_input_tokens_seen": 4795835,
      "step": 229,
      "time_per_iteration": 2.7046124935150146
    },
    {
      "auxiliary_loss_clip": 0.01352373,
      "auxiliary_loss_mlp": 0.01058024,
      "balance_loss_clip": 1.09058845,
      "balance_loss_mlp": 1.03896272,
      "epoch": 0.02765586484699092,
      "flos": 34094236907520.0,
      "grad_norm": 2.794993024392365,
      "language_loss": 0.79793596,
      "learning_rate": 3.9395945311000525e-06,
      "loss": 0.82203996,
      "num_input_tokens_seen": 4817460,
      "step": 230,
      "time_per_iteration": 2.8646390438079834
    },
    {
      "auxiliary_loss_clip": 0.01351001,
      "auxiliary_loss_mlp": 0.01062398,
      "balance_loss_clip": 1.0898087,
      "balance_loss_mlp": 1.0424664,
      "epoch": 0.027776107737630014,
      "flos": 14829922615680.0,
      "grad_norm": 3.723560560190765,
      "language_loss": 0.90817654,
      "learning_rate": 3.942737468567608e-06,
      "loss": 0.93231046,
      "num_input_tokens_seen": 4835475,
      "step": 231,
      "time_per_iteration": 2.754484176635742
    },
    {
      "auxiliary_loss_clip": 0.01349596,
      "auxiliary_loss_mlp": 0.01071364,
      "balance_loss_clip": 1.0896405,
      "balance_loss_mlp": 1.04889286,
      "epoch": 0.027896350628269104,
      "flos": 47920347066240.0,
      "grad_norm": 2.556045947182385,
      "language_loss": 0.8623867,
      "learning_rate": 3.9458668295891026e-06,
      "loss": 0.88659638,
      "num_input_tokens_seen": 4857760,
      "step": 232,
      "time_per_iteration": 2.972388982772827
    },
    {
      "auxiliary_loss_clip": 0.01344712,
      "auxiliary_loss_mlp": 0.01047586,
      "balance_loss_clip": 1.08578265,
      "balance_loss_mlp": 1.02825069,
      "epoch": 0.028016593518908194,
      "flos": 21684550734720.0,
      "grad_norm": 2.4899485523183627,
      "language_loss": 0.86840755,
      "learning_rate": 3.948982730951712e-06,
      "loss": 0.89233053,
      "num_input_tokens_seen": 4875855,
      "step": 233,
      "time_per_iteration": 2.778571128845215
    },
    {
      "auxiliary_loss_clip": 0.01352215,
      "auxiliary_loss_mlp": 0.01064662,
      "balance_loss_clip": 1.09134102,
      "balance_loss_mlp": 1.0430609,
      "epoch": 0.028136836409547287,
      "flos": 18439483305600.0,
      "grad_norm": 2.377556586178967,
      "language_loss": 0.82117772,
      "learning_rate": 3.9520852879421254e-06,
      "loss": 0.84534645,
      "num_input_tokens_seen": 4893200,
      "step": 234,
      "time_per_iteration": 2.6536307334899902
    },
    {
      "auxiliary_loss_clip": 0.01342038,
      "auxiliary_loss_mlp": 0.01057875,
      "balance_loss_clip": 1.08670402,
      "balance_loss_mlp": 1.03882504,
      "epoch": 0.028257079300186377,
      "flos": 31576934937600.0,
      "grad_norm": 2.706803517057041,
      "language_loss": 0.81870466,
      "learning_rate": 3.955174614372137e-06,
      "loss": 0.84270382,
      "num_input_tokens_seen": 4912965,
      "step": 235,
      "time_per_iteration": 2.7041773796081543
    },
    {
      "auxiliary_loss_clip": 0.01348325,
      "auxiliary_loss_mlp": 0.01067592,
      "balance_loss_clip": 1.08983624,
      "balance_loss_mlp": 1.04746985,
      "epoch": 0.028377322190825467,
      "flos": 23513337832320.0,
      "grad_norm": 3.638066717470407,
      "language_loss": 0.84350473,
      "learning_rate": 3.9582508226037045e-06,
      "loss": 0.86766392,
      "num_input_tokens_seen": 4933105,
      "step": 236,
      "time_per_iteration": 2.7125213146209717
    },
    {
      "auxiliary_loss_clip": 0.01348085,
      "auxiliary_loss_mlp": 0.0106656,
      "balance_loss_clip": 1.08931589,
      "balance_loss_mlp": 1.04486418,
      "epoch": 0.02849756508146456,
      "flos": 20479604071680.0,
      "grad_norm": 4.308818017675863,
      "language_loss": 0.9413892,
      "learning_rate": 3.9613140235734636e-06,
      "loss": 0.96553564,
      "num_input_tokens_seen": 4950085,
      "step": 237,
      "time_per_iteration": 2.6339480876922607
    },
    {
      "auxiliary_loss_clip": 0.0134467,
      "auxiliary_loss_mlp": 0.01067314,
      "balance_loss_clip": 1.0876863,
      "balance_loss_mlp": 1.04708433,
      "epoch": 0.02861780797210365,
      "flos": 14283362292480.0,
      "grad_norm": 2.0670559194349334,
      "language_loss": 0.81187743,
      "learning_rate": 3.96436432681674e-06,
      "loss": 0.83599728,
      "num_input_tokens_seen": 4968075,
      "step": 238,
      "time_per_iteration": 2.678725004196167
    },
    {
      "auxiliary_loss_clip": 0.0134601,
      "auxiliary_loss_mlp": 0.01058781,
      "balance_loss_clip": 1.08830392,
      "balance_loss_mlp": 1.03883731,
      "epoch": 0.02873805086274274,
      "flos": 25808532053760.0,
      "grad_norm": 2.434832016649358,
      "language_loss": 0.89231443,
      "learning_rate": 3.967401840491044e-06,
      "loss": 0.91636229,
      "num_input_tokens_seen": 4987355,
      "step": 239,
      "time_per_iteration": 2.7481400966644287
    },
    {
      "auxiliary_loss_clip": 0.01343489,
      "auxiliary_loss_mlp": 0.01071726,
      "balance_loss_clip": 1.08856606,
      "balance_loss_mlp": 1.05165064,
      "epoch": 0.028858293753381833,
      "flos": 17304238984320.0,
      "grad_norm": 2.238358419806448,
      "language_loss": 0.87721354,
      "learning_rate": 3.97042667139909e-06,
      "loss": 0.90136564,
      "num_input_tokens_seen": 5004680,
      "step": 240,
      "time_per_iteration": 2.664344549179077
    },
    {
      "auxiliary_loss_clip": 0.01344444,
      "auxiliary_loss_mlp": 0.01057876,
      "balance_loss_clip": 1.08965027,
      "balance_loss_mlp": 1.03671598,
      "epoch": 0.028978536644020923,
      "flos": 23038347358080.0,
      "grad_norm": 2.130332197407333,
      "language_loss": 0.87527931,
      "learning_rate": 3.973438925011327e-06,
      "loss": 0.89930248,
      "num_input_tokens_seen": 5022965,
      "step": 241,
      "time_per_iteration": 2.7137465476989746
    },
    {
      "auxiliary_loss_clip": 0.01344648,
      "auxiliary_loss_mlp": 0.01063251,
      "balance_loss_clip": 1.08736086,
      "balance_loss_mlp": 1.04385567,
      "epoch": 0.029098779534660012,
      "flos": 28329712692480.0,
      "grad_norm": 2.284064681198569,
      "language_loss": 0.91277969,
      "learning_rate": 3.976438705488002e-06,
      "loss": 0.93685871,
      "num_input_tokens_seen": 5042625,
      "step": 242,
      "time_per_iteration": 2.687079906463623
    },
    {
      "auxiliary_loss_clip": 0.01342574,
      "auxiliary_loss_mlp": 0.0105818,
      "balance_loss_clip": 1.09006476,
      "balance_loss_mlp": 1.03902268,
      "epoch": 0.029219022425299106,
      "flos": 13881665520000.0,
      "grad_norm": 2.9503076579019822,
      "language_loss": 0.92902613,
      "learning_rate": 3.9794261157007744e-06,
      "loss": 0.95303363,
      "num_input_tokens_seen": 5060380,
      "step": 243,
      "time_per_iteration": 2.6521639823913574
    },
    {
      "auxiliary_loss_clip": 0.01347869,
      "auxiliary_loss_mlp": 0.01071535,
      "balance_loss_clip": 1.09082711,
      "balance_loss_mlp": 1.04971933,
      "epoch": 0.029339265315938196,
      "flos": 19422501788160.0,
      "grad_norm": 2.0857838080122044,
      "language_loss": 0.8482095,
      "learning_rate": 3.982401257253887e-06,
      "loss": 0.8724035,
      "num_input_tokens_seen": 5078720,
      "step": 244,
      "time_per_iteration": 2.681300640106201
    },
    {
      "auxiliary_loss_clip": 0.01341591,
      "auxiliary_loss_mlp": 0.01049504,
      "balance_loss_clip": 1.08732557,
      "balance_loss_mlp": 1.03109837,
      "epoch": 0.029459508206577285,
      "flos": 15669550005120.0,
      "grad_norm": 2.3976617348060927,
      "language_loss": 0.89639938,
      "learning_rate": 3.985364230504893e-06,
      "loss": 0.92031038,
      "num_input_tokens_seen": 5096605,
      "step": 245,
      "time_per_iteration": 2.69746732711792
    },
    {
      "auxiliary_loss_clip": 0.01342913,
      "auxiliary_loss_mlp": 0.01061634,
      "balance_loss_clip": 1.08769083,
      "balance_loss_mlp": 1.04175019,
      "epoch": 0.02957975109721638,
      "flos": 28220975245440.0,
      "grad_norm": 1.9809464072377714,
      "language_loss": 0.84273672,
      "learning_rate": 3.988315134584976e-06,
      "loss": 0.86678219,
      "num_input_tokens_seen": 5116285,
      "step": 246,
      "time_per_iteration": 2.753222942352295
    },
    {
      "auxiliary_loss_clip": 0.01345533,
      "auxiliary_loss_mlp": 0.01061471,
      "balance_loss_clip": 1.08980727,
      "balance_loss_mlp": 1.04224288,
      "epoch": 0.02969999398785547,
      "flos": 24315869450880.0,
      "grad_norm": 1.797197459118882,
      "language_loss": 0.80585629,
      "learning_rate": 3.991254067418851e-06,
      "loss": 0.82992631,
      "num_input_tokens_seen": 5136825,
      "step": 247,
      "time_per_iteration": 2.7297940254211426
    },
    {
      "auxiliary_loss_clip": 0.01338759,
      "auxiliary_loss_mlp": 0.01060145,
      "balance_loss_clip": 1.08798909,
      "balance_loss_mlp": 1.04173923,
      "epoch": 0.02982023687849456,
      "flos": 35078584193280.0,
      "grad_norm": 2.260179795159027,
      "language_loss": 0.82688993,
      "learning_rate": 3.994181125744254e-06,
      "loss": 0.85087901,
      "num_input_tokens_seen": 5158630,
      "step": 248,
      "time_per_iteration": 2.771456718444824
    },
    {
      "auxiliary_loss_clip": 0.0134043,
      "auxiliary_loss_mlp": 0.01057327,
      "balance_loss_clip": 1.08870614,
      "balance_loss_mlp": 1.03834915,
      "epoch": 0.02994047976913365,
      "flos": 26177155378560.0,
      "grad_norm": 2.0669164763145074,
      "language_loss": 0.73925662,
      "learning_rate": 3.99709640513106e-06,
      "loss": 0.76323426,
      "num_input_tokens_seen": 5179510,
      "step": 249,
      "time_per_iteration": 3.649968147277832
    },
    {
      "auxiliary_loss_clip": 0.01342174,
      "auxiliary_loss_mlp": 0.01067167,
      "balance_loss_clip": 1.08887959,
      "balance_loss_mlp": 1.0471164,
      "epoch": 0.03006072265977274,
      "flos": 25625028447360.0,
      "grad_norm": 3.4602588582004534,
      "language_loss": 0.85728574,
      "learning_rate": 4e-06,
      "loss": 0.88137913,
      "num_input_tokens_seen": 5199345,
      "step": 250,
      "time_per_iteration": 2.704350709915161
    },
    {
      "auxiliary_loss_clip": 0.01345185,
      "auxiliary_loss_mlp": 0.01063719,
      "balance_loss_clip": 1.09113336,
      "balance_loss_mlp": 1.04377556,
      "epoch": 0.03018096555041183,
      "flos": 22127078292480.0,
      "grad_norm": 2.9091942192213764,
      "language_loss": 0.887025,
      "learning_rate": 3.999999848300794e-06,
      "loss": 0.91111398,
      "num_input_tokens_seen": 5218330,
      "step": 251,
      "time_per_iteration": 3.6359775066375732
    },
    {
      "auxiliary_loss_clip": 0.01334993,
      "auxiliary_loss_mlp": 0.01065801,
      "balance_loss_clip": 1.08472395,
      "balance_loss_mlp": 1.04741907,
      "epoch": 0.030301208441050925,
      "flos": 30188197359360.0,
      "grad_norm": 1.7336457611264133,
      "language_loss": 0.89083165,
      "learning_rate": 3.999999393203203e-06,
      "loss": 0.91483957,
      "num_input_tokens_seen": 5240740,
      "step": 252,
      "time_per_iteration": 3.6692140102386475
    },
    {
      "auxiliary_loss_clip": 0.01335445,
      "auxiliary_loss_mlp": 0.01062689,
      "balance_loss_clip": 1.08498359,
      "balance_loss_mlp": 1.04264951,
      "epoch": 0.030421451331690014,
      "flos": 23621392920960.0,
      "grad_norm": 2.0492958441941425,
      "language_loss": 0.85130203,
      "learning_rate": 3.999998634707293e-06,
      "loss": 0.87528336,
      "num_input_tokens_seen": 5260290,
      "step": 253,
      "time_per_iteration": 2.7231366634368896
    },
    {
      "auxiliary_loss_clip": 0.01342733,
      "auxiliary_loss_mlp": 0.01070677,
      "balance_loss_clip": 1.089993,
      "balance_loss_mlp": 1.05060196,
      "epoch": 0.030541694222329104,
      "flos": 27928446883200.0,
      "grad_norm": 2.161956721384621,
      "language_loss": 0.96453905,
      "learning_rate": 3.999997572813182e-06,
      "loss": 0.98867309,
      "num_input_tokens_seen": 5278100,
      "step": 254,
      "time_per_iteration": 3.585601806640625
    },
    {
      "auxiliary_loss_clip": 0.01334662,
      "auxiliary_loss_mlp": 0.01054046,
      "balance_loss_clip": 1.0869453,
      "balance_loss_mlp": 1.03580713,
      "epoch": 0.030661937112968194,
      "flos": 18588441006720.0,
      "grad_norm": 1.9011231721078339,
      "language_loss": 0.87489724,
      "learning_rate": 3.999996207521028e-06,
      "loss": 0.8987844,
      "num_input_tokens_seen": 5296810,
      "step": 255,
      "time_per_iteration": 2.6965291500091553
    },
    {
      "auxiliary_loss_clip": 0.01337497,
      "auxiliary_loss_mlp": 0.01061369,
      "balance_loss_clip": 1.08637857,
      "balance_loss_mlp": 1.04072237,
      "epoch": 0.030782180003607287,
      "flos": 12969139478400.0,
      "grad_norm": 2.4840798588232986,
      "language_loss": 0.81944323,
      "learning_rate": 3.999994538831039e-06,
      "loss": 0.84343195,
      "num_input_tokens_seen": 5313395,
      "step": 256,
      "time_per_iteration": 2.7494046688079834
    },
    {
      "auxiliary_loss_clip": 0.01334121,
      "auxiliary_loss_mlp": 0.01060036,
      "balance_loss_clip": 1.08442736,
      "balance_loss_mlp": 1.03999722,
      "epoch": 0.030902422894246377,
      "flos": 23335364920320.0,
      "grad_norm": 2.741851552231918,
      "language_loss": 0.85795528,
      "learning_rate": 3.99999256674347e-06,
      "loss": 0.88189685,
      "num_input_tokens_seen": 5333545,
      "step": 257,
      "time_per_iteration": 2.738823890686035
    },
    {
      "auxiliary_loss_clip": 0.01313265,
      "auxiliary_loss_mlp": 0.01018576,
      "balance_loss_clip": 1.11683297,
      "balance_loss_mlp": 1.00808549,
      "epoch": 0.031022665784885467,
      "flos": 55094151438720.0,
      "grad_norm": 1.0090079864146269,
      "language_loss": 0.53497308,
      "learning_rate": 3.999990291258618e-06,
      "loss": 0.55829144,
      "num_input_tokens_seen": 5392235,
      "step": 258,
      "time_per_iteration": 3.2375783920288086
    },
    {
      "auxiliary_loss_clip": 0.01340199,
      "auxiliary_loss_mlp": 0.01064169,
      "balance_loss_clip": 1.0909133,
      "balance_loss_mlp": 1.04466641,
      "epoch": 0.03114290867552456,
      "flos": 19317786664320.0,
      "grad_norm": 3.0243520222376272,
      "language_loss": 0.86719823,
      "learning_rate": 3.999987712376829e-06,
      "loss": 0.89124191,
      "num_input_tokens_seen": 5410555,
      "step": 259,
      "time_per_iteration": 2.63279390335083
    },
    {
      "auxiliary_loss_clip": 0.01333569,
      "auxiliary_loss_mlp": 0.01064223,
      "balance_loss_clip": 1.08639205,
      "balance_loss_mlp": 1.04375446,
      "epoch": 0.031263151566163654,
      "flos": 20959442881920.0,
      "grad_norm": 2.4065767753133187,
      "language_loss": 0.82247615,
      "learning_rate": 3.999984830098494e-06,
      "loss": 0.84645402,
      "num_input_tokens_seen": 5430135,
      "step": 260,
      "time_per_iteration": 2.7147560119628906
    },
    {
      "auxiliary_loss_clip": 0.01331466,
      "auxiliary_loss_mlp": 0.01053058,
      "balance_loss_clip": 1.08425927,
      "balance_loss_mlp": 1.03341222,
      "epoch": 0.03138339445680274,
      "flos": 14793006412800.0,
      "grad_norm": 4.66319158942458,
      "language_loss": 0.98433584,
      "learning_rate": 3.999981644424051e-06,
      "loss": 1.0081811,
      "num_input_tokens_seen": 5444935,
      "step": 261,
      "time_per_iteration": 2.68422532081604
    },
    {
      "auxiliary_loss_clip": 0.01338071,
      "auxiliary_loss_mlp": 0.01069575,
      "balance_loss_clip": 1.08772779,
      "balance_loss_mlp": 1.04873776,
      "epoch": 0.03150363734744183,
      "flos": 11655599022720.0,
      "grad_norm": 2.431294449869067,
      "language_loss": 0.86173052,
      "learning_rate": 3.999978155353982e-06,
      "loss": 0.88580704,
      "num_input_tokens_seen": 5462080,
      "step": 262,
      "time_per_iteration": 2.789813995361328
    },
    {
      "auxiliary_loss_clip": 0.01333394,
      "auxiliary_loss_mlp": 0.01059259,
      "balance_loss_clip": 1.08367026,
      "balance_loss_mlp": 1.03974426,
      "epoch": 0.03162388023808092,
      "flos": 33727732485120.0,
      "grad_norm": 2.2692784019642254,
      "language_loss": 0.80039096,
      "learning_rate": 3.9999743628888186e-06,
      "loss": 0.82431751,
      "num_input_tokens_seen": 5483870,
      "step": 263,
      "time_per_iteration": 2.789564609527588
    },
    {
      "auxiliary_loss_clip": 0.01332279,
      "auxiliary_loss_mlp": 0.0105584,
      "balance_loss_clip": 1.0839963,
      "balance_loss_mlp": 1.03723145,
      "epoch": 0.03174412312872001,
      "flos": 20810952057600.0,
      "grad_norm": 2.339746730496139,
      "language_loss": 0.8961612,
      "learning_rate": 3.999970267029133e-06,
      "loss": 0.9200424,
      "num_input_tokens_seen": 5502830,
      "step": 264,
      "time_per_iteration": 2.6982367038726807
    },
    {
      "auxiliary_loss_clip": 0.01330216,
      "auxiliary_loss_mlp": 0.01059934,
      "balance_loss_clip": 1.0854727,
      "balance_loss_mlp": 1.03977561,
      "epoch": 0.0318643660193591,
      "flos": 23727939638400.0,
      "grad_norm": 2.0060184471477878,
      "language_loss": 0.79990101,
      "learning_rate": 3.999965867775548e-06,
      "loss": 0.82380253,
      "num_input_tokens_seen": 5523225,
      "step": 265,
      "time_per_iteration": 2.6986820697784424
    },
    {
      "auxiliary_loss_clip": 0.01334351,
      "auxiliary_loss_mlp": 0.01052304,
      "balance_loss_clip": 1.08569264,
      "balance_loss_mlp": 1.03418386,
      "epoch": 0.0319846089099982,
      "flos": 13917863450880.0,
      "grad_norm": 2.4530576700240143,
      "language_loss": 0.8695246,
      "learning_rate": 3.9999611651287315e-06,
      "loss": 0.89339113,
      "num_input_tokens_seen": 5541380,
      "step": 266,
      "time_per_iteration": 2.673325777053833
    },
    {
      "auxiliary_loss_clip": 0.01332973,
      "auxiliary_loss_mlp": 0.01058216,
      "balance_loss_clip": 1.08457983,
      "balance_loss_mlp": 1.03787923,
      "epoch": 0.03210485180063729,
      "flos": 14753253035520.0,
      "grad_norm": 4.924322790995798,
      "language_loss": 0.78662729,
      "learning_rate": 3.999956159089396e-06,
      "loss": 0.81053913,
      "num_input_tokens_seen": 5558830,
      "step": 267,
      "time_per_iteration": 2.698577404022217
    },
    {
      "auxiliary_loss_clip": 0.01331411,
      "auxiliary_loss_mlp": 0.01065573,
      "balance_loss_clip": 1.0863322,
      "balance_loss_mlp": 1.04727435,
      "epoch": 0.03222509469127638,
      "flos": 28913153304960.0,
      "grad_norm": 2.0804776966187197,
      "language_loss": 0.79668897,
      "learning_rate": 3.999950849658302e-06,
      "loss": 0.8206588,
      "num_input_tokens_seen": 5577750,
      "step": 268,
      "time_per_iteration": 2.754279851913452
    },
    {
      "auxiliary_loss_clip": 0.01334471,
      "auxiliary_loss_mlp": 0.01064549,
      "balance_loss_clip": 1.08531642,
      "balance_loss_mlp": 1.04473615,
      "epoch": 0.03234533758191547,
      "flos": 16946389739520.0,
      "grad_norm": 2.6706541920455504,
      "language_loss": 0.84314668,
      "learning_rate": 3.999945236836254e-06,
      "loss": 0.86713684,
      "num_input_tokens_seen": 5596715,
      "step": 269,
      "time_per_iteration": 2.7007744312286377
    },
    {
      "auxiliary_loss_clip": 0.01335398,
      "auxiliary_loss_mlp": 0.01065218,
      "balance_loss_clip": 1.08700883,
      "balance_loss_mlp": 1.04485714,
      "epoch": 0.03246558047255456,
      "flos": 18989096284800.0,
      "grad_norm": 3.177205879447411,
      "language_loss": 0.94793367,
      "learning_rate": 3.999939320624103e-06,
      "loss": 0.9719398,
      "num_input_tokens_seen": 5611865,
      "step": 270,
      "time_per_iteration": 2.7495241165161133
    },
    {
      "auxiliary_loss_clip": 0.01329732,
      "auxiliary_loss_mlp": 0.01068375,
      "balance_loss_clip": 1.08411133,
      "balance_loss_mlp": 1.04883647,
      "epoch": 0.03258582336319365,
      "flos": 23728334688000.0,
      "grad_norm": 1.8674007917933901,
      "language_loss": 0.89979607,
      "learning_rate": 3.999933101022749e-06,
      "loss": 0.9237771,
      "num_input_tokens_seen": 5632270,
      "step": 271,
      "time_per_iteration": 2.7252659797668457
    },
    {
      "auxiliary_loss_clip": 0.01331895,
      "auxiliary_loss_mlp": 0.01052261,
      "balance_loss_clip": 1.08836126,
      "balance_loss_mlp": 1.03300881,
      "epoch": 0.032706066253832745,
      "flos": 27670823562240.0,
      "grad_norm": 1.867226103730768,
      "language_loss": 0.86988771,
      "learning_rate": 3.999926578033132e-06,
      "loss": 0.89372927,
      "num_input_tokens_seen": 5652085,
      "step": 272,
      "time_per_iteration": 2.6898910999298096
    },
    {
      "auxiliary_loss_clip": 0.01332849,
      "auxiliary_loss_mlp": 0.01060149,
      "balance_loss_clip": 1.0857079,
      "balance_loss_mlp": 1.04018128,
      "epoch": 0.032826309144471835,
      "flos": 45624685968000.0,
      "grad_norm": 2.6508221365500626,
      "language_loss": 0.63324946,
      "learning_rate": 3.999919751656244e-06,
      "loss": 0.65717947,
      "num_input_tokens_seen": 5678985,
      "step": 273,
      "time_per_iteration": 2.9197139739990234
    },
    {
      "auxiliary_loss_clip": 0.01331014,
      "auxiliary_loss_mlp": 0.01057466,
      "balance_loss_clip": 1.08564317,
      "balance_loss_mlp": 1.0391916,
      "epoch": 0.032946552035110925,
      "flos": 25812374808960.0,
      "grad_norm": 3.720886186327365,
      "language_loss": 0.75995743,
      "learning_rate": 3.9999126218931195e-06,
      "loss": 0.78384221,
      "num_input_tokens_seen": 5697020,
      "step": 274,
      "time_per_iteration": 2.6745901107788086
    },
    {
      "auxiliary_loss_clip": 0.01330418,
      "auxiliary_loss_mlp": 0.010713,
      "balance_loss_clip": 1.08678365,
      "balance_loss_mlp": 1.05356216,
      "epoch": 0.033066794925750015,
      "flos": 15121984101120.0,
      "grad_norm": 2.3093727712041865,
      "language_loss": 0.894436,
      "learning_rate": 3.99990518874484e-06,
      "loss": 0.91845322,
      "num_input_tokens_seen": 5713460,
      "step": 275,
      "time_per_iteration": 2.645709753036499
    },
    {
      "auxiliary_loss_clip": 0.01332308,
      "auxiliary_loss_mlp": 0.01067478,
      "balance_loss_clip": 1.08806872,
      "balance_loss_mlp": 1.04875016,
      "epoch": 0.033187037816389105,
      "flos": 22776593973120.0,
      "grad_norm": 5.36614730396745,
      "language_loss": 0.9265433,
      "learning_rate": 3.999897452212534e-06,
      "loss": 0.95054114,
      "num_input_tokens_seen": 5730790,
      "step": 276,
      "time_per_iteration": 3.6108014583587646
    },
    {
      "auxiliary_loss_clip": 0.01330633,
      "auxiliary_loss_mlp": 0.01069527,
      "balance_loss_clip": 1.08387327,
      "balance_loss_mlp": 1.04914188,
      "epoch": 0.033307280707028195,
      "flos": 23331414424320.0,
      "grad_norm": 2.007433340654826,
      "language_loss": 1.00389564,
      "learning_rate": 3.999889412297374e-06,
      "loss": 1.02789724,
      "num_input_tokens_seen": 5750215,
      "step": 277,
      "time_per_iteration": 3.600492000579834
    },
    {
      "auxiliary_loss_clip": 0.01329864,
      "auxiliary_loss_mlp": 0.01059298,
      "balance_loss_clip": 1.08548903,
      "balance_loss_mlp": 1.03910446,
      "epoch": 0.03342752359766729,
      "flos": 28840290566400.0,
      "grad_norm": 5.412266055259722,
      "language_loss": 0.7915169,
      "learning_rate": 3.999881069000581e-06,
      "loss": 0.81540859,
      "num_input_tokens_seen": 5769945,
      "step": 278,
      "time_per_iteration": 3.7194409370422363
    },
    {
      "auxiliary_loss_clip": 0.01326784,
      "auxiliary_loss_mlp": 0.01054641,
      "balance_loss_clip": 1.08401179,
      "balance_loss_mlp": 1.03544843,
      "epoch": 0.03354776648830638,
      "flos": 19384544090880.0,
      "grad_norm": 2.512472138473977,
      "language_loss": 0.86967021,
      "learning_rate": 3.99987242232342e-06,
      "loss": 0.89348447,
      "num_input_tokens_seen": 5784950,
      "step": 279,
      "time_per_iteration": 2.7840583324432373
    },
    {
      "auxiliary_loss_clip": 0.01330092,
      "auxiliary_loss_mlp": 0.01059539,
      "balance_loss_clip": 1.08681631,
      "balance_loss_mlp": 1.04033422,
      "epoch": 0.03366800937894547,
      "flos": 17858628472320.0,
      "grad_norm": 2.0775584614929263,
      "language_loss": 0.79687029,
      "learning_rate": 3.9998634722672026e-06,
      "loss": 0.82076657,
      "num_input_tokens_seen": 5805005,
      "step": 280,
      "time_per_iteration": 3.6582541465759277
    },
    {
      "auxiliary_loss_clip": 0.01330469,
      "auxiliary_loss_mlp": 0.01063478,
      "balance_loss_clip": 1.08722067,
      "balance_loss_mlp": 1.04410684,
      "epoch": 0.03378825226958456,
      "flos": 35951033635200.0,
      "grad_norm": 1.945510130965831,
      "language_loss": 0.78442359,
      "learning_rate": 3.999854218833286e-06,
      "loss": 0.80836308,
      "num_input_tokens_seen": 5825825,
      "step": 281,
      "time_per_iteration": 2.789396047592163
    },
    {
      "auxiliary_loss_clip": 0.01329322,
      "auxiliary_loss_mlp": 0.01074479,
      "balance_loss_clip": 1.08695626,
      "balance_loss_mlp": 1.05473757,
      "epoch": 0.03390849516022365,
      "flos": 25702488126720.0,
      "grad_norm": 3.22136030577526,
      "language_loss": 0.82282412,
      "learning_rate": 3.999844662023075e-06,
      "loss": 0.84686214,
      "num_input_tokens_seen": 5845700,
      "step": 282,
      "time_per_iteration": 2.715961456298828
    },
    {
      "auxiliary_loss_clip": 0.01321581,
      "auxiliary_loss_mlp": 0.01062309,
      "balance_loss_clip": 1.08139455,
      "balance_loss_mlp": 1.04310441,
      "epoch": 0.03402873805086274,
      "flos": 21284505987840.0,
      "grad_norm": 2.08542584073003,
      "language_loss": 0.92024875,
      "learning_rate": 3.999834801838018e-06,
      "loss": 0.94408762,
      "num_input_tokens_seen": 5864680,
      "step": 283,
      "time_per_iteration": 2.7861926555633545
    },
    {
      "auxiliary_loss_clip": 0.01327507,
      "auxiliary_loss_mlp": 0.01060625,
      "balance_loss_clip": 1.08660173,
      "balance_loss_mlp": 1.04251671,
      "epoch": 0.03414898094150183,
      "flos": 22710913954560.0,
      "grad_norm": 2.210332678685034,
      "language_loss": 0.73841798,
      "learning_rate": 3.9998246382796115e-06,
      "loss": 0.7622993,
      "num_input_tokens_seen": 5884260,
      "step": 284,
      "time_per_iteration": 2.7307846546173096
    },
    {
      "auxiliary_loss_clip": 0.01327847,
      "auxiliary_loss_mlp": 0.01061502,
      "balance_loss_clip": 1.08504248,
      "balance_loss_mlp": 1.04267883,
      "epoch": 0.03426922383214093,
      "flos": 18879927874560.0,
      "grad_norm": 5.983992762047023,
      "language_loss": 0.90770316,
      "learning_rate": 3.999814171349399e-06,
      "loss": 0.93159664,
      "num_input_tokens_seen": 5902120,
      "step": 285,
      "time_per_iteration": 2.680018663406372
    },
    {
      "auxiliary_loss_clip": 0.01323022,
      "auxiliary_loss_mlp": 0.0105247,
      "balance_loss_clip": 1.08363426,
      "balance_loss_mlp": 1.03491044,
      "epoch": 0.03438946672278002,
      "flos": 34752012716160.0,
      "grad_norm": 3.1191517137275575,
      "language_loss": 0.73603314,
      "learning_rate": 3.9998034010489655e-06,
      "loss": 0.75978804,
      "num_input_tokens_seen": 5925810,
      "step": 286,
      "time_per_iteration": 2.875169515609741
    },
    {
      "auxiliary_loss_clip": 0.01325629,
      "auxiliary_loss_mlp": 0.01059788,
      "balance_loss_clip": 1.08522511,
      "balance_loss_mlp": 1.04158437,
      "epoch": 0.03450970961341911,
      "flos": 22164102236160.0,
      "grad_norm": 2.0317332111661837,
      "language_loss": 0.76008719,
      "learning_rate": 3.999792327379946e-06,
      "loss": 0.78394133,
      "num_input_tokens_seen": 5945185,
      "step": 287,
      "time_per_iteration": 2.6936113834381104
    },
    {
      "auxiliary_loss_clip": 0.01327652,
      "auxiliary_loss_mlp": 0.01057565,
      "balance_loss_clip": 1.08708119,
      "balance_loss_mlp": 1.03825343,
      "epoch": 0.034629952504058197,
      "flos": 21725740656000.0,
      "grad_norm": 2.0885222687021066,
      "language_loss": 0.96361196,
      "learning_rate": 3.999780950344021e-06,
      "loss": 0.98746413,
      "num_input_tokens_seen": 5963375,
      "step": 288,
      "time_per_iteration": 2.710160493850708
    },
    {
      "auxiliary_loss_clip": 0.01329198,
      "auxiliary_loss_mlp": 0.01055755,
      "balance_loss_clip": 1.0864017,
      "balance_loss_mlp": 1.03570366,
      "epoch": 0.034750195394697286,
      "flos": 20047994248320.0,
      "grad_norm": 1.9079784791415575,
      "language_loss": 0.83028495,
      "learning_rate": 3.999769269942916e-06,
      "loss": 0.85413456,
      "num_input_tokens_seen": 5983415,
      "step": 289,
      "time_per_iteration": 2.733022451400757
    },
    {
      "auxiliary_loss_clip": 0.01321769,
      "auxiliary_loss_mlp": 0.01055872,
      "balance_loss_clip": 1.08379507,
      "balance_loss_mlp": 1.03664339,
      "epoch": 0.034870438285336376,
      "flos": 27965865876480.0,
      "grad_norm": 8.917787922156819,
      "language_loss": 0.81099391,
      "learning_rate": 3.999757286178402e-06,
      "loss": 0.83477032,
      "num_input_tokens_seen": 6005850,
      "step": 290,
      "time_per_iteration": 2.7998557090759277
    },
    {
      "auxiliary_loss_clip": 0.01326486,
      "auxiliary_loss_mlp": 0.01063042,
      "balance_loss_clip": 1.08628845,
      "balance_loss_mlp": 1.04417109,
      "epoch": 0.03499068117597547,
      "flos": 22017514832640.0,
      "grad_norm": 1.7574948632895206,
      "language_loss": 0.90764606,
      "learning_rate": 3.999744999052299e-06,
      "loss": 0.93154132,
      "num_input_tokens_seen": 6027240,
      "step": 291,
      "time_per_iteration": 2.7823214530944824
    },
    {
      "auxiliary_loss_clip": 0.0128704,
      "auxiliary_loss_mlp": 0.01013491,
      "balance_loss_clip": 1.10277903,
      "balance_loss_mlp": 1.00342929,
      "epoch": 0.03511092406661456,
      "flos": 57242147725440.0,
      "grad_norm": 0.964081432299059,
      "language_loss": 0.61178768,
      "learning_rate": 3.9997324085664675e-06,
      "loss": 0.63479292,
      "num_input_tokens_seen": 6087470,
      "step": 292,
      "time_per_iteration": 3.2987613677978516
    },
    {
      "auxiliary_loss_clip": 0.01322582,
      "auxiliary_loss_mlp": 0.0105824,
      "balance_loss_clip": 1.08284974,
      "balance_loss_mlp": 1.04088283,
      "epoch": 0.03523116695725365,
      "flos": 22928065626240.0,
      "grad_norm": 4.084045732936985,
      "language_loss": 0.92101932,
      "learning_rate": 3.999719514722821e-06,
      "loss": 0.94482756,
      "num_input_tokens_seen": 6107600,
      "step": 293,
      "time_per_iteration": 2.7961180210113525
    },
    {
      "auxiliary_loss_clip": 0.01316473,
      "auxiliary_loss_mlp": 0.01064212,
      "balance_loss_clip": 1.08168936,
      "balance_loss_mlp": 1.04630661,
      "epoch": 0.03535140984789274,
      "flos": 36903241226880.0,
      "grad_norm": 2.942634049713418,
      "language_loss": 0.74823284,
      "learning_rate": 3.999706317523314e-06,
      "loss": 0.77203965,
      "num_input_tokens_seen": 6126160,
      "step": 294,
      "time_per_iteration": 2.8931026458740234
    },
    {
      "auxiliary_loss_clip": 0.01319422,
      "auxiliary_loss_mlp": 0.01056571,
      "balance_loss_clip": 1.08375442,
      "balance_loss_mlp": 1.03843927,
      "epoch": 0.03547165273853183,
      "flos": 20449152316800.0,
      "grad_norm": 2.2386162267260046,
      "language_loss": 0.86008883,
      "learning_rate": 3.999692816969948e-06,
      "loss": 0.88384879,
      "num_input_tokens_seen": 6145695,
      "step": 295,
      "time_per_iteration": 2.745386838912964
    },
    {
      "auxiliary_loss_clip": 0.01277518,
      "auxiliary_loss_mlp": 0.01010998,
      "balance_loss_clip": 1.09604788,
      "balance_loss_mlp": 1.00084138,
      "epoch": 0.03559189562917092,
      "flos": 69850564871040.0,
      "grad_norm": 1.006546332665757,
      "language_loss": 0.69426966,
      "learning_rate": 3.999679013064772e-06,
      "loss": 0.71715474,
      "num_input_tokens_seen": 6212440,
      "step": 296,
      "time_per_iteration": 3.282848596572876
    },
    {
      "auxiliary_loss_clip": 0.01316366,
      "auxiliary_loss_mlp": 0.01058533,
      "balance_loss_clip": 1.08076227,
      "balance_loss_mlp": 1.04012692,
      "epoch": 0.03571213851981002,
      "flos": 21651944163840.0,
      "grad_norm": 2.4490698384917264,
      "language_loss": 0.85304523,
      "learning_rate": 3.99966490580988e-06,
      "loss": 0.87679422,
      "num_input_tokens_seen": 6229800,
      "step": 297,
      "time_per_iteration": 2.7566468715667725
    },
    {
      "auxiliary_loss_clip": 0.01322227,
      "auxiliary_loss_mlp": 0.01066187,
      "balance_loss_clip": 1.08458424,
      "balance_loss_mlp": 1.04800749,
      "epoch": 0.03583238141044911,
      "flos": 43945610757120.0,
      "grad_norm": 2.4280701355997985,
      "language_loss": 0.65834558,
      "learning_rate": 3.999650495207411e-06,
      "loss": 0.68222976,
      "num_input_tokens_seen": 6255825,
      "step": 298,
      "time_per_iteration": 2.928084373474121
    },
    {
      "auxiliary_loss_clip": 0.01320002,
      "auxiliary_loss_mlp": 0.01057809,
      "balance_loss_clip": 1.08412218,
      "balance_loss_mlp": 1.03768611,
      "epoch": 0.0359526243010882,
      "flos": 18910810592640.0,
      "grad_norm": 3.8044984192631723,
      "language_loss": 0.90089405,
      "learning_rate": 3.999635781259553e-06,
      "loss": 0.92467213,
      "num_input_tokens_seen": 6271090,
      "step": 299,
      "time_per_iteration": 2.6696536540985107
    },
    {
      "auxiliary_loss_clip": 0.01267765,
      "auxiliary_loss_mlp": 0.01013133,
      "balance_loss_clip": 1.09013486,
      "balance_loss_mlp": 1.00311947,
      "epoch": 0.03607286719172729,
      "flos": 61668892782720.0,
      "grad_norm": 0.920493545733023,
      "language_loss": 0.52254307,
      "learning_rate": 3.999620763968535e-06,
      "loss": 0.5453521,
      "num_input_tokens_seen": 6329965,
      "step": 300,
      "time_per_iteration": 3.111809492111206
    },
    {
      "auxiliary_loss_clip": 0.01316884,
      "auxiliary_loss_mlp": 0.0106411,
      "balance_loss_clip": 1.08282757,
      "balance_loss_mlp": 1.04606152,
      "epoch": 0.03619311008236638,
      "flos": 27819062991360.0,
      "grad_norm": 1.7392908028723926,
      "language_loss": 0.86369812,
      "learning_rate": 3.999605443336638e-06,
      "loss": 0.88750803,
      "num_input_tokens_seen": 6352095,
      "step": 301,
      "time_per_iteration": 2.7056374549865723
    },
    {
      "auxiliary_loss_clip": 0.01319593,
      "auxiliary_loss_mlp": 0.01062115,
      "balance_loss_clip": 1.08157349,
      "balance_loss_mlp": 1.04227853,
      "epoch": 0.03631335297300547,
      "flos": 13621133197440.0,
      "grad_norm": 2.5684757588850347,
      "language_loss": 0.89244145,
      "learning_rate": 3.999589819366185e-06,
      "loss": 0.91625857,
      "num_input_tokens_seen": 6365885,
      "step": 302,
      "time_per_iteration": 2.612623453140259
    },
    {
      "auxiliary_loss_clip": 0.01322307,
      "auxiliary_loss_mlp": 0.01056153,
      "balance_loss_clip": 1.08431971,
      "balance_loss_mlp": 1.03790164,
      "epoch": 0.036433595863644565,
      "flos": 27631788456960.0,
      "grad_norm": 2.960710930831052,
      "language_loss": 0.84683073,
      "learning_rate": 3.999573892059547e-06,
      "loss": 0.8706153,
      "num_input_tokens_seen": 6385015,
      "step": 303,
      "time_per_iteration": 3.6686787605285645
    },
    {
      "auxiliary_loss_clip": 0.01323174,
      "auxiliary_loss_mlp": 0.01053787,
      "balance_loss_clip": 1.08400941,
      "balance_loss_mlp": 1.03460622,
      "epoch": 0.036553838754283655,
      "flos": 24572020314240.0,
      "grad_norm": 1.9658339668209999,
      "language_loss": 0.81036204,
      "learning_rate": 3.999557661419138e-06,
      "loss": 0.83413166,
      "num_input_tokens_seen": 6405165,
      "step": 304,
      "time_per_iteration": 4.5113770961761475
    },
    {
      "auxiliary_loss_clip": 0.01319909,
      "auxiliary_loss_mlp": 0.01057978,
      "balance_loss_clip": 1.0834105,
      "balance_loss_mlp": 1.04023933,
      "epoch": 0.036674081644922744,
      "flos": 23404313076480.0,
      "grad_norm": 2.457993836763286,
      "language_loss": 0.81225455,
      "learning_rate": 3.9995411274474225e-06,
      "loss": 0.83603346,
      "num_input_tokens_seen": 6424445,
      "step": 305,
      "time_per_iteration": 2.813042640686035
    },
    {
      "auxiliary_loss_clip": 0.01322504,
      "auxiliary_loss_mlp": 0.01049888,
      "balance_loss_clip": 1.08614635,
      "balance_loss_mlp": 1.03150594,
      "epoch": 0.036794324535561834,
      "flos": 27489690253440.0,
      "grad_norm": 11.821719772457628,
      "language_loss": 0.81777579,
      "learning_rate": 3.999524290146908e-06,
      "loss": 0.84149969,
      "num_input_tokens_seen": 6444650,
      "step": 306,
      "time_per_iteration": 3.6807756423950195
    },
    {
      "auxiliary_loss_clip": 0.01318192,
      "auxiliary_loss_mlp": 0.01064017,
      "balance_loss_clip": 1.0812788,
      "balance_loss_mlp": 1.04531312,
      "epoch": 0.036914567426200924,
      "flos": 19463476227840.0,
      "grad_norm": 2.371346813488224,
      "language_loss": 0.92085314,
      "learning_rate": 3.9995071495201485e-06,
      "loss": 0.94467521,
      "num_input_tokens_seen": 6461755,
      "step": 307,
      "time_per_iteration": 2.708165168762207
    },
    {
      "auxiliary_loss_clip": 0.01320551,
      "auxiliary_loss_mlp": 0.01061696,
      "balance_loss_clip": 1.08453202,
      "balance_loss_mlp": 1.04218102,
      "epoch": 0.037034810316840014,
      "flos": 22309324922880.0,
      "grad_norm": 2.7102188315558973,
      "language_loss": 0.98055506,
      "learning_rate": 3.999489705569744e-06,
      "loss": 1.00437748,
      "num_input_tokens_seen": 6479455,
      "step": 308,
      "time_per_iteration": 2.7222626209259033
    },
    {
      "auxiliary_loss_clip": 0.01317948,
      "auxiliary_loss_mlp": 0.01059042,
      "balance_loss_clip": 1.0810914,
      "balance_loss_mlp": 1.0403266,
      "epoch": 0.03715505320747911,
      "flos": 18588333265920.0,
      "grad_norm": 3.0075478299910454,
      "language_loss": 0.86244941,
      "learning_rate": 3.999471958298341e-06,
      "loss": 0.88621926,
      "num_input_tokens_seen": 6498365,
      "step": 309,
      "time_per_iteration": 2.773552894592285
    },
    {
      "auxiliary_loss_clip": 0.01321597,
      "auxiliary_loss_mlp": 0.0105902,
      "balance_loss_clip": 1.08356643,
      "balance_loss_mlp": 1.03995883,
      "epoch": 0.0372752960981182,
      "flos": 35955343267200.0,
      "grad_norm": 1.8692003301890903,
      "language_loss": 0.76116639,
      "learning_rate": 3.999453907708631e-06,
      "loss": 0.78497255,
      "num_input_tokens_seen": 6520770,
      "step": 310,
      "time_per_iteration": 2.8258137702941895
    },
    {
      "auxiliary_loss_clip": 0.01317394,
      "auxiliary_loss_mlp": 0.01049011,
      "balance_loss_clip": 1.08132231,
      "balance_loss_mlp": 1.0327152,
      "epoch": 0.03739553898875729,
      "flos": 20814040627200.0,
      "grad_norm": 1.8464873603471188,
      "language_loss": 0.81401056,
      "learning_rate": 3.999435553803353e-06,
      "loss": 0.83767462,
      "num_input_tokens_seen": 6540170,
      "step": 311,
      "time_per_iteration": 2.6896297931671143
    },
    {
      "auxiliary_loss_clip": 0.01315405,
      "auxiliary_loss_mlp": 0.01050329,
      "balance_loss_clip": 1.08071184,
      "balance_loss_mlp": 1.03281713,
      "epoch": 0.03751578187939638,
      "flos": 20264140339200.0,
      "grad_norm": 2.3327441891367733,
      "language_loss": 0.83382356,
      "learning_rate": 3.999416896585292e-06,
      "loss": 0.85748088,
      "num_input_tokens_seen": 6557200,
      "step": 312,
      "time_per_iteration": 2.609515905380249
    },
    {
      "auxiliary_loss_clip": 0.01313563,
      "auxiliary_loss_mlp": 0.01059269,
      "balance_loss_clip": 1.07904983,
      "balance_loss_mlp": 1.04075599,
      "epoch": 0.03763602477003547,
      "flos": 20668063754880.0,
      "grad_norm": 8.367455210623413,
      "language_loss": 0.85775554,
      "learning_rate": 3.9993979360572775e-06,
      "loss": 0.88148379,
      "num_input_tokens_seen": 6577340,
      "step": 313,
      "time_per_iteration": 2.6352968215942383
    },
    {
      "auxiliary_loss_clip": 0.01323002,
      "auxiliary_loss_mlp": 0.01059054,
      "balance_loss_clip": 1.08635175,
      "balance_loss_mlp": 1.04057693,
      "epoch": 0.03775626766067456,
      "flos": 16691352197760.0,
      "grad_norm": 3.503847974872001,
      "language_loss": 0.82556355,
      "learning_rate": 3.999378672222185e-06,
      "loss": 0.84938413,
      "num_input_tokens_seen": 6595125,
      "step": 314,
      "time_per_iteration": 2.61641263961792
    },
    {
      "auxiliary_loss_clip": 0.01311844,
      "auxiliary_loss_mlp": 0.01068888,
      "balance_loss_clip": 1.08003175,
      "balance_loss_mlp": 1.0507803,
      "epoch": 0.03787651055131366,
      "flos": 21141797253120.0,
      "grad_norm": 1.9492685489274622,
      "language_loss": 0.82933319,
      "learning_rate": 3.9993591050829385e-06,
      "loss": 0.85314047,
      "num_input_tokens_seen": 6612990,
      "step": 315,
      "time_per_iteration": 2.754539966583252
    },
    {
      "auxiliary_loss_clip": 0.0131925,
      "auxiliary_loss_mlp": 0.01059061,
      "balance_loss_clip": 1.0850234,
      "balance_loss_mlp": 1.04047632,
      "epoch": 0.037996753441952746,
      "flos": 22018089450240.0,
      "grad_norm": 1.9007872301534665,
      "language_loss": 0.79154676,
      "learning_rate": 3.999339234642506e-06,
      "loss": 0.81532985,
      "num_input_tokens_seen": 6632740,
      "step": 316,
      "time_per_iteration": 2.7447831630706787
    },
    {
      "auxiliary_loss_clip": 0.01315253,
      "auxiliary_loss_mlp": 0.01061114,
      "balance_loss_clip": 1.08205259,
      "balance_loss_mlp": 1.04303038,
      "epoch": 0.038116996332591836,
      "flos": 27709391790720.0,
      "grad_norm": 5.9566805299857455,
      "language_loss": 0.83794707,
      "learning_rate": 3.9993190609038994e-06,
      "loss": 0.86171073,
      "num_input_tokens_seen": 6651505,
      "step": 317,
      "time_per_iteration": 2.6997358798980713
    },
    {
      "auxiliary_loss_clip": 0.01313626,
      "auxiliary_loss_mlp": 0.01055286,
      "balance_loss_clip": 1.08262026,
      "balance_loss_mlp": 1.03710675,
      "epoch": 0.038237239223230926,
      "flos": 21178067011200.0,
      "grad_norm": 2.142058489601009,
      "language_loss": 0.83120823,
      "learning_rate": 3.999298583870182e-06,
      "loss": 0.85489726,
      "num_input_tokens_seen": 6671090,
      "step": 318,
      "time_per_iteration": 2.62884783744812
    },
    {
      "auxiliary_loss_clip": 0.01313336,
      "auxiliary_loss_mlp": 0.01055347,
      "balance_loss_clip": 1.08001947,
      "balance_loss_mlp": 1.03667879,
      "epoch": 0.038357482113870016,
      "flos": 25556618995200.0,
      "grad_norm": 2.016708247899925,
      "language_loss": 0.77369535,
      "learning_rate": 3.999277803544458e-06,
      "loss": 0.79738218,
      "num_input_tokens_seen": 6691245,
      "step": 319,
      "time_per_iteration": 2.6756930351257324
    },
    {
      "auxiliary_loss_clip": 0.01241233,
      "auxiliary_loss_mlp": 0.01015842,
      "balance_loss_clip": 1.07434034,
      "balance_loss_mlp": 1.0065434,
      "epoch": 0.038477725004509106,
      "flos": 59227578034560.0,
      "grad_norm": 0.9554338029834426,
      "language_loss": 0.62385643,
      "learning_rate": 3.999256719929882e-06,
      "loss": 0.64642715,
      "num_input_tokens_seen": 6752520,
      "step": 320,
      "time_per_iteration": 3.1467599868774414
    },
    {
      "auxiliary_loss_clip": 0.01241174,
      "auxiliary_loss_mlp": 0.01012175,
      "balance_loss_clip": 1.07448196,
      "balance_loss_mlp": 1.0029242,
      "epoch": 0.0385979678951482,
      "flos": 67317676398720.0,
      "grad_norm": 1.2292841025612062,
      "language_loss": 0.67077219,
      "learning_rate": 3.999235333029651e-06,
      "loss": 0.69330573,
      "num_input_tokens_seen": 6806460,
      "step": 321,
      "time_per_iteration": 3.1459403038024902
    },
    {
      "auxiliary_loss_clip": 0.01314791,
      "auxiliary_loss_mlp": 0.01060542,
      "balance_loss_clip": 1.08322525,
      "balance_loss_mlp": 1.04331684,
      "epoch": 0.03871821078578729,
      "flos": 22746752749440.0,
      "grad_norm": 3.0889597050778166,
      "language_loss": 0.8211453,
      "learning_rate": 3.999213642847009e-06,
      "loss": 0.84489858,
      "num_input_tokens_seen": 6827045,
      "step": 322,
      "time_per_iteration": 2.719306230545044
    },
    {
      "auxiliary_loss_clip": 0.01311107,
      "auxiliary_loss_mlp": 0.01066396,
      "balance_loss_clip": 1.08145475,
      "balance_loss_mlp": 1.04892004,
      "epoch": 0.03883845367642638,
      "flos": 26280613526400.0,
      "grad_norm": 5.902250473781829,
      "language_loss": 0.91215533,
      "learning_rate": 3.999191649385247e-06,
      "loss": 0.93593037,
      "num_input_tokens_seen": 6848220,
      "step": 323,
      "time_per_iteration": 2.71667742729187
    },
    {
      "auxiliary_loss_clip": 0.01236088,
      "auxiliary_loss_mlp": 0.01009712,
      "balance_loss_clip": 1.07167816,
      "balance_loss_mlp": 1.00031817,
      "epoch": 0.03895869656706547,
      "flos": 56962835568000.0,
      "grad_norm": 0.905337378477428,
      "language_loss": 0.59770799,
      "learning_rate": 3.999169352647702e-06,
      "loss": 0.62016594,
      "num_input_tokens_seen": 6909400,
      "step": 324,
      "time_per_iteration": 3.156677722930908
    },
    {
      "auxiliary_loss_clip": 0.01316129,
      "auxiliary_loss_mlp": 0.01070261,
      "balance_loss_clip": 1.08344948,
      "balance_loss_mlp": 1.05075836,
      "epoch": 0.03907893945770456,
      "flos": 24863363527680.0,
      "grad_norm": 1.8501606357250273,
      "language_loss": 0.83130682,
      "learning_rate": 3.999146752637755e-06,
      "loss": 0.85517079,
      "num_input_tokens_seen": 6930445,
      "step": 325,
      "time_per_iteration": 2.683835029602051
    },
    {
      "auxiliary_loss_clip": 0.0131518,
      "auxiliary_loss_mlp": 0.0105782,
      "balance_loss_clip": 1.08185172,
      "balance_loss_mlp": 1.03935432,
      "epoch": 0.03919918234834365,
      "flos": 18368595815040.0,
      "grad_norm": 2.682868641336022,
      "language_loss": 0.89205086,
      "learning_rate": 3.999123849358836e-06,
      "loss": 0.9157809,
      "num_input_tokens_seen": 6948110,
      "step": 326,
      "time_per_iteration": 2.672957181930542
    },
    {
      "auxiliary_loss_clip": 0.0131163,
      "auxiliary_loss_mlp": 0.01055847,
      "balance_loss_clip": 1.08043766,
      "balance_loss_mlp": 1.03745282,
      "epoch": 0.03931942523898275,
      "flos": 25225414663680.0,
      "grad_norm": 4.261353862598198,
      "language_loss": 0.74749565,
      "learning_rate": 3.999100642814418e-06,
      "loss": 0.7711705,
      "num_input_tokens_seen": 6968550,
      "step": 327,
      "time_per_iteration": 2.6969008445739746
    },
    {
      "auxiliary_loss_clip": 0.01313191,
      "auxiliary_loss_mlp": 0.01059462,
      "balance_loss_clip": 1.08309019,
      "balance_loss_mlp": 1.04210496,
      "epoch": 0.03943966812962184,
      "flos": 23257905240960.0,
      "grad_norm": 3.2470002138302885,
      "language_loss": 0.88196522,
      "learning_rate": 3.999077133008022e-06,
      "loss": 0.90569174,
      "num_input_tokens_seen": 6987135,
      "step": 328,
      "time_per_iteration": 2.6409268379211426
    },
    {
      "auxiliary_loss_clip": 0.01315889,
      "auxiliary_loss_mlp": 0.01055558,
      "balance_loss_clip": 1.08357167,
      "balance_loss_mlp": 1.03709257,
      "epoch": 0.03955991102026093,
      "flos": 29168837291520.0,
      "grad_norm": 1.9703598520233967,
      "language_loss": 0.90680027,
      "learning_rate": 3.9990533199432145e-06,
      "loss": 0.93051475,
      "num_input_tokens_seen": 7008630,
      "step": 329,
      "time_per_iteration": 3.6552486419677734
    },
    {
      "auxiliary_loss_clip": 0.01315667,
      "auxiliary_loss_mlp": 0.01060758,
      "balance_loss_clip": 1.08353353,
      "balance_loss_mlp": 1.04231644,
      "epoch": 0.03968015391090002,
      "flos": 17602441695360.0,
      "grad_norm": 2.3818577613506404,
      "language_loss": 0.75638235,
      "learning_rate": 3.999029203623608e-06,
      "loss": 0.7801466,
      "num_input_tokens_seen": 7026350,
      "step": 330,
      "time_per_iteration": 3.5881452560424805
    },
    {
      "auxiliary_loss_clip": 0.01310566,
      "auxiliary_loss_mlp": 0.01058272,
      "balance_loss_clip": 1.08129394,
      "balance_loss_mlp": 1.04122508,
      "epoch": 0.03980039680153911,
      "flos": 21799285752960.0,
      "grad_norm": 2.107938380303504,
      "language_loss": 0.87223387,
      "learning_rate": 3.99900478405286e-06,
      "loss": 0.8959223,
      "num_input_tokens_seen": 7045660,
      "step": 331,
      "time_per_iteration": 3.5931057929992676
    },
    {
      "auxiliary_loss_clip": 0.01310661,
      "auxiliary_loss_mlp": 0.01051016,
      "balance_loss_clip": 1.08240366,
      "balance_loss_mlp": 1.03396928,
      "epoch": 0.0399206396921782,
      "flos": 15195134148480.0,
      "grad_norm": 2.5831935954435363,
      "language_loss": 0.82481706,
      "learning_rate": 3.998980061234676e-06,
      "loss": 0.84843385,
      "num_input_tokens_seen": 7063575,
      "step": 332,
      "time_per_iteration": 2.6629538536071777
    },
    {
      "auxiliary_loss_clip": 0.01316398,
      "auxiliary_loss_mlp": 0.01054225,
      "balance_loss_clip": 1.08415699,
      "balance_loss_mlp": 1.03608119,
      "epoch": 0.040040882582817294,
      "flos": 14422910630400.0,
      "grad_norm": 3.2765878375970265,
      "language_loss": 0.75989079,
      "learning_rate": 3.9989550351728055e-06,
      "loss": 0.78359705,
      "num_input_tokens_seen": 7080505,
      "step": 333,
      "time_per_iteration": 3.4921488761901855
    },
    {
      "auxiliary_loss_clip": 0.01312418,
      "auxiliary_loss_mlp": 0.010673,
      "balance_loss_clip": 1.08288419,
      "balance_loss_mlp": 1.04985976,
      "epoch": 0.040161125473456384,
      "flos": 19280906375040.0,
      "grad_norm": 2.5549747578588637,
      "language_loss": 0.84467292,
      "learning_rate": 3.998929705871046e-06,
      "loss": 0.86847007,
      "num_input_tokens_seen": 7097860,
      "step": 334,
      "time_per_iteration": 2.718377113342285
    },
    {
      "auxiliary_loss_clip": 0.0131206,
      "auxiliary_loss_mlp": 0.01054693,
      "balance_loss_clip": 1.08303678,
      "balance_loss_mlp": 1.03682327,
      "epoch": 0.040281368364095474,
      "flos": 17821101738240.0,
      "grad_norm": 2.594943846886399,
      "language_loss": 0.89008689,
      "learning_rate": 3.99890407333324e-06,
      "loss": 0.91375446,
      "num_input_tokens_seen": 7116390,
      "step": 335,
      "time_per_iteration": 2.65879225730896
    },
    {
      "auxiliary_loss_clip": 0.01305189,
      "auxiliary_loss_mlp": 0.01053122,
      "balance_loss_clip": 1.07677388,
      "balance_loss_mlp": 1.03568137,
      "epoch": 0.040401611254734564,
      "flos": 19573757959680.0,
      "grad_norm": 1.8567481919899163,
      "language_loss": 0.87158579,
      "learning_rate": 3.998878137563275e-06,
      "loss": 0.8951689,
      "num_input_tokens_seen": 7135940,
      "step": 336,
      "time_per_iteration": 2.7029807567596436
    },
    {
      "auxiliary_loss_clip": 0.01308126,
      "auxiliary_loss_mlp": 0.01062454,
      "balance_loss_clip": 1.08033013,
      "balance_loss_mlp": 1.04584837,
      "epoch": 0.040521854145373654,
      "flos": 22054466949120.0,
      "grad_norm": 5.184915187114566,
      "language_loss": 0.85417187,
      "learning_rate": 3.998851898565085e-06,
      "loss": 0.87787765,
      "num_input_tokens_seen": 7155745,
      "step": 337,
      "time_per_iteration": 2.7057945728302
    },
    {
      "auxiliary_loss_clip": 0.0131052,
      "auxiliary_loss_mlp": 0.01060716,
      "balance_loss_clip": 1.07978249,
      "balance_loss_mlp": 1.04390728,
      "epoch": 0.04064209703601274,
      "flos": 22674644196480.0,
      "grad_norm": 1.9280515583643374,
      "language_loss": 0.83206749,
      "learning_rate": 3.998825356342653e-06,
      "loss": 0.85577989,
      "num_input_tokens_seen": 7175920,
      "step": 338,
      "time_per_iteration": 2.759704828262329
    },
    {
      "auxiliary_loss_clip": 0.01309634,
      "auxiliary_loss_mlp": 0.01046004,
      "balance_loss_clip": 1.07953835,
      "balance_loss_mlp": 1.02940989,
      "epoch": 0.04076233992665183,
      "flos": 38582172783360.0,
      "grad_norm": 2.9791869434929774,
      "language_loss": 0.73254955,
      "learning_rate": 3.998798510900003e-06,
      "loss": 0.7561059,
      "num_input_tokens_seen": 7198720,
      "step": 339,
      "time_per_iteration": 2.8123061656951904
    },
    {
      "auxiliary_loss_clip": 0.01303911,
      "auxiliary_loss_mlp": 0.01054731,
      "balance_loss_clip": 1.07760644,
      "balance_loss_mlp": 1.03761244,
      "epoch": 0.04088258281729093,
      "flos": 25885309374720.0,
      "grad_norm": 2.8020377577927547,
      "language_loss": 0.83838844,
      "learning_rate": 3.998771362241207e-06,
      "loss": 0.86197484,
      "num_input_tokens_seen": 7219125,
      "step": 340,
      "time_per_iteration": 2.677311658859253
    },
    {
      "auxiliary_loss_clip": 0.01305011,
      "auxiliary_loss_mlp": 0.01055533,
      "balance_loss_clip": 1.07914829,
      "balance_loss_mlp": 1.03904581,
      "epoch": 0.04100282570793002,
      "flos": 19789832223360.0,
      "grad_norm": 2.116505356596882,
      "language_loss": 0.8788749,
      "learning_rate": 3.998743910370385e-06,
      "loss": 0.90248036,
      "num_input_tokens_seen": 7237985,
      "step": 341,
      "time_per_iteration": 2.7009551525115967
    },
    {
      "auxiliary_loss_clip": 0.01314027,
      "auxiliary_loss_mlp": 0.01051436,
      "balance_loss_clip": 1.08639824,
      "balance_loss_mlp": 1.03416204,
      "epoch": 0.04112306859856911,
      "flos": 22565152563840.0,
      "grad_norm": 2.3140281496041912,
      "language_loss": 0.73341012,
      "learning_rate": 3.998716155291702e-06,
      "loss": 0.7570647,
      "num_input_tokens_seen": 7255825,
      "step": 342,
      "time_per_iteration": 2.632542133331299
    },
    {
      "auxiliary_loss_clip": 0.01307555,
      "auxiliary_loss_mlp": 0.01043454,
      "balance_loss_clip": 1.08335137,
      "balance_loss_mlp": 1.02697909,
      "epoch": 0.0412433114892082,
      "flos": 25040654081280.0,
      "grad_norm": 1.6953134538468169,
      "language_loss": 0.90617108,
      "learning_rate": 3.998688097009366e-06,
      "loss": 0.92968118,
      "num_input_tokens_seen": 7276590,
      "step": 343,
      "time_per_iteration": 2.8527209758758545
    },
    {
      "auxiliary_loss_clip": 0.01303845,
      "auxiliary_loss_mlp": 0.01052978,
      "balance_loss_clip": 1.07916045,
      "balance_loss_mlp": 1.03610945,
      "epoch": 0.04136355437984729,
      "flos": 25191371548800.0,
      "grad_norm": 3.110155882471581,
      "language_loss": 0.79673529,
      "learning_rate": 3.998659735527636e-06,
      "loss": 0.8203035,
      "num_input_tokens_seen": 7295680,
      "step": 344,
      "time_per_iteration": 2.670529842376709
    },
    {
      "auxiliary_loss_clip": 0.01303376,
      "auxiliary_loss_mlp": 0.01060344,
      "balance_loss_clip": 1.07962394,
      "balance_loss_mlp": 1.04312992,
      "epoch": 0.04148379727048638,
      "flos": 22966777509120.0,
      "grad_norm": 2.0288079630142697,
      "language_loss": 0.77838016,
      "learning_rate": 3.998631070850813e-06,
      "loss": 0.80201733,
      "num_input_tokens_seen": 7316300,
      "step": 345,
      "time_per_iteration": 2.7030649185180664
    },
    {
      "auxiliary_loss_clip": 0.01304251,
      "auxiliary_loss_mlp": 0.01059153,
      "balance_loss_clip": 1.07916498,
      "balance_loss_mlp": 1.04283285,
      "epoch": 0.041604040161125476,
      "flos": 14063481187200.0,
      "grad_norm": 2.7149746707821008,
      "language_loss": 0.8338145,
      "learning_rate": 3.9986021029832455e-06,
      "loss": 0.85744858,
      "num_input_tokens_seen": 7333615,
      "step": 346,
      "time_per_iteration": 2.6110408306121826
    },
    {
      "auxiliary_loss_clip": 0.0130469,
      "auxiliary_loss_mlp": 0.01052665,
      "balance_loss_clip": 1.07904434,
      "balance_loss_mlp": 1.03557062,
      "epoch": 0.041724283051764566,
      "flos": 12091877614080.0,
      "grad_norm": 2.618161004038605,
      "language_loss": 0.91037285,
      "learning_rate": 3.9985728319293285e-06,
      "loss": 0.93394643,
      "num_input_tokens_seen": 7347590,
      "step": 347,
      "time_per_iteration": 2.653554916381836
    },
    {
      "auxiliary_loss_clip": 0.01308529,
      "auxiliary_loss_mlp": 0.01065578,
      "balance_loss_clip": 1.07882595,
      "balance_loss_mlp": 1.0465169,
      "epoch": 0.041844525942403656,
      "flos": 12385303816320.0,
      "grad_norm": 2.5259381614977747,
      "language_loss": 0.85050571,
      "learning_rate": 3.998543257693501e-06,
      "loss": 0.87424684,
      "num_input_tokens_seen": 7364345,
      "step": 348,
      "time_per_iteration": 2.8200273513793945
    },
    {
      "auxiliary_loss_clip": 0.01304348,
      "auxiliary_loss_mlp": 0.01054274,
      "balance_loss_clip": 1.080948,
      "balance_loss_mlp": 1.03720367,
      "epoch": 0.041964768833042745,
      "flos": 23769345041280.0,
      "grad_norm": 1.7711025230908282,
      "language_loss": 0.8803162,
      "learning_rate": 3.998513380280251e-06,
      "loss": 0.90390241,
      "num_input_tokens_seen": 7384625,
      "step": 349,
      "time_per_iteration": 2.7264652252197266
    },
    {
      "auxiliary_loss_clip": 0.01309174,
      "auxiliary_loss_mlp": 0.01068243,
      "balance_loss_clip": 1.08033311,
      "balance_loss_mlp": 1.0509932,
      "epoch": 0.042085011723681835,
      "flos": 11875336473600.0,
      "grad_norm": 3.8200484439665185,
      "language_loss": 0.95077676,
      "learning_rate": 3.99848319969411e-06,
      "loss": 0.9745509,
      "num_input_tokens_seen": 7402225,
      "step": 350,
      "time_per_iteration": 2.9306702613830566
    },
    {
      "auxiliary_loss_clip": 0.01311774,
      "auxiliary_loss_mlp": 0.01067051,
      "balance_loss_clip": 1.08323812,
      "balance_loss_mlp": 1.04934812,
      "epoch": 0.042205254614320925,
      "flos": 16873957964160.0,
      "grad_norm": 2.4216160113589194,
      "language_loss": 0.78877413,
      "learning_rate": 3.9984527159396564e-06,
      "loss": 0.81256235,
      "num_input_tokens_seen": 7420865,
      "step": 351,
      "time_per_iteration": 2.8476171493530273
    },
    {
      "auxiliary_loss_clip": 0.01302136,
      "auxiliary_loss_mlp": 0.01054442,
      "balance_loss_clip": 1.07748055,
      "balance_loss_mlp": 1.03752637,
      "epoch": 0.04232549750496002,
      "flos": 25118508810240.0,
      "grad_norm": 2.1560167146306735,
      "language_loss": 0.83963549,
      "learning_rate": 3.9984219290215154e-06,
      "loss": 0.86320132,
      "num_input_tokens_seen": 7441040,
      "step": 352,
      "time_per_iteration": 2.7566349506378174
    },
    {
      "auxiliary_loss_clip": 0.01303534,
      "auxiliary_loss_mlp": 0.01059427,
      "balance_loss_clip": 1.08061063,
      "balance_loss_mlp": 1.04301178,
      "epoch": 0.04244574039559911,
      "flos": 26724541714560.0,
      "grad_norm": 3.1966676570032213,
      "language_loss": 0.89213383,
      "learning_rate": 3.998390838944356e-06,
      "loss": 0.91576338,
      "num_input_tokens_seen": 7462545,
      "step": 353,
      "time_per_iteration": 2.7652621269226074
    },
    {
      "auxiliary_loss_clip": 0.01306941,
      "auxiliary_loss_mlp": 0.01053478,
      "balance_loss_clip": 1.07934618,
      "balance_loss_mlp": 1.0368011,
      "epoch": 0.0425659832862382,
      "flos": 20923244951040.0,
      "grad_norm": 4.429405146484284,
      "language_loss": 0.90408921,
      "learning_rate": 3.998359445712895e-06,
      "loss": 0.92769343,
      "num_input_tokens_seen": 7481650,
      "step": 354,
      "time_per_iteration": 2.69223690032959
    },
    {
      "auxiliary_loss_clip": 0.01303635,
      "auxiliary_loss_mlp": 0.01054416,
      "balance_loss_clip": 1.07802653,
      "balance_loss_mlp": 1.03771496,
      "epoch": 0.04268622617687729,
      "flos": 23331127115520.0,
      "grad_norm": 3.0986551063430645,
      "language_loss": 0.81134617,
      "learning_rate": 3.9983277493318955e-06,
      "loss": 0.83492661,
      "num_input_tokens_seen": 7500945,
      "step": 355,
      "time_per_iteration": 2.6368491649627686
    },
    {
      "auxiliary_loss_clip": 0.01302412,
      "auxiliary_loss_mlp": 0.0105426,
      "balance_loss_clip": 1.07770896,
      "balance_loss_mlp": 1.03778505,
      "epoch": 0.04280646906751638,
      "flos": 25994010908160.0,
      "grad_norm": 1.653004541875781,
      "language_loss": 0.81128716,
      "learning_rate": 3.998295749806165e-06,
      "loss": 0.83485389,
      "num_input_tokens_seen": 7522170,
      "step": 356,
      "time_per_iteration": 3.516709327697754
    },
    {
      "auxiliary_loss_clip": 0.01305651,
      "auxiliary_loss_mlp": 0.01054769,
      "balance_loss_clip": 1.07950175,
      "balance_loss_mlp": 1.03835404,
      "epoch": 0.04292671195815547,
      "flos": 26906824258560.0,
      "grad_norm": 2.0964476951655455,
      "language_loss": 0.83325434,
      "learning_rate": 3.998263447140558e-06,
      "loss": 0.85685855,
      "num_input_tokens_seen": 7542370,
      "step": 357,
      "time_per_iteration": 4.495465517044067
    },
    {
      "auxiliary_loss_clip": 0.01302657,
      "auxiliary_loss_mlp": 0.01052245,
      "balance_loss_clip": 1.07814622,
      "balance_loss_mlp": 1.03503132,
      "epoch": 0.04304695484879457,
      "flos": 39457315745280.0,
      "grad_norm": 1.9791897329572008,
      "language_loss": 0.81494331,
      "learning_rate": 3.998230841339976e-06,
      "loss": 0.83849233,
      "num_input_tokens_seen": 7564380,
      "step": 358,
      "time_per_iteration": 3.7497401237487793
    },
    {
      "auxiliary_loss_clip": 0.01304031,
      "auxiliary_loss_mlp": 0.01061637,
      "balance_loss_clip": 1.08069944,
      "balance_loss_mlp": 1.0454129,
      "epoch": 0.04316719773943366,
      "flos": 19646297475840.0,
      "grad_norm": 2.9926578627958027,
      "language_loss": 0.84986901,
      "learning_rate": 3.998197932409363e-06,
      "loss": 0.8735258,
      "num_input_tokens_seen": 7582390,
      "step": 359,
      "time_per_iteration": 2.6211178302764893
    },
    {
      "auxiliary_loss_clip": 0.01299347,
      "auxiliary_loss_mlp": 0.01052802,
      "balance_loss_clip": 1.07755017,
      "balance_loss_mlp": 1.03618407,
      "epoch": 0.04328744063007275,
      "flos": 22452320966400.0,
      "grad_norm": 2.1391725520609186,
      "language_loss": 0.86312336,
      "learning_rate": 3.9981647203537125e-06,
      "loss": 0.88664484,
      "num_input_tokens_seen": 7599890,
      "step": 360,
      "time_per_iteration": 2.684583902359009
    },
    {
      "auxiliary_loss_clip": 0.01301256,
      "auxiliary_loss_mlp": 0.01052676,
      "balance_loss_clip": 1.07748687,
      "balance_loss_mlp": 1.03559351,
      "epoch": 0.04340768352071184,
      "flos": 21283033530240.0,
      "grad_norm": 2.2748306331486976,
      "language_loss": 0.95937276,
      "learning_rate": 3.998131205178063e-06,
      "loss": 0.98291212,
      "num_input_tokens_seen": 7618360,
      "step": 361,
      "time_per_iteration": 2.64471173286438
    },
    {
      "auxiliary_loss_clip": 0.013026,
      "auxiliary_loss_mlp": 0.01056073,
      "balance_loss_clip": 1.0787288,
      "balance_loss_mlp": 1.03811979,
      "epoch": 0.04352792641135093,
      "flos": 11583705951360.0,
      "grad_norm": 11.735900798606263,
      "language_loss": 0.76661503,
      "learning_rate": 3.998097386887498e-06,
      "loss": 0.79020178,
      "num_input_tokens_seen": 7635435,
      "step": 362,
      "time_per_iteration": 2.6437597274780273
    },
    {
      "auxiliary_loss_clip": 0.01296467,
      "auxiliary_loss_mlp": 0.010562,
      "balance_loss_clip": 1.07645893,
      "balance_loss_mlp": 1.04059577,
      "epoch": 0.04364816930199002,
      "flos": 23623547736960.0,
      "grad_norm": 1.8472651974208198,
      "language_loss": 0.84921682,
      "learning_rate": 3.998063265487148e-06,
      "loss": 0.87274349,
      "num_input_tokens_seen": 7656485,
      "step": 363,
      "time_per_iteration": 2.654468536376953
    },
    {
      "auxiliary_loss_clip": 0.01300164,
      "auxiliary_loss_mlp": 0.01064081,
      "balance_loss_clip": 1.07740927,
      "balance_loss_mlp": 1.04808283,
      "epoch": 0.043768412192629114,
      "flos": 14429734214400.0,
      "grad_norm": 4.112434094423012,
      "language_loss": 0.81128711,
      "learning_rate": 3.99802884098219e-06,
      "loss": 0.83492959,
      "num_input_tokens_seen": 7674595,
      "step": 364,
      "time_per_iteration": 2.6554665565490723
    },
    {
      "auxiliary_loss_clip": 0.01298253,
      "auxiliary_loss_mlp": 0.01046731,
      "balance_loss_clip": 1.07515359,
      "balance_loss_mlp": 1.0303036,
      "epoch": 0.043888655083268203,
      "flos": 26468893641600.0,
      "grad_norm": 2.478277187301562,
      "language_loss": 0.82387006,
      "learning_rate": 3.997994113377845e-06,
      "loss": 0.8473199,
      "num_input_tokens_seen": 7693495,
      "step": 365,
      "time_per_iteration": 2.6950199604034424
    },
    {
      "auxiliary_loss_clip": 0.01302148,
      "auxiliary_loss_mlp": 0.01063067,
      "balance_loss_clip": 1.0791018,
      "balance_loss_mlp": 1.04640126,
      "epoch": 0.04400889797390729,
      "flos": 27235263242880.0,
      "grad_norm": 2.4618489894397824,
      "language_loss": 0.83243906,
      "learning_rate": 3.9979590826793815e-06,
      "loss": 0.8560912,
      "num_input_tokens_seen": 7714685,
      "step": 366,
      "time_per_iteration": 2.705496072769165
    },
    {
      "auxiliary_loss_clip": 0.01303664,
      "auxiliary_loss_mlp": 0.01054857,
      "balance_loss_clip": 1.08077598,
      "balance_loss_mlp": 1.03745198,
      "epoch": 0.04412914086454638,
      "flos": 20119528183680.0,
      "grad_norm": 2.1991399256036974,
      "language_loss": 0.81119907,
      "learning_rate": 3.997923748892113e-06,
      "loss": 0.83478421,
      "num_input_tokens_seen": 7734005,
      "step": 367,
      "time_per_iteration": 2.6525352001190186
    },
    {
      "auxiliary_loss_clip": 0.01301725,
      "auxiliary_loss_mlp": 0.01059197,
      "balance_loss_clip": 1.08004928,
      "balance_loss_mlp": 1.04267454,
      "epoch": 0.04424938375518547,
      "flos": 22604618632320.0,
      "grad_norm": 1.7871520008729351,
      "language_loss": 0.88514447,
      "learning_rate": 3.9978881120214015e-06,
      "loss": 0.90875375,
      "num_input_tokens_seen": 7755525,
      "step": 368,
      "time_per_iteration": 2.7256197929382324
    },
    {
      "auxiliary_loss_clip": 0.0130007,
      "auxiliary_loss_mlp": 0.01046724,
      "balance_loss_clip": 1.07588542,
      "balance_loss_mlp": 1.02989185,
      "epoch": 0.04436962664582456,
      "flos": 24132365844480.0,
      "grad_norm": 1.849002353469361,
      "language_loss": 0.79390275,
      "learning_rate": 3.997852172072652e-06,
      "loss": 0.81737071,
      "num_input_tokens_seen": 7776740,
      "step": 369,
      "time_per_iteration": 2.676048517227173
    },
    {
      "auxiliary_loss_clip": 0.01301548,
      "auxiliary_loss_mlp": 0.0106793,
      "balance_loss_clip": 1.07754004,
      "balance_loss_mlp": 1.04942918,
      "epoch": 0.04448986953646366,
      "flos": 18222906251520.0,
      "grad_norm": 3.0680065399828855,
      "language_loss": 0.89492077,
      "learning_rate": 3.9978159290513155e-06,
      "loss": 0.91861558,
      "num_input_tokens_seen": 7794820,
      "step": 370,
      "time_per_iteration": 2.58294677734375
    },
    {
      "auxiliary_loss_clip": 0.01307686,
      "auxiliary_loss_mlp": 0.01062697,
      "balance_loss_clip": 1.08165538,
      "balance_loss_mlp": 1.0453763,
      "epoch": 0.04461011242710275,
      "flos": 30117920400000.0,
      "grad_norm": 2.8603399139884647,
      "language_loss": 0.80434948,
      "learning_rate": 3.997779382962892e-06,
      "loss": 0.82805336,
      "num_input_tokens_seen": 7817705,
      "step": 371,
      "time_per_iteration": 2.731043815612793
    },
    {
      "auxiliary_loss_clip": 0.01298654,
      "auxiliary_loss_mlp": 0.01051143,
      "balance_loss_clip": 1.07828224,
      "balance_loss_mlp": 1.0345608,
      "epoch": 0.04473035531774184,
      "flos": 29752529299200.0,
      "grad_norm": 2.3908545636859815,
      "language_loss": 0.74133778,
      "learning_rate": 3.997742533812924e-06,
      "loss": 0.76483572,
      "num_input_tokens_seen": 7840970,
      "step": 372,
      "time_per_iteration": 2.782999038696289
    },
    {
      "auxiliary_loss_clip": 0.01299294,
      "auxiliary_loss_mlp": 0.01068106,
      "balance_loss_clip": 1.0776968,
      "balance_loss_mlp": 1.05089164,
      "epoch": 0.04485059820838093,
      "flos": 13151565676800.0,
      "grad_norm": 2.7520260148809164,
      "language_loss": 0.92503977,
      "learning_rate": 3.997705381607001e-06,
      "loss": 0.94871378,
      "num_input_tokens_seen": 7857785,
      "step": 373,
      "time_per_iteration": 2.634202003479004
    },
    {
      "auxiliary_loss_clip": 0.01229592,
      "auxiliary_loss_mlp": 0.01031015,
      "balance_loss_clip": 1.06863999,
      "balance_loss_mlp": 1.02171695,
      "epoch": 0.04497084109902002,
      "flos": 68094209548800.0,
      "grad_norm": 5.298250616423988,
      "language_loss": 0.60237736,
      "learning_rate": 3.997667926350761e-06,
      "loss": 0.62498343,
      "num_input_tokens_seen": 7916115,
      "step": 374,
      "time_per_iteration": 3.125986337661743
    },
    {
      "auxiliary_loss_clip": 0.01229545,
      "auxiliary_loss_mlp": 0.01026861,
      "balance_loss_clip": 1.06812024,
      "balance_loss_mlp": 1.01751459,
      "epoch": 0.04509108398965911,
      "flos": 64342263346560.0,
      "grad_norm": 0.9115823330686188,
      "language_loss": 0.5778172,
      "learning_rate": 3.997630168049886e-06,
      "loss": 0.60038126,
      "num_input_tokens_seen": 7974480,
      "step": 375,
      "time_per_iteration": 3.200150966644287
    },
    {
      "auxiliary_loss_clip": 0.01302189,
      "auxiliary_loss_mlp": 0.01052453,
      "balance_loss_clip": 1.07994246,
      "balance_loss_mlp": 1.03531051,
      "epoch": 0.045211326880298205,
      "flos": 22271115830400.0,
      "grad_norm": 1.9778693985668847,
      "language_loss": 0.77330089,
      "learning_rate": 3.997592106710101e-06,
      "loss": 0.79684734,
      "num_input_tokens_seen": 7993940,
      "step": 376,
      "time_per_iteration": 2.6484200954437256
    },
    {
      "auxiliary_loss_clip": 0.01297064,
      "auxiliary_loss_mlp": 0.0105497,
      "balance_loss_clip": 1.07606614,
      "balance_loss_mlp": 1.03855538,
      "epoch": 0.045331569770937295,
      "flos": 32159441796480.0,
      "grad_norm": 5.435332119991449,
      "language_loss": 0.65523994,
      "learning_rate": 3.997553742337182e-06,
      "loss": 0.67876029,
      "num_input_tokens_seen": 8013365,
      "step": 377,
      "time_per_iteration": 2.726078510284424
    },
    {
      "auxiliary_loss_clip": 0.01300217,
      "auxiliary_loss_mlp": 0.01064226,
      "balance_loss_clip": 1.07984042,
      "balance_loss_mlp": 1.04829991,
      "epoch": 0.045451812661576385,
      "flos": 22163455791360.0,
      "grad_norm": 1.9156859903029109,
      "language_loss": 0.91562378,
      "learning_rate": 3.997515074936949e-06,
      "loss": 0.93926823,
      "num_input_tokens_seen": 8034240,
      "step": 378,
      "time_per_iteration": 2.6822803020477295
    },
    {
      "auxiliary_loss_clip": 0.01297574,
      "auxiliary_loss_mlp": 0.01051214,
      "balance_loss_clip": 1.07721746,
      "balance_loss_mlp": 1.0342741,
      "epoch": 0.045572055552215475,
      "flos": 16581968305920.0,
      "grad_norm": 2.6642076871791205,
      "language_loss": 0.87371695,
      "learning_rate": 3.997476104515268e-06,
      "loss": 0.89720482,
      "num_input_tokens_seen": 8052430,
      "step": 379,
      "time_per_iteration": 2.6224355697631836
    },
    {
      "auxiliary_loss_clip": 0.01293056,
      "auxiliary_loss_mlp": 0.01052869,
      "balance_loss_clip": 1.07755327,
      "balance_loss_mlp": 1.03671646,
      "epoch": 0.045692298442854565,
      "flos": 17603375448960.0,
      "grad_norm": 2.3321547462077077,
      "language_loss": 0.77836657,
      "learning_rate": 3.9974368310780485e-06,
      "loss": 0.80182588,
      "num_input_tokens_seen": 8069605,
      "step": 380,
      "time_per_iteration": 2.659353494644165
    },
    {
      "auxiliary_loss_clip": 0.01304064,
      "auxiliary_loss_mlp": 0.01061384,
      "balance_loss_clip": 1.08022654,
      "balance_loss_mlp": 1.04424191,
      "epoch": 0.045812541333493655,
      "flos": 26761098781440.0,
      "grad_norm": 2.780408237198681,
      "language_loss": 0.74347103,
      "learning_rate": 3.997397254631251e-06,
      "loss": 0.76712549,
      "num_input_tokens_seen": 8090225,
      "step": 381,
      "time_per_iteration": 2.678143262863159
    },
    {
      "auxiliary_loss_clip": 0.01218911,
      "auxiliary_loss_mlp": 0.01012946,
      "balance_loss_clip": 1.06116951,
      "balance_loss_mlp": 1.00350475,
      "epoch": 0.04593278422413275,
      "flos": 60250349894400.0,
      "grad_norm": 0.8081417123630017,
      "language_loss": 0.60029817,
      "learning_rate": 3.997357375180878e-06,
      "loss": 0.62261677,
      "num_input_tokens_seen": 8154505,
      "step": 382,
      "time_per_iteration": 4.130557060241699
    },
    {
      "auxiliary_loss_clip": 0.01295451,
      "auxiliary_loss_mlp": 0.01054881,
      "balance_loss_clip": 1.07608986,
      "balance_loss_mlp": 1.03870416,
      "epoch": 0.04605302711477184,
      "flos": 21799249839360.0,
      "grad_norm": 1.7384428028315952,
      "language_loss": 0.75364017,
      "learning_rate": 3.997317192732979e-06,
      "loss": 0.77714348,
      "num_input_tokens_seen": 8173285,
      "step": 383,
      "time_per_iteration": 3.5492475032806396
    },
    {
      "auxiliary_loss_clip": 0.01298251,
      "auxiliary_loss_mlp": 0.01081114,
      "balance_loss_clip": 1.07620525,
      "balance_loss_mlp": 1.06467474,
      "epoch": 0.04617327000541093,
      "flos": 19459705299840.0,
      "grad_norm": 1.862336881741504,
      "language_loss": 0.82686806,
      "learning_rate": 3.99727670729365e-06,
      "loss": 0.85066175,
      "num_input_tokens_seen": 8191845,
      "step": 384,
      "time_per_iteration": 3.527864694595337
    },
    {
      "auxiliary_loss_clip": 0.01297728,
      "auxiliary_loss_mlp": 0.01052551,
      "balance_loss_clip": 1.07980442,
      "balance_loss_mlp": 1.03630292,
      "epoch": 0.04629351289605002,
      "flos": 25411468135680.0,
      "grad_norm": 1.7169656763152812,
      "language_loss": 0.77886957,
      "learning_rate": 3.997235918869033e-06,
      "loss": 0.80237234,
      "num_input_tokens_seen": 8212880,
      "step": 385,
      "time_per_iteration": 3.5657477378845215
    },
    {
      "auxiliary_loss_clip": 0.01294538,
      "auxiliary_loss_mlp": 0.01048789,
      "balance_loss_clip": 1.07687211,
      "balance_loss_mlp": 1.03304136,
      "epoch": 0.04641375578668911,
      "flos": 20558284813440.0,
      "grad_norm": 2.0990250560348804,
      "language_loss": 0.82422608,
      "learning_rate": 3.997194827465315e-06,
      "loss": 0.84765935,
      "num_input_tokens_seen": 8231475,
      "step": 386,
      "time_per_iteration": 2.694240093231201
    },
    {
      "auxiliary_loss_clip": 0.01293776,
      "auxiliary_loss_mlp": 0.01048039,
      "balance_loss_clip": 1.07469344,
      "balance_loss_mlp": 1.03165984,
      "epoch": 0.0465339986773282,
      "flos": 13188661447680.0,
      "grad_norm": 3.668457870887252,
      "language_loss": 0.9140296,
      "learning_rate": 3.997153433088728e-06,
      "loss": 0.93744779,
      "num_input_tokens_seen": 8248600,
      "step": 387,
      "time_per_iteration": 2.614888906478882
    },
    {
      "auxiliary_loss_clip": 0.01294541,
      "auxiliary_loss_mlp": 0.01048702,
      "balance_loss_clip": 1.07612348,
      "balance_loss_mlp": 1.03195333,
      "epoch": 0.0466542415679673,
      "flos": 25556547168000.0,
      "grad_norm": 3.095194407786512,
      "language_loss": 0.81485796,
      "learning_rate": 3.997111735745554e-06,
      "loss": 0.83829039,
      "num_input_tokens_seen": 8271570,
      "step": 388,
      "time_per_iteration": 2.7356226444244385
    },
    {
      "auxiliary_loss_clip": 0.01293387,
      "auxiliary_loss_mlp": 0.01058006,
      "balance_loss_clip": 1.0762105,
      "balance_loss_mlp": 1.04109049,
      "epoch": 0.04677448445860639,
      "flos": 22236749493120.0,
      "grad_norm": 1.9160458408406096,
      "language_loss": 0.82782376,
      "learning_rate": 3.997069735442118e-06,
      "loss": 0.85133767,
      "num_input_tokens_seen": 8291265,
      "step": 389,
      "time_per_iteration": 2.6425600051879883
    },
    {
      "auxiliary_loss_clip": 0.01289865,
      "auxiliary_loss_mlp": 0.01049579,
      "balance_loss_clip": 1.07428908,
      "balance_loss_mlp": 1.03452289,
      "epoch": 0.04689472734924548,
      "flos": 28147825198080.0,
      "grad_norm": 1.8054039265887218,
      "language_loss": 0.80316806,
      "learning_rate": 3.997027432184792e-06,
      "loss": 0.82656252,
      "num_input_tokens_seen": 8315925,
      "step": 390,
      "time_per_iteration": 2.7311086654663086
    },
    {
      "auxiliary_loss_clip": 0.01292172,
      "auxiliary_loss_mlp": 0.01043596,
      "balance_loss_clip": 1.07509053,
      "balance_loss_mlp": 1.02720487,
      "epoch": 0.04701497023988457,
      "flos": 23148952312320.0,
      "grad_norm": 1.9855015041288748,
      "language_loss": 0.89557445,
      "learning_rate": 3.99698482597999e-06,
      "loss": 0.91893214,
      "num_input_tokens_seen": 8333605,
      "step": 391,
      "time_per_iteration": 2.6503472328186035
    },
    {
      "auxiliary_loss_clip": 0.01209534,
      "auxiliary_loss_mlp": 0.01008373,
      "balance_loss_clip": 1.05423355,
      "balance_loss_mlp": 0.99917048,
      "epoch": 0.04713521313052366,
      "flos": 64827668764800.0,
      "grad_norm": 0.8819561541659392,
      "language_loss": 0.63975763,
      "learning_rate": 3.99694191683418e-06,
      "loss": 0.66193664,
      "num_input_tokens_seen": 8394405,
      "step": 392,
      "time_per_iteration": 3.2271018028259277
    },
    {
      "auxiliary_loss_clip": 0.01300491,
      "auxiliary_loss_mlp": 0.01049527,
      "balance_loss_clip": 1.08060265,
      "balance_loss_mlp": 1.0334574,
      "epoch": 0.047255456021162746,
      "flos": 18771585477120.0,
      "grad_norm": 2.1087823950029008,
      "language_loss": 0.81835318,
      "learning_rate": 3.996898704753867e-06,
      "loss": 0.84185338,
      "num_input_tokens_seen": 8412355,
      "step": 393,
      "time_per_iteration": 2.7225568294525146
    },
    {
      "auxiliary_loss_clip": 0.01293015,
      "auxiliary_loss_mlp": 0.01060726,
      "balance_loss_clip": 1.07374549,
      "balance_loss_mlp": 1.04400134,
      "epoch": 0.04737569891180184,
      "flos": 22053820504320.0,
      "grad_norm": 2.4285516811479195,
      "language_loss": 0.87830842,
      "learning_rate": 3.996855189745609e-06,
      "loss": 0.90184581,
      "num_input_tokens_seen": 8431620,
      "step": 394,
      "time_per_iteration": 2.6995530128479004
    },
    {
      "auxiliary_loss_clip": 0.01291689,
      "auxiliary_loss_mlp": 0.01046846,
      "balance_loss_clip": 1.07387996,
      "balance_loss_mlp": 1.03215957,
      "epoch": 0.04749594180244093,
      "flos": 29057370410880.0,
      "grad_norm": 2.2024421180345954,
      "language_loss": 0.92685658,
      "learning_rate": 3.996811371816007e-06,
      "loss": 0.95024192,
      "num_input_tokens_seen": 8454045,
      "step": 395,
      "time_per_iteration": 2.738726854324341
    },
    {
      "auxiliary_loss_clip": 0.01293593,
      "auxiliary_loss_mlp": 0.01050369,
      "balance_loss_clip": 1.07729363,
      "balance_loss_mlp": 1.0348475,
      "epoch": 0.04761618469308002,
      "flos": 35112268172160.0,
      "grad_norm": 1.9898024634541902,
      "language_loss": 0.77986836,
      "learning_rate": 3.996767250971707e-06,
      "loss": 0.80330795,
      "num_input_tokens_seen": 8476785,
      "step": 396,
      "time_per_iteration": 2.787046194076538
    },
    {
      "auxiliary_loss_clip": 0.01293384,
      "auxiliary_loss_mlp": 0.01056578,
      "balance_loss_clip": 1.07635403,
      "balance_loss_mlp": 1.03984058,
      "epoch": 0.04773642758371911,
      "flos": 25630702796160.0,
      "grad_norm": 1.772569646175126,
      "language_loss": 0.86864698,
      "learning_rate": 3.996722827219403e-06,
      "loss": 0.89214659,
      "num_input_tokens_seen": 8498400,
      "step": 397,
      "time_per_iteration": 2.6741998195648193
    },
    {
      "auxiliary_loss_clip": 0.01292788,
      "auxiliary_loss_mlp": 0.01053077,
      "balance_loss_clip": 1.07613659,
      "balance_loss_mlp": 1.03673363,
      "epoch": 0.0478566704743582,
      "flos": 20631506688000.0,
      "grad_norm": 2.3970860501203863,
      "language_loss": 0.82831717,
      "learning_rate": 3.996678100565833e-06,
      "loss": 0.85177588,
      "num_input_tokens_seen": 8517455,
      "step": 398,
      "time_per_iteration": 2.639420509338379
    },
    {
      "auxiliary_loss_clip": 0.01289822,
      "auxiliary_loss_mlp": 0.01064366,
      "balance_loss_clip": 1.07460463,
      "balance_loss_mlp": 1.04774797,
      "epoch": 0.04797691336499729,
      "flos": 18835721210880.0,
      "grad_norm": 2.1516839352813135,
      "language_loss": 0.88451064,
      "learning_rate": 3.996633071017783e-06,
      "loss": 0.90805256,
      "num_input_tokens_seen": 8534085,
      "step": 399,
      "time_per_iteration": 2.664177656173706
    },
    {
      "auxiliary_loss_clip": 0.01290936,
      "auxiliary_loss_mlp": 0.01052556,
      "balance_loss_clip": 1.07636344,
      "balance_loss_mlp": 1.03689194,
      "epoch": 0.04809715625563638,
      "flos": 21099673578240.0,
      "grad_norm": 2.669274155804717,
      "language_loss": 0.82088906,
      "learning_rate": 3.996587738582084e-06,
      "loss": 0.84432399,
      "num_input_tokens_seen": 8550885,
      "step": 400,
      "time_per_iteration": 2.641785144805908
    },
    {
      "auxiliary_loss_clip": 0.01293293,
      "auxiliary_loss_mlp": 0.01048966,
      "balance_loss_clip": 1.07610917,
      "balance_loss_mlp": 1.03380287,
      "epoch": 0.04821739914627548,
      "flos": 23805650712960.0,
      "grad_norm": 2.9053764207972157,
      "language_loss": 0.86313266,
      "learning_rate": 3.9965421032656115e-06,
      "loss": 0.88655519,
      "num_input_tokens_seen": 8570815,
      "step": 401,
      "time_per_iteration": 2.7073943614959717
    },
    {
      "auxiliary_loss_clip": 0.0129221,
      "auxiliary_loss_mlp": 0.01051412,
      "balance_loss_clip": 1.07578194,
      "balance_loss_mlp": 1.03575957,
      "epoch": 0.04833764203691457,
      "flos": 22200587475840.0,
      "grad_norm": 3.0223910462283547,
      "language_loss": 0.94327593,
      "learning_rate": 3.99649616507529e-06,
      "loss": 0.96671212,
      "num_input_tokens_seen": 8589910,
      "step": 402,
      "time_per_iteration": 2.639754056930542
    },
    {
      "auxiliary_loss_clip": 0.01200891,
      "auxiliary_loss_mlp": 0.01010472,
      "balance_loss_clip": 1.04910183,
      "balance_loss_mlp": 1.00165009,
      "epoch": 0.04845788492755366,
      "flos": 65904376896000.0,
      "grad_norm": 0.8911483503930873,
      "language_loss": 0.63280851,
      "learning_rate": 3.996449924018088e-06,
      "loss": 0.65492213,
      "num_input_tokens_seen": 8650370,
      "step": 403,
      "time_per_iteration": 3.1136527061462402
    },
    {
      "auxiliary_loss_clip": 0.0128811,
      "auxiliary_loss_mlp": 0.01048019,
      "balance_loss_clip": 1.07227468,
      "balance_loss_mlp": 1.03276062,
      "epoch": 0.04857812781819275,
      "flos": 19281301424640.0,
      "grad_norm": 2.1235192192881294,
      "language_loss": 0.79580522,
      "learning_rate": 3.99640338010102e-06,
      "loss": 0.81916654,
      "num_input_tokens_seen": 8669475,
      "step": 404,
      "time_per_iteration": 2.7265443801879883
    },
    {
      "auxiliary_loss_clip": 0.0128719,
      "auxiliary_loss_mlp": 0.01049882,
      "balance_loss_clip": 1.07264519,
      "balance_loss_mlp": 1.03495753,
      "epoch": 0.04869837070883184,
      "flos": 24062376193920.0,
      "grad_norm": 1.9531659266552335,
      "language_loss": 0.78571874,
      "learning_rate": 3.996356533331146e-06,
      "loss": 0.80908942,
      "num_input_tokens_seen": 8691345,
      "step": 405,
      "time_per_iteration": 2.651461362838745
    },
    {
      "auxiliary_loss_clip": 0.01293542,
      "auxiliary_loss_mlp": 0.01048509,
      "balance_loss_clip": 1.0729543,
      "balance_loss_mlp": 1.03105664,
      "epoch": 0.04881861359947093,
      "flos": 25187169657600.0,
      "grad_norm": 2.888319179534624,
      "language_loss": 0.61595905,
      "learning_rate": 3.996309383715573e-06,
      "loss": 0.6393795,
      "num_input_tokens_seen": 8710125,
      "step": 406,
      "time_per_iteration": 2.6398096084594727
    },
    {
      "auxiliary_loss_clip": 0.01292289,
      "auxiliary_loss_mlp": 0.01054171,
      "balance_loss_clip": 1.07597327,
      "balance_loss_mlp": 1.03665948,
      "epoch": 0.048938856490110025,
      "flos": 16362913213440.0,
      "grad_norm": 1.971694364650014,
      "language_loss": 0.7383033,
      "learning_rate": 3.996261931261454e-06,
      "loss": 0.76176786,
      "num_input_tokens_seen": 8728705,
      "step": 407,
      "time_per_iteration": 2.645510196685791
    },
    {
      "auxiliary_loss_clip": 0.01295603,
      "auxiliary_loss_mlp": 0.01062239,
      "balance_loss_clip": 1.0782758,
      "balance_loss_mlp": 1.04571676,
      "epoch": 0.049059099380749115,
      "flos": 29895094379520.0,
      "grad_norm": 1.9163212213885072,
      "language_loss": 0.86655384,
      "learning_rate": 3.996214175975987e-06,
      "loss": 0.89013231,
      "num_input_tokens_seen": 8749225,
      "step": 408,
      "time_per_iteration": 3.588085174560547
    },
    {
      "auxiliary_loss_clip": 0.0129555,
      "auxiliary_loss_mlp": 0.01049339,
      "balance_loss_clip": 1.07680726,
      "balance_loss_mlp": 1.03232801,
      "epoch": 0.049179342271388204,
      "flos": 35918858027520.0,
      "grad_norm": 2.2743332457168464,
      "language_loss": 0.79276836,
      "learning_rate": 3.996166117866417e-06,
      "loss": 0.81621718,
      "num_input_tokens_seen": 8771160,
      "step": 409,
      "time_per_iteration": 2.7322707176208496
    },
    {
      "auxiliary_loss_clip": 0.01286842,
      "auxiliary_loss_mlp": 0.01048021,
      "balance_loss_clip": 1.07119548,
      "balance_loss_mlp": 1.03320312,
      "epoch": 0.049299585162027294,
      "flos": 14611226659200.0,
      "grad_norm": 2.174391695318995,
      "language_loss": 0.86905503,
      "learning_rate": 3.996117756940035e-06,
      "loss": 0.89240366,
      "num_input_tokens_seen": 8787845,
      "step": 410,
      "time_per_iteration": 2.577420234680176
    },
    {
      "auxiliary_loss_clip": 0.01287664,
      "auxiliary_loss_mlp": 0.01053522,
      "balance_loss_clip": 1.07339787,
      "balance_loss_mlp": 1.03821516,
      "epoch": 0.049419828052666384,
      "flos": 19567939956480.0,
      "grad_norm": 2.2102803330424576,
      "language_loss": 0.97707361,
      "learning_rate": 3.996069093204175e-06,
      "loss": 1.00048542,
      "num_input_tokens_seen": 8803805,
      "step": 411,
      "time_per_iteration": 4.427610397338867
    },
    {
      "auxiliary_loss_clip": 0.01294352,
      "auxiliary_loss_mlp": 0.01046866,
      "balance_loss_clip": 1.07771516,
      "balance_loss_mlp": 1.03244209,
      "epoch": 0.049540070943305474,
      "flos": 13659916907520.0,
      "grad_norm": 2.352711557636332,
      "language_loss": 0.88274282,
      "learning_rate": 3.996020126666221e-06,
      "loss": 0.90615499,
      "num_input_tokens_seen": 8820785,
      "step": 412,
      "time_per_iteration": 2.5822505950927734
    },
    {
      "auxiliary_loss_clip": 0.01292134,
      "auxiliary_loss_mlp": 0.0105314,
      "balance_loss_clip": 1.07694256,
      "balance_loss_mlp": 1.03809571,
      "epoch": 0.04966031383394457,
      "flos": 21832035978240.0,
      "grad_norm": 2.0381910350327015,
      "language_loss": 0.82182908,
      "learning_rate": 3.995970857333601e-06,
      "loss": 0.84528184,
      "num_input_tokens_seen": 8841195,
      "step": 413,
      "time_per_iteration": 2.7868926525115967
    },
    {
      "auxiliary_loss_clip": 0.01289397,
      "auxiliary_loss_mlp": 0.01055598,
      "balance_loss_clip": 1.07488227,
      "balance_loss_mlp": 1.03911114,
      "epoch": 0.04978055672458366,
      "flos": 28618793349120.0,
      "grad_norm": 1.7648625853458038,
      "language_loss": 0.79743814,
      "learning_rate": 3.995921285213789e-06,
      "loss": 0.8208881,
      "num_input_tokens_seen": 8861455,
      "step": 414,
      "time_per_iteration": 2.665783643722534
    },
    {
      "auxiliary_loss_clip": 0.01288925,
      "auxiliary_loss_mlp": 0.01044105,
      "balance_loss_clip": 1.07373428,
      "balance_loss_mlp": 1.02864313,
      "epoch": 0.04990079961522275,
      "flos": 19828220883840.0,
      "grad_norm": 2.301923957699768,
      "language_loss": 0.80692011,
      "learning_rate": 3.995871410314305e-06,
      "loss": 0.83025038,
      "num_input_tokens_seen": 8880015,
      "step": 415,
      "time_per_iteration": 2.768411159515381
    },
    {
      "auxiliary_loss_clip": 0.01184486,
      "auxiliary_loss_mlp": 0.01009151,
      "balance_loss_clip": 1.04825163,
      "balance_loss_mlp": 1.00047255,
      "epoch": 0.05002104250586184,
      "flos": 62735045293440.0,
      "grad_norm": 0.9030998695797329,
      "language_loss": 0.59643245,
      "learning_rate": 3.995821232642714e-06,
      "loss": 0.6183688,
      "num_input_tokens_seen": 8938420,
      "step": 416,
      "time_per_iteration": 3.3072218894958496
    },
    {
      "auxiliary_loss_clip": 0.01269257,
      "auxiliary_loss_mlp": 0.01052915,
      "balance_loss_clip": 1.07436073,
      "balance_loss_mlp": 1.03677452,
      "epoch": 0.05014128539650093,
      "flos": 27928518710400.0,
      "grad_norm": 5.421790085037207,
      "language_loss": 0.8243736,
      "learning_rate": 3.995770752206629e-06,
      "loss": 0.84759533,
      "num_input_tokens_seen": 8959495,
      "step": 417,
      "time_per_iteration": 2.7052743434906006
    },
    {
      "auxiliary_loss_clip": 0.01291332,
      "auxiliary_loss_mlp": 0.0104693,
      "balance_loss_clip": 1.07639861,
      "balance_loss_mlp": 1.03053904,
      "epoch": 0.05026152828714002,
      "flos": 17705576620800.0,
      "grad_norm": 2.0778779795885534,
      "language_loss": 0.97145164,
      "learning_rate": 3.995719969013709e-06,
      "loss": 0.9948343,
      "num_input_tokens_seen": 8976675,
      "step": 418,
      "time_per_iteration": 2.7102177143096924
    },
    {
      "auxiliary_loss_clip": 0.01253479,
      "auxiliary_loss_mlp": 0.01048254,
      "balance_loss_clip": 1.06865847,
      "balance_loss_mlp": 1.03196967,
      "epoch": 0.05038177117777912,
      "flos": 19133277477120.0,
      "grad_norm": 15.389809140542523,
      "language_loss": 0.85902262,
      "learning_rate": 3.995668883071655e-06,
      "loss": 0.8820399,
      "num_input_tokens_seen": 8992900,
      "step": 419,
      "time_per_iteration": 2.7032995223999023
    },
    {
      "auxiliary_loss_clip": 0.01292975,
      "auxiliary_loss_mlp": 0.0105044,
      "balance_loss_clip": 1.07574022,
      "balance_loss_mlp": 1.034549,
      "epoch": 0.050502014068418206,
      "flos": 20667704618880.0,
      "grad_norm": 2.279737523478281,
      "language_loss": 0.91151357,
      "learning_rate": 3.995617494388219e-06,
      "loss": 0.93494773,
      "num_input_tokens_seen": 9011020,
      "step": 420,
      "time_per_iteration": 2.6725447177886963
    },
    {
      "auxiliary_loss_clip": 0.012497,
      "auxiliary_loss_mlp": 0.0105205,
      "balance_loss_clip": 1.06641006,
      "balance_loss_mlp": 1.0353967,
      "epoch": 0.050622256959057296,
      "flos": 21361103740800.0,
      "grad_norm": 2.3897843969472796,
      "language_loss": 0.80586004,
      "learning_rate": 3.995565802971196e-06,
      "loss": 0.82887757,
      "num_input_tokens_seen": 9030995,
      "step": 421,
      "time_per_iteration": 2.718824863433838
    },
    {
      "auxiliary_loss_clip": 0.01250918,
      "auxiliary_loss_mlp": 0.01057063,
      "balance_loss_clip": 1.06974006,
      "balance_loss_mlp": 1.04156613,
      "epoch": 0.050742499849696386,
      "flos": 27673588909440.0,
      "grad_norm": 1.9902930089143869,
      "language_loss": 0.67726701,
      "learning_rate": 3.995513808828427e-06,
      "loss": 0.70034683,
      "num_input_tokens_seen": 9053790,
      "step": 422,
      "time_per_iteration": 2.796081066131592
    },
    {
      "auxiliary_loss_clip": 0.01248079,
      "auxiliary_loss_mlp": 0.01048972,
      "balance_loss_clip": 1.06643867,
      "balance_loss_mlp": 1.03370106,
      "epoch": 0.050862742740335476,
      "flos": 19865999013120.0,
      "grad_norm": 2.3613161439233155,
      "language_loss": 0.7661202,
      "learning_rate": 3.9954615119678e-06,
      "loss": 0.78909063,
      "num_input_tokens_seen": 9072345,
      "step": 423,
      "time_per_iteration": 2.702320098876953
    },
    {
      "auxiliary_loss_clip": 0.01257682,
      "auxiliary_loss_mlp": 0.01043432,
      "balance_loss_clip": 1.06791151,
      "balance_loss_mlp": 1.0288291,
      "epoch": 0.050982985630974566,
      "flos": 22085098272000.0,
      "grad_norm": 3.4490032738996503,
      "language_loss": 0.80611616,
      "learning_rate": 3.995408912397248e-06,
      "loss": 0.82912737,
      "num_input_tokens_seen": 9090240,
      "step": 424,
      "time_per_iteration": 2.7180817127227783
    },
    {
      "auxiliary_loss_clip": 0.01253727,
      "auxiliary_loss_mlp": 0.01054966,
      "balance_loss_clip": 1.07150888,
      "balance_loss_mlp": 1.03944468,
      "epoch": 0.05110322852161366,
      "flos": 20740962407040.0,
      "grad_norm": 2.4383242354792465,
      "language_loss": 0.93390119,
      "learning_rate": 3.99535601012475e-06,
      "loss": 0.95698822,
      "num_input_tokens_seen": 9105570,
      "step": 425,
      "time_per_iteration": 2.72660756111145
    },
    {
      "auxiliary_loss_clip": 0.01240089,
      "auxiliary_loss_mlp": 0.0076717,
      "balance_loss_clip": 1.06959558,
      "balance_loss_mlp": 1.00052357,
      "epoch": 0.05122347141225275,
      "flos": 28547295327360.0,
      "grad_norm": 1.8595131581305586,
      "language_loss": 0.75537908,
      "learning_rate": 3.995302805158333e-06,
      "loss": 0.77545166,
      "num_input_tokens_seen": 9128225,
      "step": 426,
      "time_per_iteration": 2.803004741668701
    },
    {
      "auxiliary_loss_clip": 0.01242193,
      "auxiliary_loss_mlp": 0.01059346,
      "balance_loss_clip": 1.06623244,
      "balance_loss_mlp": 1.04293144,
      "epoch": 0.05134371430289184,
      "flos": 19722679747200.0,
      "grad_norm": 1.8942337146221704,
      "language_loss": 0.83664978,
      "learning_rate": 3.9952492975060665e-06,
      "loss": 0.85966516,
      "num_input_tokens_seen": 9148295,
      "step": 427,
      "time_per_iteration": 2.8120460510253906
    },
    {
      "auxiliary_loss_clip": 0.01268533,
      "auxiliary_loss_mlp": 0.01050392,
      "balance_loss_clip": 1.07163215,
      "balance_loss_mlp": 1.03432274,
      "epoch": 0.05146395719353093,
      "flos": 34458945649920.0,
      "grad_norm": 4.037054317180963,
      "language_loss": 0.84818387,
      "learning_rate": 3.995195487176067e-06,
      "loss": 0.87137312,
      "num_input_tokens_seen": 9168525,
      "step": 428,
      "time_per_iteration": 2.7524492740631104
    },
    {
      "auxiliary_loss_clip": 0.01285384,
      "auxiliary_loss_mlp": 0.01058417,
      "balance_loss_clip": 1.07193899,
      "balance_loss_mlp": 1.0431459,
      "epoch": 0.05158420008417002,
      "flos": 21760286561280.0,
      "grad_norm": 1.77901377161528,
      "language_loss": 0.85738152,
      "learning_rate": 3.995141374176499e-06,
      "loss": 0.88081956,
      "num_input_tokens_seen": 9186920,
      "step": 429,
      "time_per_iteration": 2.7009294033050537
    },
    {
      "auxiliary_loss_clip": 0.01148415,
      "auxiliary_loss_mlp": 0.0075989,
      "balance_loss_clip": 1.04057348,
      "balance_loss_mlp": 1.00046813,
      "epoch": 0.05170444297480911,
      "flos": 72553956226560.0,
      "grad_norm": 0.8783884477988154,
      "language_loss": 0.63112628,
      "learning_rate": 3.995086958515572e-06,
      "loss": 0.65020931,
      "num_input_tokens_seen": 9244940,
      "step": 430,
      "time_per_iteration": 3.312316656112671
    },
    {
      "auxiliary_loss_clip": 0.0119209,
      "auxiliary_loss_mlp": 0.00759751,
      "balance_loss_clip": 1.04711723,
      "balance_loss_mlp": 1.00055933,
      "epoch": 0.05182468586544821,
      "flos": 62416159326720.0,
      "grad_norm": 0.8572921777334838,
      "language_loss": 0.59979105,
      "learning_rate": 3.995032240201538e-06,
      "loss": 0.61930943,
      "num_input_tokens_seen": 9307335,
      "step": 431,
      "time_per_iteration": 3.166560649871826
    },
    {
      "auxiliary_loss_clip": 0.01169036,
      "auxiliary_loss_mlp": 0.01012862,
      "balance_loss_clip": 1.0438782,
      "balance_loss_mlp": 1.00480366,
      "epoch": 0.0519449287560873,
      "flos": 41225989432320.0,
      "grad_norm": 0.9299667868115054,
      "language_loss": 0.63100147,
      "learning_rate": 3.9949772192427e-06,
      "loss": 0.65282047,
      "num_input_tokens_seen": 9353960,
      "step": 432,
      "time_per_iteration": 2.9726219177246094
    },
    {
      "auxiliary_loss_clip": 0.01249352,
      "auxiliary_loss_mlp": 0.01050096,
      "balance_loss_clip": 1.06600463,
      "balance_loss_mlp": 1.03540373,
      "epoch": 0.05206517164672639,
      "flos": 17494530261120.0,
      "grad_norm": 1.8212929727111247,
      "language_loss": 0.79519093,
      "learning_rate": 3.994921895647405e-06,
      "loss": 0.81818545,
      "num_input_tokens_seen": 9372130,
      "step": 433,
      "time_per_iteration": 3.6096909046173096
    },
    {
      "auxiliary_loss_clip": 0.01188595,
      "auxiliary_loss_mlp": 0.01008375,
      "balance_loss_clip": 1.04452968,
      "balance_loss_mlp": 1.00045919,
      "epoch": 0.05218541453736548,
      "flos": 64002762973440.0,
      "grad_norm": 0.8436779026574405,
      "language_loss": 0.55351132,
      "learning_rate": 3.994866269424043e-06,
      "loss": 0.575481,
      "num_input_tokens_seen": 9428500,
      "step": 434,
      "time_per_iteration": 3.1480226516723633
    },
    {
      "auxiliary_loss_clip": 0.01198131,
      "auxiliary_loss_mlp": 0.010505,
      "balance_loss_clip": 1.05566788,
      "balance_loss_mlp": 1.03530049,
      "epoch": 0.05230565742800457,
      "flos": 19317319787520.0,
      "grad_norm": 2.7931773097865755,
      "language_loss": 0.78612697,
      "learning_rate": 3.9948103405810545e-06,
      "loss": 0.8086133,
      "num_input_tokens_seen": 9447450,
      "step": 435,
      "time_per_iteration": 2.82161545753479
    },
    {
      "auxiliary_loss_clip": 0.01213337,
      "auxiliary_loss_mlp": 0.01054466,
      "balance_loss_clip": 1.06108499,
      "balance_loss_mlp": 1.03948092,
      "epoch": 0.05242590031864366,
      "flos": 25298636538240.0,
      "grad_norm": 1.9585017290646973,
      "language_loss": 0.86196637,
      "learning_rate": 3.994754109126923e-06,
      "loss": 0.88464439,
      "num_input_tokens_seen": 9468945,
      "step": 436,
      "time_per_iteration": 3.732786178588867
    },
    {
      "auxiliary_loss_clip": 0.01190644,
      "auxiliary_loss_mlp": 0.01058864,
      "balance_loss_clip": 1.06077528,
      "balance_loss_mlp": 1.04425502,
      "epoch": 0.052546143209282754,
      "flos": 26211629456640.0,
      "grad_norm": 3.102487603203787,
      "language_loss": 0.93438071,
      "learning_rate": 3.994697575070181e-06,
      "loss": 0.9568758,
      "num_input_tokens_seen": 9488405,
      "step": 437,
      "time_per_iteration": 4.724217414855957
    },
    {
      "auxiliary_loss_clip": 0.01252051,
      "auxiliary_loss_mlp": 0.01046012,
      "balance_loss_clip": 1.07170343,
      "balance_loss_mlp": 1.03194571,
      "epoch": 0.052666386099921844,
      "flos": 22158140578560.0,
      "grad_norm": 1.8777815103365687,
      "language_loss": 0.91609454,
      "learning_rate": 3.994640738419402e-06,
      "loss": 0.93907517,
      "num_input_tokens_seen": 9507780,
      "step": 438,
      "time_per_iteration": 2.7505035400390625
    },
    {
      "auxiliary_loss_clip": 0.01263034,
      "auxiliary_loss_mlp": 0.01054488,
      "balance_loss_clip": 1.06862211,
      "balance_loss_mlp": 1.04023635,
      "epoch": 0.052786628990560934,
      "flos": 23881817502720.0,
      "grad_norm": 2.1098545982706076,
      "language_loss": 0.80693662,
      "learning_rate": 3.9945835991832075e-06,
      "loss": 0.83011186,
      "num_input_tokens_seen": 9529665,
      "step": 439,
      "time_per_iteration": 2.7022833824157715
    },
    {
      "auxiliary_loss_clip": 0.01284868,
      "auxiliary_loss_mlp": 0.01048281,
      "balance_loss_clip": 1.07439876,
      "balance_loss_mlp": 1.03372598,
      "epoch": 0.052906871881200024,
      "flos": 24605021934720.0,
      "grad_norm": 2.195139202210898,
      "language_loss": 0.92865914,
      "learning_rate": 3.994526157370268e-06,
      "loss": 0.9519906,
      "num_input_tokens_seen": 9548280,
      "step": 440,
      "time_per_iteration": 2.71628999710083
    },
    {
      "auxiliary_loss_clip": 0.01162758,
      "auxiliary_loss_mlp": 0.01009402,
      "balance_loss_clip": 1.03917003,
      "balance_loss_mlp": 1.00120056,
      "epoch": 0.053027114771839114,
      "flos": 56461631143680.0,
      "grad_norm": 0.8919355744608258,
      "language_loss": 0.59321082,
      "learning_rate": 3.994468412989296e-06,
      "loss": 0.61493242,
      "num_input_tokens_seen": 9609690,
      "step": 441,
      "time_per_iteration": 3.394660234451294
    },
    {
      "auxiliary_loss_clip": 0.01226917,
      "auxiliary_loss_mlp": 0.01049701,
      "balance_loss_clip": 1.06322169,
      "balance_loss_mlp": 1.03479958,
      "epoch": 0.053147357662478203,
      "flos": 17311098481920.0,
      "grad_norm": 2.7623135640937884,
      "language_loss": 0.9256649,
      "learning_rate": 3.994410366049052e-06,
      "loss": 0.94843102,
      "num_input_tokens_seen": 9627550,
      "step": 442,
      "time_per_iteration": 2.733097791671753
    },
    {
      "auxiliary_loss_clip": 0.01261248,
      "auxiliary_loss_mlp": 0.010415,
      "balance_loss_clip": 1.06894374,
      "balance_loss_mlp": 1.02699256,
      "epoch": 0.0532676005531173,
      "flos": 17164977955200.0,
      "grad_norm": 40.59394791320157,
      "language_loss": 0.83087289,
      "learning_rate": 3.994352016558341e-06,
      "loss": 0.85390043,
      "num_input_tokens_seen": 9644855,
      "step": 443,
      "time_per_iteration": 2.833704948425293
    },
    {
      "auxiliary_loss_clip": 0.01263772,
      "auxiliary_loss_mlp": 0.01055575,
      "balance_loss_clip": 1.07125056,
      "balance_loss_mlp": 1.0408411,
      "epoch": 0.05338784344375639,
      "flos": 27819960831360.0,
      "grad_norm": 1.9552318702250655,
      "language_loss": 0.73857021,
      "learning_rate": 3.994293364526014e-06,
      "loss": 0.76176369,
      "num_input_tokens_seen": 9665740,
      "step": 444,
      "time_per_iteration": 2.785085916519165
    },
    {
      "auxiliary_loss_clip": 0.01241292,
      "auxiliary_loss_mlp": 0.01049108,
      "balance_loss_clip": 1.06850517,
      "balance_loss_mlp": 1.03283572,
      "epoch": 0.05350808633439548,
      "flos": 21507691144320.0,
      "grad_norm": 2.2609387871723112,
      "language_loss": 0.84899586,
      "learning_rate": 3.99423440996097e-06,
      "loss": 0.87189984,
      "num_input_tokens_seen": 9685280,
      "step": 445,
      "time_per_iteration": 2.7917685508728027
    },
    {
      "auxiliary_loss_clip": 0.01254602,
      "auxiliary_loss_mlp": 0.01054763,
      "balance_loss_clip": 1.07142305,
      "balance_loss_mlp": 1.03919411,
      "epoch": 0.05362832922503457,
      "flos": 20084299920000.0,
      "grad_norm": 31.025720900556042,
      "language_loss": 0.81410706,
      "learning_rate": 3.994175152872152e-06,
      "loss": 0.8372007,
      "num_input_tokens_seen": 9704365,
      "step": 446,
      "time_per_iteration": 2.7781684398651123
    },
    {
      "auxiliary_loss_clip": 0.01265154,
      "auxiliary_loss_mlp": 0.01060476,
      "balance_loss_clip": 1.06833529,
      "balance_loss_mlp": 1.04597461,
      "epoch": 0.05374857211567366,
      "flos": 26137222433280.0,
      "grad_norm": 2.5944538226610674,
      "language_loss": 0.78682315,
      "learning_rate": 3.994115593268548e-06,
      "loss": 0.81007946,
      "num_input_tokens_seen": 9724145,
      "step": 447,
      "time_per_iteration": 2.7722597122192383
    },
    {
      "auxiliary_loss_clip": 0.0128086,
      "auxiliary_loss_mlp": 0.01051094,
      "balance_loss_clip": 1.07252359,
      "balance_loss_mlp": 1.03625226,
      "epoch": 0.05386881500631275,
      "flos": 27486817165440.0,
      "grad_norm": 3.4450317371169272,
      "language_loss": 0.82241881,
      "learning_rate": 3.994055731159195e-06,
      "loss": 0.84573835,
      "num_input_tokens_seen": 9741615,
      "step": 448,
      "time_per_iteration": 2.8158562183380127
    },
    {
      "auxiliary_loss_clip": 0.01270235,
      "auxiliary_loss_mlp": 0.01053666,
      "balance_loss_clip": 1.0761708,
      "balance_loss_mlp": 1.03918767,
      "epoch": 0.053989057896951846,
      "flos": 23585087249280.0,
      "grad_norm": 1.987517126185603,
      "language_loss": 0.87227154,
      "learning_rate": 3.993995566553172e-06,
      "loss": 0.89551055,
      "num_input_tokens_seen": 9760580,
      "step": 449,
      "time_per_iteration": 2.805704355239868
    },
    {
      "auxiliary_loss_clip": 0.01233077,
      "auxiliary_loss_mlp": 0.01046941,
      "balance_loss_clip": 1.06122851,
      "balance_loss_mlp": 1.03199208,
      "epoch": 0.054109300787590936,
      "flos": 25228862369280.0,
      "grad_norm": 1.5755731373425914,
      "language_loss": 0.77104473,
      "learning_rate": 3.993935099459607e-06,
      "loss": 0.79384494,
      "num_input_tokens_seen": 9782195,
      "step": 450,
      "time_per_iteration": 2.8505988121032715
    },
    {
      "auxiliary_loss_clip": 0.01274172,
      "auxiliary_loss_mlp": 0.01043252,
      "balance_loss_clip": 1.06929421,
      "balance_loss_mlp": 1.02987683,
      "epoch": 0.054229543678230026,
      "flos": 23841525421440.0,
      "grad_norm": 3.9498097552442273,
      "language_loss": 0.74033761,
      "learning_rate": 3.993874329887673e-06,
      "loss": 0.7635119,
      "num_input_tokens_seen": 9800850,
      "step": 451,
      "time_per_iteration": 2.732203483581543
    },
    {
      "auxiliary_loss_clip": 0.01265463,
      "auxiliary_loss_mlp": 0.01049317,
      "balance_loss_clip": 1.07213509,
      "balance_loss_mlp": 1.03590584,
      "epoch": 0.054349786568869116,
      "flos": 16320933192960.0,
      "grad_norm": 2.5003782429865873,
      "language_loss": 0.86026299,
      "learning_rate": 3.993813257846589e-06,
      "loss": 0.88341081,
      "num_input_tokens_seen": 9817605,
      "step": 452,
      "time_per_iteration": 2.789525270462036
    },
    {
      "auxiliary_loss_clip": 0.01269379,
      "auxiliary_loss_mlp": 0.01047437,
      "balance_loss_clip": 1.07324314,
      "balance_loss_mlp": 1.03290558,
      "epoch": 0.054470029459508205,
      "flos": 18660729127680.0,
      "grad_norm": 2.1871569514474327,
      "language_loss": 0.92873228,
      "learning_rate": 3.993751883345619e-06,
      "loss": 0.95190042,
      "num_input_tokens_seen": 9835965,
      "step": 453,
      "time_per_iteration": 2.73264741897583
    },
    {
      "auxiliary_loss_clip": 0.01245576,
      "auxiliary_loss_mlp": 0.01052734,
      "balance_loss_clip": 1.06951928,
      "balance_loss_mlp": 1.03723717,
      "epoch": 0.054590272350147295,
      "flos": 17785298856960.0,
      "grad_norm": 2.381938006186589,
      "language_loss": 0.87716043,
      "learning_rate": 3.993690206394073e-06,
      "loss": 0.9001435,
      "num_input_tokens_seen": 9852265,
      "step": 454,
      "time_per_iteration": 2.7695465087890625
    },
    {
      "auxiliary_loss_clip": 0.01256693,
      "auxiliary_loss_mlp": 0.01052512,
      "balance_loss_clip": 1.07246852,
      "balance_loss_mlp": 1.03707409,
      "epoch": 0.054710515240786385,
      "flos": 17785945301760.0,
      "grad_norm": 2.1616230759216974,
      "language_loss": 0.87349772,
      "learning_rate": 3.993628227001307e-06,
      "loss": 0.89658976,
      "num_input_tokens_seen": 9870465,
      "step": 455,
      "time_per_iteration": 2.713808298110962
    },
    {
      "auxiliary_loss_clip": 0.01246105,
      "auxiliary_loss_mlp": 0.01049585,
      "balance_loss_clip": 1.06506479,
      "balance_loss_mlp": 1.03551793,
      "epoch": 0.05483075813142548,
      "flos": 48210900180480.0,
      "grad_norm": 1.9550749895072734,
      "language_loss": 0.71251023,
      "learning_rate": 3.993565945176726e-06,
      "loss": 0.73546714,
      "num_input_tokens_seen": 9891490,
      "step": 456,
      "time_per_iteration": 2.9432170391082764
    },
    {
      "auxiliary_loss_clip": 0.01248289,
      "auxiliary_loss_mlp": 0.01056822,
      "balance_loss_clip": 1.06962991,
      "balance_loss_mlp": 1.04238582,
      "epoch": 0.05495100102206457,
      "flos": 19682244011520.0,
      "grad_norm": 2.2005711799880294,
      "language_loss": 0.83943439,
      "learning_rate": 3.993503360929776e-06,
      "loss": 0.86248553,
      "num_input_tokens_seen": 9910375,
      "step": 457,
      "time_per_iteration": 2.6838905811309814
    },
    {
      "auxiliary_loss_clip": 0.01187822,
      "auxiliary_loss_mlp": 0.0105221,
      "balance_loss_clip": 1.06155205,
      "balance_loss_mlp": 1.03690338,
      "epoch": 0.05507124391270366,
      "flos": 26360048453760.0,
      "grad_norm": 1.6093704727088132,
      "language_loss": 0.81191719,
      "learning_rate": 3.99344047426995e-06,
      "loss": 0.83431751,
      "num_input_tokens_seen": 9931635,
      "step": 458,
      "time_per_iteration": 3.3502273559570312
    },
    {
      "auxiliary_loss_clip": 0.01220043,
      "auxiliary_loss_mlp": 0.0104713,
      "balance_loss_clip": 1.06316185,
      "balance_loss_mlp": 1.03209734,
      "epoch": 0.05519148680334275,
      "flos": 22601314581120.0,
      "grad_norm": 2.462518909384051,
      "language_loss": 0.93592596,
      "learning_rate": 3.993377285206789e-06,
      "loss": 0.95859766,
      "num_input_tokens_seen": 9951420,
      "step": 459,
      "time_per_iteration": 4.177008628845215
    },
    {
      "auxiliary_loss_clip": 0.01206826,
      "auxiliary_loss_mlp": 0.01045207,
      "balance_loss_clip": 1.06054378,
      "balance_loss_mlp": 1.03016281,
      "epoch": 0.05531172969398184,
      "flos": 40552519380480.0,
      "grad_norm": 1.6872489127973465,
      "language_loss": 0.86609066,
      "learning_rate": 3.99331379374988e-06,
      "loss": 0.88861102,
      "num_input_tokens_seen": 9975025,
      "step": 460,
      "time_per_iteration": 2.9176807403564453
    },
    {
      "auxiliary_loss_clip": 0.01248593,
      "auxiliary_loss_mlp": 0.01049551,
      "balance_loss_clip": 1.0618906,
      "balance_loss_mlp": 1.03538251,
      "epoch": 0.05543197258462093,
      "flos": 23477894087040.0,
      "grad_norm": 1.9805208460318446,
      "language_loss": 0.79866087,
      "learning_rate": 3.993249999908852e-06,
      "loss": 0.82164228,
      "num_input_tokens_seen": 9995175,
      "step": 461,
      "time_per_iteration": 2.7650017738342285
    },
    {
      "auxiliary_loss_clip": 0.01278323,
      "auxiliary_loss_mlp": 0.01053803,
      "balance_loss_clip": 1.06990552,
      "balance_loss_mlp": 1.03959322,
      "epoch": 0.05555221547526003,
      "flos": 18624603024000.0,
      "grad_norm": 1.9998614293548942,
      "language_loss": 0.87319207,
      "learning_rate": 3.993185903693384e-06,
      "loss": 0.89651334,
      "num_input_tokens_seen": 10011975,
      "step": 462,
      "time_per_iteration": 4.56042218208313
    },
    {
      "auxiliary_loss_clip": 0.01251184,
      "auxiliary_loss_mlp": 0.0104919,
      "balance_loss_clip": 1.06962848,
      "balance_loss_mlp": 1.03504014,
      "epoch": 0.05567245836589912,
      "flos": 23587098410880.0,
      "grad_norm": 3.0132774333864525,
      "language_loss": 0.82631147,
      "learning_rate": 3.9931215051131995e-06,
      "loss": 0.84931517,
      "num_input_tokens_seen": 10032620,
      "step": 463,
      "time_per_iteration": 3.7687015533447266
    },
    {
      "auxiliary_loss_clip": 0.01251473,
      "auxiliary_loss_mlp": 0.01051817,
      "balance_loss_clip": 1.06551528,
      "balance_loss_mlp": 1.03670144,
      "epoch": 0.05579270125653821,
      "flos": 27746667129600.0,
      "grad_norm": 1.6337387684428168,
      "language_loss": 0.80217737,
      "learning_rate": 3.993056804178068e-06,
      "loss": 0.82521027,
      "num_input_tokens_seen": 10054165,
      "step": 464,
      "time_per_iteration": 2.761094331741333
    },
    {
      "auxiliary_loss_clip": 0.01212272,
      "auxiliary_loss_mlp": 0.01050303,
      "balance_loss_clip": 1.06697702,
      "balance_loss_mlp": 1.0353657,
      "epoch": 0.0559129441471773,
      "flos": 27014161075200.0,
      "grad_norm": 1.888333267000692,
      "language_loss": 0.84200609,
      "learning_rate": 3.992991800897803e-06,
      "loss": 0.86463189,
      "num_input_tokens_seen": 10073970,
      "step": 465,
      "time_per_iteration": 2.839029550552368
    },
    {
      "auxiliary_loss_clip": 0.01278467,
      "auxiliary_loss_mlp": 0.01045914,
      "balance_loss_clip": 1.07003558,
      "balance_loss_mlp": 1.03174043,
      "epoch": 0.05603318703781639,
      "flos": 15229787794560.0,
      "grad_norm": 2.5974444680140016,
      "language_loss": 0.89868486,
      "learning_rate": 3.9929264952822665e-06,
      "loss": 0.9219287,
      "num_input_tokens_seen": 10091505,
      "step": 466,
      "time_per_iteration": 2.618161201477051
    },
    {
      "auxiliary_loss_clip": 0.01268164,
      "auxiliary_loss_mlp": 0.01046575,
      "balance_loss_clip": 1.06962729,
      "balance_loss_mlp": 1.03172183,
      "epoch": 0.05615342992845548,
      "flos": 22266482976000.0,
      "grad_norm": 2.219720269199115,
      "language_loss": 0.88394034,
      "learning_rate": 3.992860887341366e-06,
      "loss": 0.90708768,
      "num_input_tokens_seen": 10109675,
      "step": 467,
      "time_per_iteration": 2.6831932067871094
    },
    {
      "auxiliary_loss_clip": 0.01218039,
      "auxiliary_loss_mlp": 0.01052648,
      "balance_loss_clip": 1.06198335,
      "balance_loss_mlp": 1.03824794,
      "epoch": 0.056273672819094574,
      "flos": 23584979508480.0,
      "grad_norm": 2.0733929362780934,
      "language_loss": 0.8127926,
      "learning_rate": 3.992794977085052e-06,
      "loss": 0.83549953,
      "num_input_tokens_seen": 10127675,
      "step": 468,
      "time_per_iteration": 2.9471590518951416
    },
    {
      "auxiliary_loss_clip": 0.01231524,
      "auxiliary_loss_mlp": 0.010541,
      "balance_loss_clip": 1.06622851,
      "balance_loss_mlp": 1.0384475,
      "epoch": 0.056393915709733664,
      "flos": 19858708552320.0,
      "grad_norm": 1.9863433249078764,
      "language_loss": 0.85034794,
      "learning_rate": 3.992728764523326e-06,
      "loss": 0.87320423,
      "num_input_tokens_seen": 10146620,
      "step": 469,
      "time_per_iteration": 2.8474512100219727
    },
    {
      "auxiliary_loss_clip": 0.01245232,
      "auxiliary_loss_mlp": 0.01044407,
      "balance_loss_clip": 1.06789255,
      "balance_loss_mlp": 1.02966046,
      "epoch": 0.05651415860037275,
      "flos": 22163779013760.0,
      "grad_norm": 1.8213229695629185,
      "language_loss": 0.8086611,
      "learning_rate": 3.99266224966623e-06,
      "loss": 0.83155745,
      "num_input_tokens_seen": 10167535,
      "step": 470,
      "time_per_iteration": 2.7712759971618652
    },
    {
      "auxiliary_loss_clip": 0.01235439,
      "auxiliary_loss_mlp": 0.01050439,
      "balance_loss_clip": 1.06718624,
      "balance_loss_mlp": 1.03572869,
      "epoch": 0.05663440149101184,
      "flos": 19463548055040.0,
      "grad_norm": 2.4560209365963654,
      "language_loss": 0.87823743,
      "learning_rate": 3.992595432523855e-06,
      "loss": 0.90109622,
      "num_input_tokens_seen": 10184825,
      "step": 471,
      "time_per_iteration": 2.7032275199890137
    },
    {
      "auxiliary_loss_clip": 0.0122151,
      "auxiliary_loss_mlp": 0.0104761,
      "balance_loss_clip": 1.0654614,
      "balance_loss_mlp": 1.03284049,
      "epoch": 0.05675464438165093,
      "flos": 22670226823680.0,
      "grad_norm": 1.9344620106606596,
      "language_loss": 0.85925663,
      "learning_rate": 3.992528313106338e-06,
      "loss": 0.88194782,
      "num_input_tokens_seen": 10203025,
      "step": 472,
      "time_per_iteration": 2.768698215484619
    },
    {
      "auxiliary_loss_clip": 0.01281234,
      "auxiliary_loss_mlp": 0.00766933,
      "balance_loss_clip": 1.07621026,
      "balance_loss_mlp": 1.00052273,
      "epoch": 0.05687488727229002,
      "flos": 16901177495040.0,
      "grad_norm": 2.608944679978251,
      "language_loss": 0.8213855,
      "learning_rate": 3.9924608914238595e-06,
      "loss": 0.84186721,
      "num_input_tokens_seen": 10218020,
      "step": 473,
      "time_per_iteration": 2.6292810440063477
    },
    {
      "auxiliary_loss_clip": 0.01265457,
      "auxiliary_loss_mlp": 0.01050484,
      "balance_loss_clip": 1.07285976,
      "balance_loss_mlp": 1.03624475,
      "epoch": 0.05699513016292912,
      "flos": 29168980945920.0,
      "grad_norm": 2.3892819267688936,
      "language_loss": 0.84047496,
      "learning_rate": 3.992393167486648e-06,
      "loss": 0.86363435,
      "num_input_tokens_seen": 10237170,
      "step": 474,
      "time_per_iteration": 2.787321090698242
    },
    {
      "auxiliary_loss_clip": 0.01281804,
      "auxiliary_loss_mlp": 0.01048677,
      "balance_loss_clip": 1.07408786,
      "balance_loss_mlp": 1.0342288,
      "epoch": 0.05711537305356821,
      "flos": 18916197632640.0,
      "grad_norm": 2.1886057922156605,
      "language_loss": 0.80717599,
      "learning_rate": 3.992325141304977e-06,
      "loss": 0.83048081,
      "num_input_tokens_seen": 10255125,
      "step": 475,
      "time_per_iteration": 2.690633773803711
    },
    {
      "auxiliary_loss_clip": 0.01211362,
      "auxiliary_loss_mlp": 0.01052831,
      "balance_loss_clip": 1.06264138,
      "balance_loss_mlp": 1.03932464,
      "epoch": 0.0572356159442073,
      "flos": 26758979879040.0,
      "grad_norm": 4.199083840588372,
      "language_loss": 0.86807591,
      "learning_rate": 3.992256812889166e-06,
      "loss": 0.89071786,
      "num_input_tokens_seen": 10271230,
      "step": 476,
      "time_per_iteration": 2.7647175788879395
    },
    {
      "auxiliary_loss_clip": 0.01281466,
      "auxiliary_loss_mlp": 0.01047677,
      "balance_loss_clip": 1.0757134,
      "balance_loss_mlp": 1.03187037,
      "epoch": 0.05735585883484639,
      "flos": 35116146840960.0,
      "grad_norm": 2.7396963555069553,
      "language_loss": 0.76558501,
      "learning_rate": 3.992188182249582e-06,
      "loss": 0.78887641,
      "num_input_tokens_seen": 10293125,
      "step": 477,
      "time_per_iteration": 2.7652716636657715
    },
    {
      "auxiliary_loss_clip": 0.01247579,
      "auxiliary_loss_mlp": 0.01043338,
      "balance_loss_clip": 1.07075047,
      "balance_loss_mlp": 1.02824628,
      "epoch": 0.05747610172548548,
      "flos": 18734381965440.0,
      "grad_norm": 2.0954455570447457,
      "language_loss": 0.9078356,
      "learning_rate": 3.992119249396633e-06,
      "loss": 0.93074483,
      "num_input_tokens_seen": 10311810,
      "step": 478,
      "time_per_iteration": 2.7219924926757812
    },
    {
      "auxiliary_loss_clip": 0.01238912,
      "auxiliary_loss_mlp": 0.00767005,
      "balance_loss_clip": 1.06746769,
      "balance_loss_mlp": 1.00043774,
      "epoch": 0.05759634461612457,
      "flos": 27964752554880.0,
      "grad_norm": 2.0770656657548385,
      "language_loss": 0.82204419,
      "learning_rate": 3.992050014340778e-06,
      "loss": 0.84210336,
      "num_input_tokens_seen": 10332165,
      "step": 479,
      "time_per_iteration": 2.8032612800598145
    },
    {
      "auxiliary_loss_clip": 0.01166483,
      "auxiliary_loss_mlp": 0.01014648,
      "balance_loss_clip": 1.04443502,
      "balance_loss_mlp": 1.00768602,
      "epoch": 0.057716587506763666,
      "flos": 69292009405440.0,
      "grad_norm": 0.910905180236274,
      "language_loss": 0.55053842,
      "learning_rate": 3.99198047709252e-06,
      "loss": 0.57234979,
      "num_input_tokens_seen": 10393685,
      "step": 480,
      "time_per_iteration": 3.3368217945098877
    },
    {
      "auxiliary_loss_clip": 0.01227499,
      "auxiliary_loss_mlp": 0.01044985,
      "balance_loss_clip": 1.06382024,
      "balance_loss_mlp": 1.03000081,
      "epoch": 0.057836830397402755,
      "flos": 25009196745600.0,
      "grad_norm": 1.8878339558894823,
      "language_loss": 0.7901237,
      "learning_rate": 3.991910637662408e-06,
      "loss": 0.81284857,
      "num_input_tokens_seen": 10413975,
      "step": 481,
      "time_per_iteration": 2.8201568126678467
    },
    {
      "auxiliary_loss_clip": 0.01275848,
      "auxiliary_loss_mlp": 0.01048846,
      "balance_loss_clip": 1.07207441,
      "balance_loss_mlp": 1.03450465,
      "epoch": 0.057957073288041845,
      "flos": 25593894334080.0,
      "grad_norm": 2.3609264202607214,
      "language_loss": 0.80945981,
      "learning_rate": 3.9918404960610355e-06,
      "loss": 0.83270675,
      "num_input_tokens_seen": 10433005,
      "step": 482,
      "time_per_iteration": 2.705639123916626
    },
    {
      "auxiliary_loss_clip": 0.01261971,
      "auxiliary_loss_mlp": 0.01045315,
      "balance_loss_clip": 1.06633317,
      "balance_loss_mlp": 1.03083134,
      "epoch": 0.058077316178680935,
      "flos": 20777411733120.0,
      "grad_norm": 2.382669675124695,
      "language_loss": 0.77533102,
      "learning_rate": 3.991770052299043e-06,
      "loss": 0.79840386,
      "num_input_tokens_seen": 10451235,
      "step": 483,
      "time_per_iteration": 2.689882516860962
    },
    {
      "auxiliary_loss_clip": 0.0124657,
      "auxiliary_loss_mlp": 0.01049021,
      "balance_loss_clip": 1.06750917,
      "balance_loss_mlp": 1.03503799,
      "epoch": 0.058197559069320025,
      "flos": 18916484941440.0,
      "grad_norm": 2.438731510245966,
      "language_loss": 0.87622619,
      "learning_rate": 3.991699306387118e-06,
      "loss": 0.89918208,
      "num_input_tokens_seen": 10469705,
      "step": 484,
      "time_per_iteration": 2.7403030395507812
    },
    {
      "auxiliary_loss_clip": 0.01264442,
      "auxiliary_loss_mlp": 0.01049458,
      "balance_loss_clip": 1.07049692,
      "balance_loss_mlp": 1.03627968,
      "epoch": 0.058317801959959115,
      "flos": 24863327614080.0,
      "grad_norm": 1.9620359525949826,
      "language_loss": 0.78014684,
      "learning_rate": 3.991628258335991e-06,
      "loss": 0.80328584,
      "num_input_tokens_seen": 10491910,
      "step": 485,
      "time_per_iteration": 2.74094820022583
    },
    {
      "auxiliary_loss_clip": 0.01220403,
      "auxiliary_loss_mlp": 0.01054618,
      "balance_loss_clip": 1.06550455,
      "balance_loss_mlp": 1.0404501,
      "epoch": 0.05843804485059821,
      "flos": 23257977068160.0,
      "grad_norm": 3.390579577241081,
      "language_loss": 0.88070476,
      "learning_rate": 3.991556908156442e-06,
      "loss": 0.9034549,
      "num_input_tokens_seen": 10508435,
      "step": 486,
      "time_per_iteration": 3.697441577911377
    },
    {
      "auxiliary_loss_clip": 0.01243501,
      "auxiliary_loss_mlp": 0.01047741,
      "balance_loss_clip": 1.06441545,
      "balance_loss_mlp": 1.03260148,
      "epoch": 0.0585582877412373,
      "flos": 23150532510720.0,
      "grad_norm": 1.945461884322535,
      "language_loss": 0.88192964,
      "learning_rate": 3.9914852558592914e-06,
      "loss": 0.90484208,
      "num_input_tokens_seen": 10529485,
      "step": 487,
      "time_per_iteration": 2.7593801021575928
    },
    {
      "auxiliary_loss_clip": 0.01261337,
      "auxiliary_loss_mlp": 0.01050059,
      "balance_loss_clip": 1.07014239,
      "balance_loss_mlp": 1.03451419,
      "epoch": 0.05867853063187639,
      "flos": 23506406507520.0,
      "grad_norm": 3.1497987968350474,
      "language_loss": 0.80457318,
      "learning_rate": 3.991413301455413e-06,
      "loss": 0.82768714,
      "num_input_tokens_seen": 10545935,
      "step": 488,
      "time_per_iteration": 3.6796257495880127
    },
    {
      "auxiliary_loss_clip": 0.01230689,
      "auxiliary_loss_mlp": 0.01050321,
      "balance_loss_clip": 1.06328082,
      "balance_loss_mlp": 1.036731,
      "epoch": 0.05879877352251548,
      "flos": 29495803818240.0,
      "grad_norm": 2.664112196381184,
      "language_loss": 0.78087038,
      "learning_rate": 3.991341044955719e-06,
      "loss": 0.80368048,
      "num_input_tokens_seen": 10565690,
      "step": 489,
      "time_per_iteration": 3.920098066329956
    },
    {
      "auxiliary_loss_clip": 0.01254453,
      "auxiliary_loss_mlp": 0.00768141,
      "balance_loss_clip": 1.06582403,
      "balance_loss_mlp": 1.00043344,
      "epoch": 0.05891901641315457,
      "flos": 20157485880960.0,
      "grad_norm": 2.179350087072872,
      "language_loss": 0.81960994,
      "learning_rate": 3.991268486371172e-06,
      "loss": 0.83983588,
      "num_input_tokens_seen": 10584245,
      "step": 490,
      "time_per_iteration": 3.7074098587036133
    },
    {
      "auxiliary_loss_clip": 0.01242201,
      "auxiliary_loss_mlp": 0.01048752,
      "balance_loss_clip": 1.06426835,
      "balance_loss_mlp": 1.03377962,
      "epoch": 0.05903925930379366,
      "flos": 24644200694400.0,
      "grad_norm": 3.1678542666405574,
      "language_loss": 0.87982762,
      "learning_rate": 3.991195625712779e-06,
      "loss": 0.90273714,
      "num_input_tokens_seen": 10601210,
      "step": 491,
      "time_per_iteration": 2.7503597736358643
    },
    {
      "auxiliary_loss_clip": 0.0127464,
      "auxiliary_loss_mlp": 0.01044076,
      "balance_loss_clip": 1.07094598,
      "balance_loss_mlp": 1.0291512,
      "epoch": 0.05915950219443276,
      "flos": 21250391045760.0,
      "grad_norm": 2.215587816392955,
      "language_loss": 0.81682879,
      "learning_rate": 3.991122462991592e-06,
      "loss": 0.84001595,
      "num_input_tokens_seen": 10620730,
      "step": 492,
      "time_per_iteration": 2.694131851196289
    },
    {
      "auxiliary_loss_clip": 0.0128078,
      "auxiliary_loss_mlp": 0.01050445,
      "balance_loss_clip": 1.0711875,
      "balance_loss_mlp": 1.03534126,
      "epoch": 0.05927974508507185,
      "flos": 9902727319680.0,
      "grad_norm": 3.6912335622416554,
      "language_loss": 0.81623566,
      "learning_rate": 3.991048998218712e-06,
      "loss": 0.83954793,
      "num_input_tokens_seen": 10634035,
      "step": 493,
      "time_per_iteration": 2.549713373184204
    },
    {
      "auxiliary_loss_clip": 0.01258553,
      "auxiliary_loss_mlp": 0.01049885,
      "balance_loss_clip": 1.06489217,
      "balance_loss_mlp": 1.03441119,
      "epoch": 0.05939998797571094,
      "flos": 18259499232000.0,
      "grad_norm": 2.472098871333579,
      "language_loss": 0.76432514,
      "learning_rate": 3.990975231405281e-06,
      "loss": 0.78740948,
      "num_input_tokens_seen": 10652485,
      "step": 494,
      "time_per_iteration": 2.6785836219787598
    },
    {
      "auxiliary_loss_clip": 0.01255903,
      "auxiliary_loss_mlp": 0.01051895,
      "balance_loss_clip": 1.06590652,
      "balance_loss_mlp": 1.03771532,
      "epoch": 0.05952023086635003,
      "flos": 28256598558720.0,
      "grad_norm": 7.228829903976104,
      "language_loss": 0.78898716,
      "learning_rate": 3.990901162562491e-06,
      "loss": 0.81206506,
      "num_input_tokens_seen": 10673175,
      "step": 495,
      "time_per_iteration": 2.6979448795318604
    },
    {
      "auxiliary_loss_clip": 0.01225617,
      "auxiliary_loss_mlp": 0.00767687,
      "balance_loss_clip": 1.0589391,
      "balance_loss_mlp": 1.00032973,
      "epoch": 0.05964047375698912,
      "flos": 14902498045440.0,
      "grad_norm": 2.6481135021640774,
      "language_loss": 0.90667564,
      "learning_rate": 3.9908267917015765e-06,
      "loss": 0.92660868,
      "num_input_tokens_seen": 10691235,
      "step": 496,
      "time_per_iteration": 2.7400946617126465
    },
    {
      "auxiliary_loss_clip": 0.01247088,
      "auxiliary_loss_mlp": 0.01055202,
      "balance_loss_clip": 1.06414557,
      "balance_loss_mlp": 1.04073548,
      "epoch": 0.059760716647628206,
      "flos": 23185581206400.0,
      "grad_norm": 2.1674049968123783,
      "language_loss": 0.93090594,
      "learning_rate": 3.990752118833821e-06,
      "loss": 0.95392895,
      "num_input_tokens_seen": 10708675,
      "step": 497,
      "time_per_iteration": 2.6797261238098145
    },
    {
      "auxiliary_loss_clip": 0.01275609,
      "auxiliary_loss_mlp": 0.01056638,
      "balance_loss_clip": 1.07051015,
      "balance_loss_mlp": 1.0419569,
      "epoch": 0.0598809595382673,
      "flos": 22746968231040.0,
      "grad_norm": 2.216177291441323,
      "language_loss": 0.77842438,
      "learning_rate": 3.990677143970553e-06,
      "loss": 0.80174685,
      "num_input_tokens_seen": 10729485,
      "step": 498,
      "time_per_iteration": 2.6700661182403564
    },
    {
      "auxiliary_loss_clip": 0.01226063,
      "auxiliary_loss_mlp": 0.0104723,
      "balance_loss_clip": 1.06796455,
      "balance_loss_mlp": 1.03323472,
      "epoch": 0.06000120242890639,
      "flos": 22127221946880.0,
      "grad_norm": 2.0947280020378205,
      "language_loss": 0.81575596,
      "learning_rate": 3.990601867123144e-06,
      "loss": 0.83848882,
      "num_input_tokens_seen": 10749210,
      "step": 499,
      "time_per_iteration": 2.748843193054199
    },
    {
      "auxiliary_loss_clip": 0.01209316,
      "auxiliary_loss_mlp": 0.01048629,
      "balance_loss_clip": 1.06244588,
      "balance_loss_mlp": 1.03459811,
      "epoch": 0.06012144531954548,
      "flos": 19171773878400.0,
      "grad_norm": 2.8240671772503063,
      "language_loss": 0.85292506,
      "learning_rate": 3.990526288303014e-06,
      "loss": 0.87550449,
      "num_input_tokens_seen": 10768000,
      "step": 500,
      "time_per_iteration": 2.7673041820526123
    },
    {
      "auxiliary_loss_clip": 0.01240141,
      "auxiliary_loss_mlp": 0.00766985,
      "balance_loss_clip": 1.06591153,
      "balance_loss_mlp": 1.00040317,
      "epoch": 0.06024168821018457,
      "flos": 22783345729920.0,
      "grad_norm": 2.0749212144003173,
      "language_loss": 0.90921527,
      "learning_rate": 3.9904504075216295e-06,
      "loss": 0.92928654,
      "num_input_tokens_seen": 10788760,
      "step": 501,
      "time_per_iteration": 2.70145583152771
    },
    {
      "auxiliary_loss_clip": 0.01225496,
      "auxiliary_loss_mlp": 0.01045285,
      "balance_loss_clip": 1.06028736,
      "balance_loss_mlp": 1.03080082,
      "epoch": 0.06036193110082366,
      "flos": 18770687637120.0,
      "grad_norm": 2.259991164034686,
      "language_loss": 0.94295013,
      "learning_rate": 3.990374224790501e-06,
      "loss": 0.96565795,
      "num_input_tokens_seen": 10806965,
      "step": 502,
      "time_per_iteration": 2.8935465812683105
    },
    {
      "auxiliary_loss_clip": 0.01245966,
      "auxiliary_loss_mlp": 0.01053143,
      "balance_loss_clip": 1.0682739,
      "balance_loss_mlp": 1.03862309,
      "epoch": 0.06048217399146275,
      "flos": 17201570935680.0,
      "grad_norm": 2.313371838230714,
      "language_loss": 0.71181738,
      "learning_rate": 3.990297740121185e-06,
      "loss": 0.7348085,
      "num_input_tokens_seen": 10824900,
      "step": 503,
      "time_per_iteration": 2.769035577774048
    },
    {
      "auxiliary_loss_clip": 0.01258025,
      "auxiliary_loss_mlp": 0.00766571,
      "balance_loss_clip": 1.0665257,
      "balance_loss_mlp": 1.00045276,
      "epoch": 0.06060241688210185,
      "flos": 24024131187840.0,
      "grad_norm": 2.040325261108514,
      "language_loss": 0.78292251,
      "learning_rate": 3.990220953525284e-06,
      "loss": 0.80316848,
      "num_input_tokens_seen": 10842010,
      "step": 504,
      "time_per_iteration": 2.7679576873779297
    },
    {
      "auxiliary_loss_clip": 0.01226785,
      "auxiliary_loss_mlp": 0.0104379,
      "balance_loss_clip": 1.06022692,
      "balance_loss_mlp": 1.02940106,
      "epoch": 0.06072265977274094,
      "flos": 14611190745600.0,
      "grad_norm": 2.695157711280703,
      "language_loss": 0.74182594,
      "learning_rate": 3.9901438650144465e-06,
      "loss": 0.76453167,
      "num_input_tokens_seen": 10858260,
      "step": 505,
      "time_per_iteration": 2.728879928588867
    },
    {
      "auxiliary_loss_clip": 0.01249233,
      "auxiliary_loss_mlp": 0.01049029,
      "balance_loss_clip": 1.06654334,
      "balance_loss_mlp": 1.03416371,
      "epoch": 0.06084290266338003,
      "flos": 20558284813440.0,
      "grad_norm": 2.416505483927437,
      "language_loss": 0.91950572,
      "learning_rate": 3.990066474600367e-06,
      "loss": 0.94248831,
      "num_input_tokens_seen": 10876230,
      "step": 506,
      "time_per_iteration": 2.8097784519195557
    },
    {
      "auxiliary_loss_clip": 0.01240689,
      "auxiliary_loss_mlp": 0.01039262,
      "balance_loss_clip": 1.06100965,
      "balance_loss_mlp": 1.02555323,
      "epoch": 0.06096314555401912,
      "flos": 22309217182080.0,
      "grad_norm": 1.8419810860514019,
      "language_loss": 0.67904162,
      "learning_rate": 3.989988782294786e-06,
      "loss": 0.70184112,
      "num_input_tokens_seen": 10896320,
      "step": 507,
      "time_per_iteration": 2.724440097808838
    },
    {
      "auxiliary_loss_clip": 0.01208248,
      "auxiliary_loss_mlp": 0.01053328,
      "balance_loss_clip": 1.05954123,
      "balance_loss_mlp": 1.03902268,
      "epoch": 0.06108338844465821,
      "flos": 19131374056320.0,
      "grad_norm": 1.750057442343561,
      "language_loss": 0.95109737,
      "learning_rate": 3.989910788109489e-06,
      "loss": 0.97371316,
      "num_input_tokens_seen": 10912970,
      "step": 508,
      "time_per_iteration": 2.712320327758789
    },
    {
      "auxiliary_loss_clip": 0.01225518,
      "auxiliary_loss_mlp": 0.01054832,
      "balance_loss_clip": 1.06422436,
      "balance_loss_mlp": 1.04016888,
      "epoch": 0.0612036313352973,
      "flos": 33584018169600.0,
      "grad_norm": 3.2343305592553415,
      "language_loss": 0.74750209,
      "learning_rate": 3.989832492056307e-06,
      "loss": 0.77030563,
      "num_input_tokens_seen": 10933995,
      "step": 509,
      "time_per_iteration": 2.8885419368743896
    },
    {
      "auxiliary_loss_clip": 0.01261598,
      "auxiliary_loss_mlp": 0.01045326,
      "balance_loss_clip": 1.07048535,
      "balance_loss_mlp": 1.02906609,
      "epoch": 0.06132387422593639,
      "flos": 27490552179840.0,
      "grad_norm": 2.231000134122838,
      "language_loss": 0.8104229,
      "learning_rate": 3.989753894147119e-06,
      "loss": 0.83349204,
      "num_input_tokens_seen": 10954120,
      "step": 510,
      "time_per_iteration": 2.694120168685913
    },
    {
      "auxiliary_loss_clip": 0.01255507,
      "auxiliary_loss_mlp": 0.01053669,
      "balance_loss_clip": 1.07117128,
      "balance_loss_mlp": 1.03963196,
      "epoch": 0.061444117116575485,
      "flos": 25885057979520.0,
      "grad_norm": 2.5728429134818493,
      "language_loss": 0.80061519,
      "learning_rate": 3.989674994393846e-06,
      "loss": 0.82370698,
      "num_input_tokens_seen": 10973595,
      "step": 511,
      "time_per_iteration": 3.660067558288574
    },
    {
      "auxiliary_loss_clip": 0.01257031,
      "auxiliary_loss_mlp": 0.01044658,
      "balance_loss_clip": 1.06993628,
      "balance_loss_mlp": 1.02997136,
      "epoch": 0.061564360007214575,
      "flos": 28512031150080.0,
      "grad_norm": 4.286099785148859,
      "language_loss": 0.94174403,
      "learning_rate": 3.98959579280846e-06,
      "loss": 0.96476096,
      "num_input_tokens_seen": 10991995,
      "step": 512,
      "time_per_iteration": 2.8348708152770996
    },
    {
      "auxiliary_loss_clip": 0.01197894,
      "auxiliary_loss_mlp": 0.01049659,
      "balance_loss_clip": 1.06280112,
      "balance_loss_mlp": 1.03482962,
      "epoch": 0.061684602897853665,
      "flos": 12094355652480.0,
      "grad_norm": 2.0043007164421436,
      "language_loss": 0.82988131,
      "learning_rate": 3.989516289402973e-06,
      "loss": 0.85235685,
      "num_input_tokens_seen": 11007625,
      "step": 513,
      "time_per_iteration": 2.980318546295166
    },
    {
      "auxiliary_loss_clip": 0.01170921,
      "auxiliary_loss_mlp": 0.0105,
      "balance_loss_clip": 1.05143929,
      "balance_loss_mlp": 1.03497946,
      "epoch": 0.061804845788492754,
      "flos": 19532639865600.0,
      "grad_norm": 3.063410894067564,
      "language_loss": 0.8056106,
      "learning_rate": 3.989436484189447e-06,
      "loss": 0.82781982,
      "num_input_tokens_seen": 11025570,
      "step": 514,
      "time_per_iteration": 4.8572142124176025
    },
    {
      "auxiliary_loss_clip": 0.01259078,
      "auxiliary_loss_mlp": 0.01043137,
      "balance_loss_clip": 1.06657028,
      "balance_loss_mlp": 1.02834296,
      "epoch": 0.061925088679131844,
      "flos": 15341111020800.0,
      "grad_norm": 2.5145404893792445,
      "language_loss": 0.81089461,
      "learning_rate": 3.9893563771799885e-06,
      "loss": 0.83391678,
      "num_input_tokens_seen": 11042045,
      "step": 515,
      "time_per_iteration": 2.8499398231506348
    },
    {
      "auxiliary_loss_clip": 0.01275873,
      "auxiliary_loss_mlp": 0.01048766,
      "balance_loss_clip": 1.07002401,
      "balance_loss_mlp": 1.03359032,
      "epoch": 0.062045331569770934,
      "flos": 25919927107200.0,
      "grad_norm": 2.36639266706255,
      "language_loss": 0.86121714,
      "learning_rate": 3.989275968386749e-06,
      "loss": 0.88446349,
      "num_input_tokens_seen": 11059955,
      "step": 516,
      "time_per_iteration": 3.676647663116455
    },
    {
      "auxiliary_loss_clip": 0.01228246,
      "auxiliary_loss_mlp": 0.01053011,
      "balance_loss_clip": 1.06115413,
      "balance_loss_mlp": 1.03791952,
      "epoch": 0.06216557446041003,
      "flos": 28110621686400.0,
      "grad_norm": 1.9739139529991114,
      "language_loss": 0.76928186,
      "learning_rate": 3.989195257821926e-06,
      "loss": 0.79209447,
      "num_input_tokens_seen": 11078440,
      "step": 517,
      "time_per_iteration": 2.7835304737091064
    },
    {
      "auxiliary_loss_clip": 0.01245917,
      "auxiliary_loss_mlp": 0.01049132,
      "balance_loss_clip": 1.07345212,
      "balance_loss_mlp": 1.03488648,
      "epoch": 0.06228581735104912,
      "flos": 23478181395840.0,
      "grad_norm": 2.4143862240817393,
      "language_loss": 0.84428084,
      "learning_rate": 3.989114245497765e-06,
      "loss": 0.86723131,
      "num_input_tokens_seen": 11098240,
      "step": 518,
      "time_per_iteration": 2.771448850631714
    },
    {
      "auxiliary_loss_clip": 0.0125904,
      "auxiliary_loss_mlp": 0.01043045,
      "balance_loss_clip": 1.06549549,
      "balance_loss_mlp": 1.02870452,
      "epoch": 0.06240606024168821,
      "flos": 15195205975680.0,
      "grad_norm": 2.257281658869347,
      "language_loss": 0.9506821,
      "learning_rate": 3.989032931426554e-06,
      "loss": 0.97370303,
      "num_input_tokens_seen": 11115395,
      "step": 519,
      "time_per_iteration": 2.6714537143707275
    },
    {
      "auxiliary_loss_clip": 0.01242529,
      "auxiliary_loss_mlp": 0.01044203,
      "balance_loss_clip": 1.06873524,
      "balance_loss_mlp": 1.02979088,
      "epoch": 0.06252630313232731,
      "flos": 20631829910400.0,
      "grad_norm": 2.535284038704183,
      "language_loss": 0.86963487,
      "learning_rate": 3.9889513156206295e-06,
      "loss": 0.89250207,
      "num_input_tokens_seen": 11134835,
      "step": 520,
      "time_per_iteration": 2.763765335083008
    },
    {
      "auxiliary_loss_clip": 0.01230604,
      "auxiliary_loss_mlp": 0.01053405,
      "balance_loss_clip": 1.06579816,
      "balance_loss_mlp": 1.03905785,
      "epoch": 0.06264654602296639,
      "flos": 20778058177920.0,
      "grad_norm": 2.8743371383942735,
      "language_loss": 0.73683721,
      "learning_rate": 3.988869398092371e-06,
      "loss": 0.75967729,
      "num_input_tokens_seen": 11154745,
      "step": 521,
      "time_per_iteration": 2.7117271423339844
    },
    {
      "auxiliary_loss_clip": 0.01244974,
      "auxiliary_loss_mlp": 0.01049164,
      "balance_loss_clip": 1.06826365,
      "balance_loss_mlp": 1.0340488,
      "epoch": 0.06276678891360549,
      "flos": 29605798241280.0,
      "grad_norm": 2.3423722175490744,
      "language_loss": 0.78474742,
      "learning_rate": 3.988787178854206e-06,
      "loss": 0.80768883,
      "num_input_tokens_seen": 11174280,
      "step": 522,
      "time_per_iteration": 2.7223896980285645
    },
    {
      "auxiliary_loss_clip": 0.01276394,
      "auxiliary_loss_mlp": 0.01048299,
      "balance_loss_clip": 1.07242155,
      "balance_loss_mlp": 1.03295708,
      "epoch": 0.06288703180424457,
      "flos": 22126288193280.0,
      "grad_norm": 2.322000712726739,
      "language_loss": 0.8777746,
      "learning_rate": 3.988704657918608e-06,
      "loss": 0.9010216,
      "num_input_tokens_seen": 11193340,
      "step": 523,
      "time_per_iteration": 2.669917106628418
    },
    {
      "auxiliary_loss_clip": 0.01257446,
      "auxiliary_loss_mlp": 0.01048018,
      "balance_loss_clip": 1.07014477,
      "balance_loss_mlp": 1.03426719,
      "epoch": 0.06300727469488367,
      "flos": 14976689587200.0,
      "grad_norm": 2.5131830499663623,
      "language_loss": 0.79819208,
      "learning_rate": 3.988621835298094e-06,
      "loss": 0.82124674,
      "num_input_tokens_seen": 11210555,
      "step": 524,
      "time_per_iteration": 2.6599583625793457
    },
    {
      "auxiliary_loss_clip": 0.0127327,
      "auxiliary_loss_mlp": 0.01039213,
      "balance_loss_clip": 1.07193589,
      "balance_loss_mlp": 1.02505636,
      "epoch": 0.06312751758552275,
      "flos": 24535391420160.0,
      "grad_norm": 2.06004887946143,
      "language_loss": 0.91888976,
      "learning_rate": 3.988538711005229e-06,
      "loss": 0.94201458,
      "num_input_tokens_seen": 11230010,
      "step": 525,
      "time_per_iteration": 2.729264259338379
    },
    {
      "auxiliary_loss_clip": 0.0125538,
      "auxiliary_loss_mlp": 0.01053775,
      "balance_loss_clip": 1.07118225,
      "balance_loss_mlp": 1.04026866,
      "epoch": 0.06324776047616185,
      "flos": 21507008785920.0,
      "grad_norm": 3.5750501163421586,
      "language_loss": 0.88528144,
      "learning_rate": 3.988455285052622e-06,
      "loss": 0.908373,
      "num_input_tokens_seen": 11246190,
      "step": 526,
      "time_per_iteration": 2.643002986907959
    },
    {
      "auxiliary_loss_clip": 0.01252495,
      "auxiliary_loss_mlp": 0.01049781,
      "balance_loss_clip": 1.06780243,
      "balance_loss_mlp": 1.03482044,
      "epoch": 0.06336800336680094,
      "flos": 21688034353920.0,
      "grad_norm": 2.428830019846273,
      "language_loss": 0.83990163,
      "learning_rate": 3.98837155745293e-06,
      "loss": 0.86292446,
      "num_input_tokens_seen": 11264230,
      "step": 527,
      "time_per_iteration": 2.7995424270629883
    },
    {
      "auxiliary_loss_clip": 0.01257398,
      "auxiliary_loss_mlp": 0.01039289,
      "balance_loss_clip": 1.07203686,
      "balance_loss_mlp": 1.0245434,
      "epoch": 0.06348824625744003,
      "flos": 19500895221120.0,
      "grad_norm": 3.4494860404568577,
      "language_loss": 0.75985074,
      "learning_rate": 3.988287528218854e-06,
      "loss": 0.7828176,
      "num_input_tokens_seen": 11283015,
      "step": 528,
      "time_per_iteration": 2.6696109771728516
    },
    {
      "auxiliary_loss_clip": 0.01256566,
      "auxiliary_loss_mlp": 0.01048261,
      "balance_loss_clip": 1.07279873,
      "balance_loss_mlp": 1.03397417,
      "epoch": 0.06360848914807912,
      "flos": 15481233976320.0,
      "grad_norm": 2.4149909919898866,
      "language_loss": 0.90539092,
      "learning_rate": 3.98820319736314e-06,
      "loss": 0.92843926,
      "num_input_tokens_seen": 11299630,
      "step": 529,
      "time_per_iteration": 2.6579387187957764
    },
    {
      "auxiliary_loss_clip": 0.01226875,
      "auxiliary_loss_mlp": 0.01051584,
      "balance_loss_clip": 1.0654254,
      "balance_loss_mlp": 1.03806591,
      "epoch": 0.0637287320387182,
      "flos": 20593369422720.0,
      "grad_norm": 1.8238754252008533,
      "language_loss": 0.85424072,
      "learning_rate": 3.988118564898582e-06,
      "loss": 0.87702525,
      "num_input_tokens_seen": 11319170,
      "step": 530,
      "time_per_iteration": 2.654066324234009
    },
    {
      "auxiliary_loss_clip": 0.01214974,
      "auxiliary_loss_mlp": 0.00767196,
      "balance_loss_clip": 1.06378567,
      "balance_loss_mlp": 1.00054157,
      "epoch": 0.0638489749293573,
      "flos": 17412222245760.0,
      "grad_norm": 2.9178790893381708,
      "language_loss": 0.89050734,
      "learning_rate": 3.988033630838019e-06,
      "loss": 0.9103291,
      "num_input_tokens_seen": 11333210,
      "step": 531,
      "time_per_iteration": 2.7737925052642822
    },
    {
      "auxiliary_loss_clip": 0.0125666,
      "auxiliary_loss_mlp": 0.0105158,
      "balance_loss_clip": 1.0688653,
      "balance_loss_mlp": 1.03716755,
      "epoch": 0.0639692178199964,
      "flos": 23807661874560.0,
      "grad_norm": 1.8265616736489187,
      "language_loss": 0.88043255,
      "learning_rate": 3.987948395194334e-06,
      "loss": 0.90351498,
      "num_input_tokens_seen": 11355590,
      "step": 532,
      "time_per_iteration": 2.766798257827759
    },
    {
      "auxiliary_loss_clip": 0.0124825,
      "auxiliary_loss_mlp": 0.01048904,
      "balance_loss_clip": 1.06594467,
      "balance_loss_mlp": 1.03421712,
      "epoch": 0.06408946071063548,
      "flos": 18477225521280.0,
      "grad_norm": 10.20076848838827,
      "language_loss": 0.76806599,
      "learning_rate": 3.987862857980458e-06,
      "loss": 0.79103756,
      "num_input_tokens_seen": 11371535,
      "step": 533,
      "time_per_iteration": 2.8007936477661133
    },
    {
      "auxiliary_loss_clip": 0.01230543,
      "auxiliary_loss_mlp": 0.01045137,
      "balance_loss_clip": 1.06867361,
      "balance_loss_mlp": 1.02974737,
      "epoch": 0.06420970360127458,
      "flos": 27162220936320.0,
      "grad_norm": 6.3713772372351825,
      "language_loss": 0.76764715,
      "learning_rate": 3.987777019209368e-06,
      "loss": 0.79040396,
      "num_input_tokens_seen": 11392050,
      "step": 534,
      "time_per_iteration": 2.83030366897583
    },
    {
      "auxiliary_loss_clip": 0.01271308,
      "auxiliary_loss_mlp": 0.01054732,
      "balance_loss_clip": 1.07112694,
      "balance_loss_mlp": 1.04096949,
      "epoch": 0.06432994649191366,
      "flos": 23659673840640.0,
      "grad_norm": 2.05288008398172,
      "language_loss": 0.81301761,
      "learning_rate": 3.987690878894084e-06,
      "loss": 0.83627808,
      "num_input_tokens_seen": 11411765,
      "step": 535,
      "time_per_iteration": 2.722294330596924
    },
    {
      "auxiliary_loss_clip": 0.01241391,
      "auxiliary_loss_mlp": 0.01051371,
      "balance_loss_clip": 1.06501627,
      "balance_loss_mlp": 1.03693461,
      "epoch": 0.06445018938255276,
      "flos": 23403953940480.0,
      "grad_norm": 2.447577008159149,
      "language_loss": 0.85060394,
      "learning_rate": 3.987604437047673e-06,
      "loss": 0.87353158,
      "num_input_tokens_seen": 11431565,
      "step": 536,
      "time_per_iteration": 2.749311685562134
    },
    {
      "auxiliary_loss_clip": 0.01255792,
      "auxiliary_loss_mlp": 0.01042214,
      "balance_loss_clip": 1.06940413,
      "balance_loss_mlp": 1.02815282,
      "epoch": 0.06457043227319184,
      "flos": 19646692525440.0,
      "grad_norm": 2.3287565981211746,
      "language_loss": 0.77363759,
      "learning_rate": 3.987517693683251e-06,
      "loss": 0.79661763,
      "num_input_tokens_seen": 11450140,
      "step": 537,
      "time_per_iteration": 3.509599447250366
    },
    {
      "auxiliary_loss_clip": 0.0123756,
      "auxiliary_loss_mlp": 0.01051341,
      "balance_loss_clip": 1.06958568,
      "balance_loss_mlp": 1.0375843,
      "epoch": 0.06469067516383094,
      "flos": 16978744915200.0,
      "grad_norm": 2.936730091552885,
      "language_loss": 0.96156335,
      "learning_rate": 3.9874306488139745e-06,
      "loss": 0.98445237,
      "num_input_tokens_seen": 11465400,
      "step": 538,
      "time_per_iteration": 2.7631750106811523
    },
    {
      "auxiliary_loss_clip": 0.0121615,
      "auxiliary_loss_mlp": 0.01043107,
      "balance_loss_clip": 1.06392276,
      "balance_loss_mlp": 1.02815771,
      "epoch": 0.06481091805447003,
      "flos": 23296401642240.0,
      "grad_norm": 2.0120076715757427,
      "language_loss": 0.87921613,
      "learning_rate": 3.987343302453049e-06,
      "loss": 0.90180874,
      "num_input_tokens_seen": 11486675,
      "step": 539,
      "time_per_iteration": 2.7747802734375
    },
    {
      "auxiliary_loss_clip": 0.01239816,
      "auxiliary_loss_mlp": 0.01042236,
      "balance_loss_clip": 1.06955314,
      "balance_loss_mlp": 1.02789545,
      "epoch": 0.06493116094510912,
      "flos": 29172356824320.0,
      "grad_norm": 1.6999519124180849,
      "language_loss": 0.82468295,
      "learning_rate": 3.987255654613724e-06,
      "loss": 0.84750342,
      "num_input_tokens_seen": 11510440,
      "step": 540,
      "time_per_iteration": 3.764968156814575
    },
    {
      "auxiliary_loss_clip": 0.01219453,
      "auxiliary_loss_mlp": 0.01047656,
      "balance_loss_clip": 1.0619483,
      "balance_loss_mlp": 1.03324389,
      "epoch": 0.06505140383574821,
      "flos": 19865065259520.0,
      "grad_norm": 4.00320770289931,
      "language_loss": 0.7058996,
      "learning_rate": 3.987167705309296e-06,
      "loss": 0.72857082,
      "num_input_tokens_seen": 11529715,
      "step": 541,
      "time_per_iteration": 2.823775291442871
    },
    {
      "auxiliary_loss_clip": 0.0125731,
      "auxiliary_loss_mlp": 0.00766216,
      "balance_loss_clip": 1.0697664,
      "balance_loss_mlp": 1.00043917,
      "epoch": 0.0651716467263873,
      "flos": 17924703540480.0,
      "grad_norm": 2.1050409643189774,
      "language_loss": 0.95358723,
      "learning_rate": 3.987079454553108e-06,
      "loss": 0.97382247,
      "num_input_tokens_seen": 11547665,
      "step": 542,
      "time_per_iteration": 3.5835654735565186
    },
    {
      "auxiliary_loss_clip": 0.0121912,
      "auxiliary_loss_mlp": 0.01045673,
      "balance_loss_clip": 1.06745076,
      "balance_loss_mlp": 1.03213048,
      "epoch": 0.0652918896170264,
      "flos": 20842840356480.0,
      "grad_norm": 1.9559651223684358,
      "language_loss": 0.91122413,
      "learning_rate": 3.986990902358546e-06,
      "loss": 0.93387204,
      "num_input_tokens_seen": 11564605,
      "step": 543,
      "time_per_iteration": 2.7830164432525635
    },
    {
      "auxiliary_loss_clip": 0.0125283,
      "auxiliary_loss_mlp": 0.01047175,
      "balance_loss_clip": 1.06628931,
      "balance_loss_mlp": 1.03380573,
      "epoch": 0.06541213250766549,
      "flos": 21872507627520.0,
      "grad_norm": 2.2198598916031447,
      "language_loss": 0.93593681,
      "learning_rate": 3.986902048739045e-06,
      "loss": 0.95893681,
      "num_input_tokens_seen": 11584550,
      "step": 544,
      "time_per_iteration": 2.7090060710906982
    },
    {
      "auxiliary_loss_clip": 0.01241851,
      "auxiliary_loss_mlp": 0.01052658,
      "balance_loss_clip": 1.06984937,
      "balance_loss_mlp": 1.03856778,
      "epoch": 0.06553237539830457,
      "flos": 23110743219840.0,
      "grad_norm": 6.239805815128727,
      "language_loss": 0.80086696,
      "learning_rate": 3.986812893708082e-06,
      "loss": 0.82381201,
      "num_input_tokens_seen": 11600740,
      "step": 545,
      "time_per_iteration": 2.6782820224761963
    },
    {
      "auxiliary_loss_clip": 0.01241062,
      "auxiliary_loss_mlp": 0.01046016,
      "balance_loss_clip": 1.06542981,
      "balance_loss_mlp": 1.03146052,
      "epoch": 0.06565261828894367,
      "flos": 17923769786880.0,
      "grad_norm": 2.2774721940902776,
      "language_loss": 0.81477523,
      "learning_rate": 3.9867234372791826e-06,
      "loss": 0.83764601,
      "num_input_tokens_seen": 11618695,
      "step": 546,
      "time_per_iteration": 2.653667688369751
    },
    {
      "auxiliary_loss_clip": 0.01249971,
      "auxiliary_loss_mlp": 0.01054942,
      "balance_loss_clip": 1.06697071,
      "balance_loss_mlp": 1.04181671,
      "epoch": 0.06577286117958275,
      "flos": 22783058421120.0,
      "grad_norm": 1.59376221803739,
      "language_loss": 0.87219083,
      "learning_rate": 3.986633679465918e-06,
      "loss": 0.89523995,
      "num_input_tokens_seen": 11638850,
      "step": 547,
      "time_per_iteration": 2.567714214324951
    },
    {
      "auxiliary_loss_clip": 0.01208839,
      "auxiliary_loss_mlp": 0.01051653,
      "balance_loss_clip": 1.06465578,
      "balance_loss_mlp": 1.03799164,
      "epoch": 0.06589310407022185,
      "flos": 23696194993920.0,
      "grad_norm": 2.4668450168390654,
      "language_loss": 0.80736482,
      "learning_rate": 3.986543620281904e-06,
      "loss": 0.8299697,
      "num_input_tokens_seen": 11658500,
      "step": 548,
      "time_per_iteration": 2.6978933811187744
    },
    {
      "auxiliary_loss_clip": 0.01223552,
      "auxiliary_loss_mlp": 0.01048668,
      "balance_loss_clip": 1.06436563,
      "balance_loss_mlp": 1.0342561,
      "epoch": 0.06601334696086093,
      "flos": 26864772410880.0,
      "grad_norm": 2.1551949584263013,
      "language_loss": 0.91550529,
      "learning_rate": 3.986453259740802e-06,
      "loss": 0.93822742,
      "num_input_tokens_seen": 11676670,
      "step": 549,
      "time_per_iteration": 2.6462385654449463
    },
    {
      "auxiliary_loss_clip": 0.01234891,
      "auxiliary_loss_mlp": 0.01047171,
      "balance_loss_clip": 1.06887388,
      "balance_loss_mlp": 1.03359878,
      "epoch": 0.06613358985150003,
      "flos": 12567694101120.0,
      "grad_norm": 2.754400949292195,
      "language_loss": 0.78949249,
      "learning_rate": 3.986362597856319e-06,
      "loss": 0.81231308,
      "num_input_tokens_seen": 11693170,
      "step": 550,
      "time_per_iteration": 2.6591908931732178
    },
    {
      "auxiliary_loss_clip": 0.01230178,
      "auxiliary_loss_mlp": 0.00767288,
      "balance_loss_clip": 1.06253052,
      "balance_loss_mlp": 1.00037336,
      "epoch": 0.06625383274213913,
      "flos": 18332505624960.0,
      "grad_norm": 2.659444106771651,
      "language_loss": 0.81692833,
      "learning_rate": 3.986271634642211e-06,
      "loss": 0.83690298,
      "num_input_tokens_seen": 11710150,
      "step": 551,
      "time_per_iteration": 2.6699206829071045
    },
    {
      "auxiliary_loss_clip": 0.01268998,
      "auxiliary_loss_mlp": 0.01040801,
      "balance_loss_clip": 1.07280934,
      "balance_loss_mlp": 1.02634048,
      "epoch": 0.06637407563277821,
      "flos": 15375585098880.0,
      "grad_norm": 1.9851056983928699,
      "language_loss": 0.81252748,
      "learning_rate": 3.986180370112274e-06,
      "loss": 0.83562547,
      "num_input_tokens_seen": 11726670,
      "step": 552,
      "time_per_iteration": 2.7054550647735596
    },
    {
      "auxiliary_loss_clip": 0.01259226,
      "auxiliary_loss_mlp": 0.00766912,
      "balance_loss_clip": 1.07126045,
      "balance_loss_mlp": 1.00041723,
      "epoch": 0.0664943185234173,
      "flos": 24025244509440.0,
      "grad_norm": 2.2758320065447264,
      "language_loss": 0.74467236,
      "learning_rate": 3.986088804280354e-06,
      "loss": 0.76493376,
      "num_input_tokens_seen": 11746400,
      "step": 553,
      "time_per_iteration": 2.719813108444214
    },
    {
      "auxiliary_loss_clip": 0.01236337,
      "auxiliary_loss_mlp": 0.01056458,
      "balance_loss_clip": 1.06397545,
      "balance_loss_mlp": 1.04248714,
      "epoch": 0.06661456141405639,
      "flos": 20957503547520.0,
      "grad_norm": 2.5990400507914204,
      "language_loss": 0.93894434,
      "learning_rate": 3.985996937160342e-06,
      "loss": 0.96187228,
      "num_input_tokens_seen": 11765590,
      "step": 554,
      "time_per_iteration": 2.6903107166290283
    },
    {
      "auxiliary_loss_clip": 0.01253625,
      "auxiliary_loss_mlp": 0.01050007,
      "balance_loss_clip": 1.07058775,
      "balance_loss_mlp": 1.03611946,
      "epoch": 0.06673480430469549,
      "flos": 52223953322880.0,
      "grad_norm": 2.14986177124641,
      "language_loss": 0.68706322,
      "learning_rate": 3.985904768766173e-06,
      "loss": 0.71009958,
      "num_input_tokens_seen": 11788365,
      "step": 555,
      "time_per_iteration": 2.8759732246398926
    },
    {
      "auxiliary_loss_clip": 0.01222389,
      "auxiliary_loss_mlp": 0.01046574,
      "balance_loss_clip": 1.0646708,
      "balance_loss_mlp": 1.03210235,
      "epoch": 0.06685504719533458,
      "flos": 16217079995520.0,
      "grad_norm": 3.657305334973761,
      "language_loss": 0.76252913,
      "learning_rate": 3.98581229911183e-06,
      "loss": 0.78521883,
      "num_input_tokens_seen": 11807285,
      "step": 556,
      "time_per_iteration": 2.747011184692383
    },
    {
      "auxiliary_loss_clip": 0.01249546,
      "auxiliary_loss_mlp": 0.01048557,
      "balance_loss_clip": 1.06304729,
      "balance_loss_mlp": 1.03464532,
      "epoch": 0.06697529008597367,
      "flos": 22491535639680.0,
      "grad_norm": 2.380812906226626,
      "language_loss": 0.92250198,
      "learning_rate": 3.985719528211341e-06,
      "loss": 0.94548297,
      "num_input_tokens_seen": 11826655,
      "step": 557,
      "time_per_iteration": 2.7364466190338135
    },
    {
      "auxiliary_loss_clip": 0.011654,
      "auxiliary_loss_mlp": 0.01007206,
      "balance_loss_clip": 1.05401134,
      "balance_loss_mlp": 1.0015316,
      "epoch": 0.06709553297661276,
      "flos": 62688216936960.0,
      "grad_norm": 0.8487815222614941,
      "language_loss": 0.63040733,
      "learning_rate": 3.985626456078777e-06,
      "loss": 0.65213341,
      "num_input_tokens_seen": 11891310,
      "step": 558,
      "time_per_iteration": 3.384471893310547
    },
    {
      "auxiliary_loss_clip": 0.01222713,
      "auxiliary_loss_mlp": 0.01047848,
      "balance_loss_clip": 1.0660733,
      "balance_loss_mlp": 1.03278005,
      "epoch": 0.06721577586725185,
      "flos": 11216590997760.0,
      "grad_norm": 2.7199048749086,
      "language_loss": 0.86275542,
      "learning_rate": 3.985533082728259e-06,
      "loss": 0.88546103,
      "num_input_tokens_seen": 11906965,
      "step": 559,
      "time_per_iteration": 2.749204397201538
    },
    {
      "auxiliary_loss_clip": 0.01272863,
      "auxiliary_loss_mlp": 0.01051082,
      "balance_loss_clip": 1.07133543,
      "balance_loss_mlp": 1.03695583,
      "epoch": 0.06733601875789094,
      "flos": 25922189664000.0,
      "grad_norm": 2.2750028965365403,
      "language_loss": 0.74877608,
      "learning_rate": 3.985439408173951e-06,
      "loss": 0.77201557,
      "num_input_tokens_seen": 11927190,
      "step": 560,
      "time_per_iteration": 2.6557657718658447
    },
    {
      "auxiliary_loss_clip": 0.01272439,
      "auxiliary_loss_mlp": 0.01049129,
      "balance_loss_clip": 1.07173944,
      "balance_loss_mlp": 1.03497291,
      "epoch": 0.06745626164853002,
      "flos": 20813645577600.0,
      "grad_norm": 1.9945421253271232,
      "language_loss": 0.71002114,
      "learning_rate": 3.9853454324300634e-06,
      "loss": 0.73323691,
      "num_input_tokens_seen": 11946400,
      "step": 561,
      "time_per_iteration": 2.7306649684906006
    },
    {
      "auxiliary_loss_clip": 0.01194991,
      "auxiliary_loss_mlp": 0.01044293,
      "balance_loss_clip": 1.05986178,
      "balance_loss_mlp": 1.03100133,
      "epoch": 0.06757650453916912,
      "flos": 19829262378240.0,
      "grad_norm": 2.011743113616089,
      "language_loss": 0.78157258,
      "learning_rate": 3.985251155510852e-06,
      "loss": 0.80396539,
      "num_input_tokens_seen": 11965430,
      "step": 562,
      "time_per_iteration": 2.837284564971924
    },
    {
      "auxiliary_loss_clip": 0.01207426,
      "auxiliary_loss_mlp": 0.01046646,
      "balance_loss_clip": 1.06768811,
      "balance_loss_mlp": 1.03248966,
      "epoch": 0.06769674742980822,
      "flos": 25739224761600.0,
      "grad_norm": 2.1370400284696727,
      "language_loss": 0.80088758,
      "learning_rate": 3.98515657743062e-06,
      "loss": 0.82342833,
      "num_input_tokens_seen": 11984895,
      "step": 563,
      "time_per_iteration": 3.6496822834014893
    },
    {
      "auxiliary_loss_clip": 0.01243193,
      "auxiliary_loss_mlp": 0.01046531,
      "balance_loss_clip": 1.06788337,
      "balance_loss_mlp": 1.03338194,
      "epoch": 0.0678169903204473,
      "flos": 13074788355840.0,
      "grad_norm": 3.0052785581367867,
      "language_loss": 0.77968907,
      "learning_rate": 3.985061698203711e-06,
      "loss": 0.80258638,
      "num_input_tokens_seen": 12002010,
      "step": 564,
      "time_per_iteration": 2.7362117767333984
    },
    {
      "auxiliary_loss_clip": 0.0118718,
      "auxiliary_loss_mlp": 0.01006799,
      "balance_loss_clip": 1.05417049,
      "balance_loss_mlp": 1.00141037,
      "epoch": 0.0679372332110864,
      "flos": 70865830788480.0,
      "grad_norm": 0.8955883533641473,
      "language_loss": 0.63854265,
      "learning_rate": 3.984966517844523e-06,
      "loss": 0.66048247,
      "num_input_tokens_seen": 12057255,
      "step": 565,
      "time_per_iteration": 3.1373822689056396
    },
    {
      "auxiliary_loss_clip": 0.01273452,
      "auxiliary_loss_mlp": 0.01048702,
      "balance_loss_clip": 1.07199895,
      "balance_loss_mlp": 1.03394413,
      "epoch": 0.06805747610172548,
      "flos": 28256418990720.0,
      "grad_norm": 2.2968524713936165,
      "language_loss": 0.80726737,
      "learning_rate": 3.984871036367492e-06,
      "loss": 0.83048898,
      "num_input_tokens_seen": 12077280,
      "step": 566,
      "time_per_iteration": 3.5436887741088867
    },
    {
      "auxiliary_loss_clip": 0.01257163,
      "auxiliary_loss_mlp": 0.0076734,
      "balance_loss_clip": 1.07057774,
      "balance_loss_mlp": 1.0002656,
      "epoch": 0.06817771899236458,
      "flos": 20120533764480.0,
      "grad_norm": 2.1935862722085675,
      "language_loss": 0.83063781,
      "learning_rate": 3.984775253787102e-06,
      "loss": 0.85088289,
      "num_input_tokens_seen": 12095570,
      "step": 567,
      "time_per_iteration": 3.795020818710327
    },
    {
      "auxiliary_loss_clip": 0.01250827,
      "auxiliary_loss_mlp": 0.01048465,
      "balance_loss_clip": 1.0637486,
      "balance_loss_mlp": 1.03501213,
      "epoch": 0.06829796188300366,
      "flos": 17930629284480.0,
      "grad_norm": 2.8327262826445816,
      "language_loss": 0.88006186,
      "learning_rate": 3.984679170117885e-06,
      "loss": 0.90305477,
      "num_input_tokens_seen": 12111775,
      "step": 568,
      "time_per_iteration": 3.6362342834472656
    },
    {
      "auxiliary_loss_clip": 0.01250128,
      "auxiliary_loss_mlp": 0.01043643,
      "balance_loss_clip": 1.06749773,
      "balance_loss_mlp": 1.02919483,
      "epoch": 0.06841820477364276,
      "flos": 14501627285760.0,
      "grad_norm": 3.2440940418556896,
      "language_loss": 0.78774881,
      "learning_rate": 3.984582785374415e-06,
      "loss": 0.81068653,
      "num_input_tokens_seen": 12129215,
      "step": 569,
      "time_per_iteration": 2.6611690521240234
    },
    {
      "auxiliary_loss_clip": 0.01238,
      "auxiliary_loss_mlp": 0.00766378,
      "balance_loss_clip": 1.06898904,
      "balance_loss_mlp": 1.00023174,
      "epoch": 0.06853844766428185,
      "flos": 21938474954880.0,
      "grad_norm": 30.441531417228404,
      "language_loss": 0.80686831,
      "learning_rate": 3.9844860995713155e-06,
      "loss": 0.82691216,
      "num_input_tokens_seen": 12148755,
      "step": 570,
      "time_per_iteration": 2.7592830657958984
    },
    {
      "auxiliary_loss_clip": 0.01257795,
      "auxiliary_loss_mlp": 0.01046896,
      "balance_loss_clip": 1.07308006,
      "balance_loss_mlp": 1.03371167,
      "epoch": 0.06865869055492094,
      "flos": 16800628348800.0,
      "grad_norm": 3.7910060663192895,
      "language_loss": 0.83214062,
      "learning_rate": 3.9843891127232524e-06,
      "loss": 0.85518754,
      "num_input_tokens_seen": 12166290,
      "step": 571,
      "time_per_iteration": 2.7251033782958984
    },
    {
      "auxiliary_loss_clip": 0.01199549,
      "auxiliary_loss_mlp": 0.01059317,
      "balance_loss_clip": 1.06237507,
      "balance_loss_mlp": 1.04658532,
      "epoch": 0.06877893344556003,
      "flos": 19937281553280.0,
      "grad_norm": 2.3592642325040116,
      "language_loss": 0.66511369,
      "learning_rate": 3.984291824844938e-06,
      "loss": 0.6877023,
      "num_input_tokens_seen": 12181385,
      "step": 572,
      "time_per_iteration": 2.9484503269195557
    },
    {
      "auxiliary_loss_clip": 0.01270819,
      "auxiliary_loss_mlp": 0.01043832,
      "balance_loss_clip": 1.07211494,
      "balance_loss_mlp": 1.02952135,
      "epoch": 0.06889917633619912,
      "flos": 23039388852480.0,
      "grad_norm": 4.7443234943812955,
      "language_loss": 0.8524062,
      "learning_rate": 3.984194235951132e-06,
      "loss": 0.87555271,
      "num_input_tokens_seen": 12197530,
      "step": 573,
      "time_per_iteration": 2.6607229709625244
    },
    {
      "auxiliary_loss_clip": 0.01271617,
      "auxiliary_loss_mlp": 0.01039962,
      "balance_loss_clip": 1.07430792,
      "balance_loss_mlp": 1.02653956,
      "epoch": 0.06901941922683821,
      "flos": 20960556203520.0,
      "grad_norm": 8.899396618097825,
      "language_loss": 0.84563845,
      "learning_rate": 3.9840963460566375e-06,
      "loss": 0.86875427,
      "num_input_tokens_seen": 12216310,
      "step": 574,
      "time_per_iteration": 2.711838960647583
    },
    {
      "auxiliary_loss_clip": 0.01174776,
      "auxiliary_loss_mlp": 0.0104771,
      "balance_loss_clip": 1.06174684,
      "balance_loss_mlp": 1.03395319,
      "epoch": 0.06913966211747731,
      "flos": 24821850384000.0,
      "grad_norm": 1.5805904695135182,
      "language_loss": 0.89651769,
      "learning_rate": 3.983998155176305e-06,
      "loss": 0.91874254,
      "num_input_tokens_seen": 12236670,
      "step": 575,
      "time_per_iteration": 2.8479719161987305
    },
    {
      "auxiliary_loss_clip": 0.01180911,
      "auxiliary_loss_mlp": 0.01007486,
      "balance_loss_clip": 1.04950523,
      "balance_loss_mlp": 1.00235963,
      "epoch": 0.06925990500811639,
      "flos": 58367446957440.0,
      "grad_norm": 0.8196954944579226,
      "language_loss": 0.57096517,
      "learning_rate": 3.9838996633250305e-06,
      "loss": 0.59284914,
      "num_input_tokens_seen": 12297185,
      "step": 576,
      "time_per_iteration": 3.144300699234009
    },
    {
      "auxiliary_loss_clip": 0.01252735,
      "auxiliary_loss_mlp": 0.01046969,
      "balance_loss_clip": 1.06681657,
      "balance_loss_mlp": 1.0339154,
      "epoch": 0.06938014789875549,
      "flos": 12749940731520.0,
      "grad_norm": 2.5485545576455357,
      "language_loss": 0.88342685,
      "learning_rate": 3.983800870517753e-06,
      "loss": 0.90642381,
      "num_input_tokens_seen": 12313975,
      "step": 577,
      "time_per_iteration": 2.702695846557617
    },
    {
      "auxiliary_loss_clip": 0.01251556,
      "auxiliary_loss_mlp": 0.01048593,
      "balance_loss_clip": 1.07170403,
      "balance_loss_mlp": 1.03561711,
      "epoch": 0.06950039078939457,
      "flos": 22820226019200.0,
      "grad_norm": 9.549613743961567,
      "language_loss": 0.79185432,
      "learning_rate": 3.983701776769463e-06,
      "loss": 0.81485581,
      "num_input_tokens_seen": 12331385,
      "step": 578,
      "time_per_iteration": 2.7667999267578125
    },
    {
      "auxiliary_loss_clip": 0.01242359,
      "auxiliary_loss_mlp": 0.01047474,
      "balance_loss_clip": 1.06755733,
      "balance_loss_mlp": 1.03471851,
      "epoch": 0.06962063368003367,
      "flos": 21941348042880.0,
      "grad_norm": 2.2733352728811007,
      "language_loss": 0.85532874,
      "learning_rate": 3.9836023820951885e-06,
      "loss": 0.87822711,
      "num_input_tokens_seen": 12350600,
      "step": 579,
      "time_per_iteration": 2.6785950660705566
    },
    {
      "auxiliary_loss_clip": 0.0121925,
      "auxiliary_loss_mlp": 0.01043829,
      "balance_loss_clip": 1.06424427,
      "balance_loss_mlp": 1.03070974,
      "epoch": 0.06974087657067275,
      "flos": 20706021452160.0,
      "grad_norm": 2.2620166790646943,
      "language_loss": 0.68504947,
      "learning_rate": 3.983502686510011e-06,
      "loss": 0.70768028,
      "num_input_tokens_seen": 12371430,
      "step": 580,
      "time_per_iteration": 2.7175381183624268
    },
    {
      "auxiliary_loss_clip": 0.01252935,
      "auxiliary_loss_mlp": 0.00766217,
      "balance_loss_clip": 1.06700361,
      "balance_loss_mlp": 1.00024271,
      "epoch": 0.06986111946131185,
      "flos": 22638230784000.0,
      "grad_norm": 1.7979739131655443,
      "language_loss": 0.7366519,
      "learning_rate": 3.9834026900290525e-06,
      "loss": 0.75684345,
      "num_input_tokens_seen": 12390825,
      "step": 581,
      "time_per_iteration": 2.679097890853882
    },
    {
      "auxiliary_loss_clip": 0.01267995,
      "auxiliary_loss_mlp": 0.01046965,
      "balance_loss_clip": 1.07138169,
      "balance_loss_mlp": 1.03474557,
      "epoch": 0.06998136235195095,
      "flos": 26943453152640.0,
      "grad_norm": 2.5813490428616768,
      "language_loss": 1.00304878,
      "learning_rate": 3.983302392667482e-06,
      "loss": 1.02619839,
      "num_input_tokens_seen": 12411670,
      "step": 582,
      "time_per_iteration": 2.6406190395355225
    },
    {
      "auxiliary_loss_clip": 0.01254476,
      "auxiliary_loss_mlp": 0.01047893,
      "balance_loss_clip": 1.07412779,
      "balance_loss_mlp": 1.03493476,
      "epoch": 0.07010160524259003,
      "flos": 22492505306880.0,
      "grad_norm": 1.77888315778107,
      "language_loss": 0.93458587,
      "learning_rate": 3.983201794440517e-06,
      "loss": 0.95760953,
      "num_input_tokens_seen": 12431245,
      "step": 583,
      "time_per_iteration": 2.694089412689209
    },
    {
      "auxiliary_loss_clip": 0.01221923,
      "auxiliary_loss_mlp": 0.01046933,
      "balance_loss_clip": 1.06499863,
      "balance_loss_mlp": 1.03376055,
      "epoch": 0.07022184813322913,
      "flos": 18332541538560.0,
      "grad_norm": 1.8831338267792685,
      "language_loss": 0.67743397,
      "learning_rate": 3.9831008953634165e-06,
      "loss": 0.70012254,
      "num_input_tokens_seen": 12450535,
      "step": 584,
      "time_per_iteration": 2.699342727661133
    },
    {
      "auxiliary_loss_clip": 0.01184686,
      "auxiliary_loss_mlp": 0.01053351,
      "balance_loss_clip": 1.05762947,
      "balance_loss_mlp": 1.03982711,
      "epoch": 0.07034209102386821,
      "flos": 24675550289280.0,
      "grad_norm": 3.5120915460034565,
      "language_loss": 0.81365317,
      "learning_rate": 3.9829996954514864e-06,
      "loss": 0.83603358,
      "num_input_tokens_seen": 12469675,
      "step": 585,
      "time_per_iteration": 2.8128342628479004
    },
    {
      "auxiliary_loss_clip": 0.01236455,
      "auxiliary_loss_mlp": 0.0104549,
      "balance_loss_clip": 1.06577754,
      "balance_loss_mlp": 1.03222191,
      "epoch": 0.0704623339145073,
      "flos": 25995878415360.0,
      "grad_norm": 2.0319309724450023,
      "language_loss": 0.84288979,
      "learning_rate": 3.982898194720079e-06,
      "loss": 0.8657093,
      "num_input_tokens_seen": 12490405,
      "step": 586,
      "time_per_iteration": 2.7475976943969727
    },
    {
      "auxiliary_loss_clip": 0.01230416,
      "auxiliary_loss_mlp": 0.00766669,
      "balance_loss_clip": 1.06834865,
      "balance_loss_mlp": 1.00027406,
      "epoch": 0.0705825768051464,
      "flos": 25338318088320.0,
      "grad_norm": 2.1926900433233367,
      "language_loss": 0.82552886,
      "learning_rate": 3.982796393184592e-06,
      "loss": 0.84549975,
      "num_input_tokens_seen": 12509485,
      "step": 587,
      "time_per_iteration": 2.812030076980591
    },
    {
      "auxiliary_loss_clip": 0.01157836,
      "auxiliary_loss_mlp": 0.01010665,
      "balance_loss_clip": 1.04122329,
      "balance_loss_mlp": 1.00534868,
      "epoch": 0.07070281969578548,
      "flos": 66047552507520.0,
      "grad_norm": 0.8190602304442607,
      "language_loss": 0.62651038,
      "learning_rate": 3.98269429086047e-06,
      "loss": 0.64819539,
      "num_input_tokens_seen": 12567325,
      "step": 588,
      "time_per_iteration": 3.1467673778533936
    },
    {
      "auxiliary_loss_clip": 0.01231877,
      "auxiliary_loss_mlp": 0.01043038,
      "balance_loss_clip": 1.06761074,
      "balance_loss_mlp": 1.02875113,
      "epoch": 0.07082306258642458,
      "flos": 23653568528640.0,
      "grad_norm": 3.48101706130193,
      "language_loss": 0.86017406,
      "learning_rate": 3.982591887763199e-06,
      "loss": 0.88292325,
      "num_input_tokens_seen": 12584785,
      "step": 589,
      "time_per_iteration": 3.6911938190460205
    },
    {
      "auxiliary_loss_clip": 0.01191512,
      "auxiliary_loss_mlp": 0.01045387,
      "balance_loss_clip": 1.05399489,
      "balance_loss_mlp": 1.0323931,
      "epoch": 0.07094330547706366,
      "flos": 13880049408000.0,
      "grad_norm": 7.329092078268937,
      "language_loss": 0.81627172,
      "learning_rate": 3.982489183908316e-06,
      "loss": 0.83864069,
      "num_input_tokens_seen": 12601205,
      "step": 590,
      "time_per_iteration": 2.7544798851013184
    },
    {
      "auxiliary_loss_clip": 0.01160077,
      "auxiliary_loss_mlp": 0.010471,
      "balance_loss_clip": 1.05072844,
      "balance_loss_mlp": 1.03422546,
      "epoch": 0.07106354836770276,
      "flos": 24645098534400.0,
      "grad_norm": 1.9239549997030303,
      "language_loss": 0.84582204,
      "learning_rate": 3.982386179311399e-06,
      "loss": 0.86789382,
      "num_input_tokens_seen": 12621725,
      "step": 591,
      "time_per_iteration": 2.8666436672210693
    },
    {
      "auxiliary_loss_clip": 0.01255452,
      "auxiliary_loss_mlp": 0.01041887,
      "balance_loss_clip": 1.07002628,
      "balance_loss_mlp": 1.02780867,
      "epoch": 0.07118379125834184,
      "flos": 16217223649920.0,
      "grad_norm": 2.458339054808149,
      "language_loss": 0.87708449,
      "learning_rate": 3.982282873988075e-06,
      "loss": 0.90005791,
      "num_input_tokens_seen": 12639600,
      "step": 592,
      "time_per_iteration": 3.6496973037719727
    },
    {
      "auxiliary_loss_clip": 0.01227703,
      "auxiliary_loss_mlp": 0.01044866,
      "balance_loss_clip": 1.06587934,
      "balance_loss_mlp": 1.03203344,
      "epoch": 0.07130403414898094,
      "flos": 19719986227200.0,
      "grad_norm": 1.6552514988420344,
      "language_loss": 0.87064189,
      "learning_rate": 3.982179267954016e-06,
      "loss": 0.89336753,
      "num_input_tokens_seen": 12660030,
      "step": 593,
      "time_per_iteration": 3.7004263401031494
    },
    {
      "auxiliary_loss_clip": 0.01264504,
      "auxiliary_loss_mlp": 0.01038995,
      "balance_loss_clip": 1.0689106,
      "balance_loss_mlp": 1.02568567,
      "epoch": 0.07142427703962004,
      "flos": 21871933009920.0,
      "grad_norm": 2.209868235734169,
      "language_loss": 0.96041369,
      "learning_rate": 3.982075361224937e-06,
      "loss": 0.98344874,
      "num_input_tokens_seen": 12678395,
      "step": 594,
      "time_per_iteration": 3.6022818088531494
    },
    {
      "auxiliary_loss_clip": 0.01247762,
      "auxiliary_loss_mlp": 0.00766417,
      "balance_loss_clip": 1.06928706,
      "balance_loss_mlp": 1.00025761,
      "epoch": 0.07154451993025912,
      "flos": 18296595002880.0,
      "grad_norm": 1.8683091469593507,
      "language_loss": 0.88101614,
      "learning_rate": 3.981971153816602e-06,
      "loss": 0.90115798,
      "num_input_tokens_seen": 12696000,
      "step": 595,
      "time_per_iteration": 2.675199508666992
    },
    {
      "auxiliary_loss_clip": 0.01264722,
      "auxiliary_loss_mlp": 0.01041331,
      "balance_loss_clip": 1.07266831,
      "balance_loss_mlp": 1.0293808,
      "epoch": 0.07166476282089822,
      "flos": 22160690444160.0,
      "grad_norm": 1.629961884715407,
      "language_loss": 0.96228844,
      "learning_rate": 3.981866645744819e-06,
      "loss": 0.98534894,
      "num_input_tokens_seen": 12716715,
      "step": 596,
      "time_per_iteration": 2.757694959640503
    },
    {
      "auxiliary_loss_clip": 0.01270674,
      "auxiliary_loss_mlp": 0.00766009,
      "balance_loss_clip": 1.07304859,
      "balance_loss_mlp": 1.00021815,
      "epoch": 0.0717850057115373,
      "flos": 14136343925760.0,
      "grad_norm": 3.0376376238313934,
      "language_loss": 0.8172909,
      "learning_rate": 3.9817618370254416e-06,
      "loss": 0.83765775,
      "num_input_tokens_seen": 12733370,
      "step": 597,
      "time_per_iteration": 2.6664061546325684
    },
    {
      "auxiliary_loss_clip": 0.01269022,
      "auxiliary_loss_mlp": 0.0104643,
      "balance_loss_clip": 1.07080579,
      "balance_loss_mlp": 1.03358519,
      "epoch": 0.0719052486021764,
      "flos": 30917794412160.0,
      "grad_norm": 2.434008856111168,
      "language_loss": 0.87579149,
      "learning_rate": 3.9816567276743684e-06,
      "loss": 0.89894605,
      "num_input_tokens_seen": 12753235,
      "step": 598,
      "time_per_iteration": 2.774862289428711
    },
    {
      "auxiliary_loss_clip": 0.01227757,
      "auxiliary_loss_mlp": 0.0104794,
      "balance_loss_clip": 1.06666911,
      "balance_loss_mlp": 1.03462386,
      "epoch": 0.0720254914928155,
      "flos": 21287019939840.0,
      "grad_norm": 2.056732012514836,
      "language_loss": 0.77598363,
      "learning_rate": 3.9815513177075466e-06,
      "loss": 0.79874063,
      "num_input_tokens_seen": 12772020,
      "step": 599,
      "time_per_iteration": 2.7556369304656982
    },
    {
      "auxiliary_loss_clip": 0.01244839,
      "auxiliary_loss_mlp": 0.01051119,
      "balance_loss_clip": 1.06821334,
      "balance_loss_mlp": 1.03895974,
      "epoch": 0.07214573438345458,
      "flos": 27819170732160.0,
      "grad_norm": 1.7020740694118823,
      "language_loss": 0.70175874,
      "learning_rate": 3.9814456071409646e-06,
      "loss": 0.72471833,
      "num_input_tokens_seen": 12792555,
      "step": 600,
      "time_per_iteration": 2.7655434608459473
    },
    {
      "auxiliary_loss_clip": 0.01209077,
      "auxiliary_loss_mlp": 0.01046935,
      "balance_loss_clip": 1.06405997,
      "balance_loss_mlp": 1.03300488,
      "epoch": 0.07226597727409367,
      "flos": 25483576688640.0,
      "grad_norm": 3.6105507838394804,
      "language_loss": 0.85834688,
      "learning_rate": 3.981339595990659e-06,
      "loss": 0.880907,
      "num_input_tokens_seen": 12811085,
      "step": 601,
      "time_per_iteration": 2.8356716632843018
    },
    {
      "auxiliary_loss_clip": 0.01252855,
      "auxiliary_loss_mlp": 0.01041463,
      "balance_loss_clip": 1.07107198,
      "balance_loss_mlp": 1.02890992,
      "epoch": 0.07238622016473276,
      "flos": 23513840622720.0,
      "grad_norm": 2.0299282391403906,
      "language_loss": 0.81401658,
      "learning_rate": 3.981233284272713e-06,
      "loss": 0.83695978,
      "num_input_tokens_seen": 12830830,
      "step": 602,
      "time_per_iteration": 2.6941051483154297
    },
    {
      "auxiliary_loss_clip": 0.0121604,
      "auxiliary_loss_mlp": 0.01052257,
      "balance_loss_clip": 1.06212783,
      "balance_loss_mlp": 1.03900087,
      "epoch": 0.07250646305537185,
      "flos": 25453519983360.0,
      "grad_norm": 1.8075552276703384,
      "language_loss": 0.90148711,
      "learning_rate": 3.981126672003253e-06,
      "loss": 0.92417008,
      "num_input_tokens_seen": 12853505,
      "step": 603,
      "time_per_iteration": 2.889545202255249
    },
    {
      "auxiliary_loss_clip": 0.01237688,
      "auxiliary_loss_mlp": 0.01043326,
      "balance_loss_clip": 1.06099033,
      "balance_loss_mlp": 1.03092861,
      "epoch": 0.07262670594601094,
      "flos": 27155038216320.0,
      "grad_norm": 2.6393995751501595,
      "language_loss": 0.78409088,
      "learning_rate": 3.981019759198451e-06,
      "loss": 0.80690104,
      "num_input_tokens_seen": 12872455,
      "step": 604,
      "time_per_iteration": 2.789447069168091
    },
    {
      "auxiliary_loss_clip": 0.01232467,
      "auxiliary_loss_mlp": 0.01042948,
      "balance_loss_clip": 1.06223249,
      "balance_loss_mlp": 1.0302527,
      "epoch": 0.07274694883665003,
      "flos": 26651607148800.0,
      "grad_norm": 2.4519764470448218,
      "language_loss": 0.84386587,
      "learning_rate": 3.980912545874528e-06,
      "loss": 0.86662006,
      "num_input_tokens_seen": 12892620,
      "step": 605,
      "time_per_iteration": 2.774458885192871
    },
    {
      "auxiliary_loss_clip": 0.01243484,
      "auxiliary_loss_mlp": 0.00766968,
      "balance_loss_clip": 1.06741714,
      "balance_loss_mlp": 1.00031376,
      "epoch": 0.07286719172728913,
      "flos": 29862344154240.0,
      "grad_norm": 2.481239914483947,
      "language_loss": 0.85620475,
      "learning_rate": 3.980805032047746e-06,
      "loss": 0.87630928,
      "num_input_tokens_seen": 12914090,
      "step": 606,
      "time_per_iteration": 2.7429800033569336
    },
    {
      "auxiliary_loss_clip": 0.01234129,
      "auxiliary_loss_mlp": 0.01042824,
      "balance_loss_clip": 1.06574273,
      "balance_loss_mlp": 1.02864969,
      "epoch": 0.07298743461792821,
      "flos": 17382057799680.0,
      "grad_norm": 1.8972346640635336,
      "language_loss": 0.81001484,
      "learning_rate": 3.980697217734415e-06,
      "loss": 0.83278441,
      "num_input_tokens_seen": 12931830,
      "step": 607,
      "time_per_iteration": 2.731691837310791
    },
    {
      "auxiliary_loss_clip": 0.01205776,
      "auxiliary_loss_mlp": 0.00767112,
      "balance_loss_clip": 1.06333268,
      "balance_loss_mlp": 1.00031447,
      "epoch": 0.07310767750856731,
      "flos": 19498201701120.0,
      "grad_norm": 1.8378206354242137,
      "language_loss": 0.91722327,
      "learning_rate": 3.980589102950891e-06,
      "loss": 0.93695217,
      "num_input_tokens_seen": 12949995,
      "step": 608,
      "time_per_iteration": 2.836153030395508
    },
    {
      "auxiliary_loss_clip": 0.01233678,
      "auxiliary_loss_mlp": 0.01054547,
      "balance_loss_clip": 1.06867993,
      "balance_loss_mlp": 1.04028976,
      "epoch": 0.07322792039920639,
      "flos": 29168693637120.0,
      "grad_norm": 2.531920049188976,
      "language_loss": 0.7605654,
      "learning_rate": 3.9804806877135755e-06,
      "loss": 0.78344762,
      "num_input_tokens_seen": 12968040,
      "step": 609,
      "time_per_iteration": 2.8863418102264404
    },
    {
      "auxiliary_loss_clip": 0.01249549,
      "auxiliary_loss_mlp": 0.00766881,
      "balance_loss_clip": 1.06228352,
      "balance_loss_mlp": 1.00036931,
      "epoch": 0.07334816328984549,
      "flos": 23477822259840.0,
      "grad_norm": 5.25238257178374,
      "language_loss": 0.86292863,
      "learning_rate": 3.980371972038915e-06,
      "loss": 0.88309294,
      "num_input_tokens_seen": 12988530,
      "step": 610,
      "time_per_iteration": 2.688387155532837
    },
    {
      "auxiliary_loss_clip": 0.01270096,
      "auxiliary_loss_mlp": 0.0105437,
      "balance_loss_clip": 1.07211399,
      "balance_loss_mlp": 1.041049,
      "epoch": 0.07346840618048459,
      "flos": 22962467877120.0,
      "grad_norm": 1.7129578481044332,
      "language_loss": 0.84240663,
      "learning_rate": 3.980262955943399e-06,
      "loss": 0.86565137,
      "num_input_tokens_seen": 13008195,
      "step": 611,
      "time_per_iteration": 2.7256715297698975
    },
    {
      "auxiliary_loss_clip": 0.01228872,
      "auxiliary_loss_mlp": 0.01047901,
      "balance_loss_clip": 1.06818843,
      "balance_loss_mlp": 1.03454971,
      "epoch": 0.07358864907112367,
      "flos": 17673903803520.0,
      "grad_norm": 2.4424740787434844,
      "language_loss": 0.86838496,
      "learning_rate": 3.980153639443569e-06,
      "loss": 0.89115262,
      "num_input_tokens_seen": 13024180,
      "step": 612,
      "time_per_iteration": 2.695604085922241
    },
    {
      "auxiliary_loss_clip": 0.01242194,
      "auxiliary_loss_mlp": 0.01050704,
      "balance_loss_clip": 1.06687391,
      "balance_loss_mlp": 1.03685212,
      "epoch": 0.07370889196176277,
      "flos": 24097029840000.0,
      "grad_norm": 2.921389844023767,
      "language_loss": 0.80493581,
      "learning_rate": 3.980044022556005e-06,
      "loss": 0.82786471,
      "num_input_tokens_seen": 13043865,
      "step": 613,
      "time_per_iteration": 2.6714742183685303
    },
    {
      "auxiliary_loss_clip": 0.01252289,
      "auxiliary_loss_mlp": 0.0104948,
      "balance_loss_clip": 1.07033384,
      "balance_loss_mlp": 1.03592598,
      "epoch": 0.07382913485240185,
      "flos": 25885919905920.0,
      "grad_norm": 2.3485167481016016,
      "language_loss": 0.73033941,
      "learning_rate": 3.9799341052973375e-06,
      "loss": 0.75335717,
      "num_input_tokens_seen": 13063700,
      "step": 614,
      "time_per_iteration": 2.7138421535491943
    },
    {
      "auxiliary_loss_clip": 0.01235436,
      "auxiliary_loss_mlp": 0.01044639,
      "balance_loss_clip": 1.06867981,
      "balance_loss_mlp": 1.03130555,
      "epoch": 0.07394937774304094,
      "flos": 16873850223360.0,
      "grad_norm": 2.680992258556214,
      "language_loss": 0.75166273,
      "learning_rate": 3.979823887684241e-06,
      "loss": 0.77446347,
      "num_input_tokens_seen": 13082640,
      "step": 615,
      "time_per_iteration": 3.618635416030884
    },
    {
      "auxiliary_loss_clip": 0.01266576,
      "auxiliary_loss_mlp": 0.01045982,
      "balance_loss_clip": 1.07110822,
      "balance_loss_mlp": 1.03242159,
      "epoch": 0.07406962063368003,
      "flos": 20703471586560.0,
      "grad_norm": 2.633816256754086,
      "language_loss": 0.85084724,
      "learning_rate": 3.979713369733434e-06,
      "loss": 0.87397283,
      "num_input_tokens_seen": 13100505,
      "step": 616,
      "time_per_iteration": 2.662513494491577
    },
    {
      "auxiliary_loss_clip": 0.012454,
      "auxiliary_loss_mlp": 0.01047147,
      "balance_loss_clip": 1.07087803,
      "balance_loss_mlp": 1.03309226,
      "epoch": 0.07418986352431912,
      "flos": 21430985650560.0,
      "grad_norm": 2.7952238589016423,
      "language_loss": 0.84869671,
      "learning_rate": 3.979602551461683e-06,
      "loss": 0.8716222,
      "num_input_tokens_seen": 13121285,
      "step": 617,
      "time_per_iteration": 2.720712184906006
    },
    {
      "auxiliary_loss_clip": 0.01235157,
      "auxiliary_loss_mlp": 0.01048245,
      "balance_loss_clip": 1.06905508,
      "balance_loss_mlp": 1.03458369,
      "epoch": 0.07431010641495822,
      "flos": 12021133777920.0,
      "grad_norm": 2.435347664266673,
      "language_loss": 0.91493261,
      "learning_rate": 3.979491432885799e-06,
      "loss": 0.93776667,
      "num_input_tokens_seen": 13137550,
      "step": 618,
      "time_per_iteration": 3.547816514968872
    },
    {
      "auxiliary_loss_clip": 0.01199864,
      "auxiliary_loss_mlp": 0.00766417,
      "balance_loss_clip": 1.05987644,
      "balance_loss_mlp": 1.0004369,
      "epoch": 0.0744303493055973,
      "flos": 20957575374720.0,
      "grad_norm": 2.067191568152436,
      "language_loss": 0.82908398,
      "learning_rate": 3.97938001402264e-06,
      "loss": 0.84874678,
      "num_input_tokens_seen": 13156675,
      "step": 619,
      "time_per_iteration": 3.6499555110931396
    },
    {
      "auxiliary_loss_clip": 0.01211546,
      "auxiliary_loss_mlp": 0.01054885,
      "balance_loss_clip": 1.06240642,
      "balance_loss_mlp": 1.04005504,
      "epoch": 0.0745505921962364,
      "flos": 16253134272000.0,
      "grad_norm": 3.0093697918777047,
      "language_loss": 0.79752105,
      "learning_rate": 3.979268294889105e-06,
      "loss": 0.82018542,
      "num_input_tokens_seen": 13172225,
      "step": 620,
      "time_per_iteration": 3.6573145389556885
    },
    {
      "auxiliary_loss_clip": 0.01266458,
      "auxiliary_loss_mlp": 0.01041317,
      "balance_loss_clip": 1.07057202,
      "balance_loss_mlp": 1.02773285,
      "epoch": 0.07467083508687548,
      "flos": 50944635550080.0,
      "grad_norm": 2.0206801149425124,
      "language_loss": 0.74122834,
      "learning_rate": 3.979156275502143e-06,
      "loss": 0.76430607,
      "num_input_tokens_seen": 13195885,
      "step": 621,
      "time_per_iteration": 2.954556941986084
    },
    {
      "auxiliary_loss_clip": 0.01215603,
      "auxiliary_loss_mlp": 0.01041908,
      "balance_loss_clip": 1.06341398,
      "balance_loss_mlp": 1.02816343,
      "epoch": 0.07479107797751458,
      "flos": 17529686697600.0,
      "grad_norm": 3.8145206079769207,
      "language_loss": 0.91734397,
      "learning_rate": 3.979043955878749e-06,
      "loss": 0.93991911,
      "num_input_tokens_seen": 13213730,
      "step": 622,
      "time_per_iteration": 2.7942793369293213
    },
    {
      "auxiliary_loss_clip": 0.01230689,
      "auxiliary_loss_mlp": 0.01050822,
      "balance_loss_clip": 1.06696057,
      "balance_loss_mlp": 1.03687501,
      "epoch": 0.07491132086815366,
      "flos": 23473943591040.0,
      "grad_norm": 2.2196476407419707,
      "language_loss": 0.8316986,
      "learning_rate": 3.978931336035959e-06,
      "loss": 0.85451376,
      "num_input_tokens_seen": 13232540,
      "step": 623,
      "time_per_iteration": 2.7696945667266846
    },
    {
      "auxiliary_loss_clip": 0.0125169,
      "auxiliary_loss_mlp": 0.01048543,
      "balance_loss_clip": 1.07109416,
      "balance_loss_mlp": 1.03614533,
      "epoch": 0.07503156375879276,
      "flos": 20157557708160.0,
      "grad_norm": 2.363437918970197,
      "language_loss": 0.82621062,
      "learning_rate": 3.9788184159908595e-06,
      "loss": 0.84921288,
      "num_input_tokens_seen": 13249670,
      "step": 624,
      "time_per_iteration": 2.696173906326294
    },
    {
      "auxiliary_loss_clip": 0.01222074,
      "auxiliary_loss_mlp": 0.01045803,
      "balance_loss_clip": 1.06092668,
      "balance_loss_mlp": 1.03303576,
      "epoch": 0.07515180664943186,
      "flos": 15115519653120.0,
      "grad_norm": 6.995444982549287,
      "language_loss": 0.82714766,
      "learning_rate": 3.97870519576058e-06,
      "loss": 0.84982646,
      "num_input_tokens_seen": 13266095,
      "step": 625,
      "time_per_iteration": 2.6936166286468506
    },
    {
      "auxiliary_loss_clip": 0.01216905,
      "auxiliary_loss_mlp": 0.00766601,
      "balance_loss_clip": 1.06291795,
      "balance_loss_mlp": 1.00058293,
      "epoch": 0.07527204954007094,
      "flos": 21287702298240.0,
      "grad_norm": 2.5252897124888647,
      "language_loss": 0.80981338,
      "learning_rate": 3.978591675362295e-06,
      "loss": 0.82964844,
      "num_input_tokens_seen": 13284810,
      "step": 626,
      "time_per_iteration": 2.8920786380767822
    },
    {
      "auxiliary_loss_clip": 0.01204825,
      "auxiliary_loss_mlp": 0.01045569,
      "balance_loss_clip": 1.06814778,
      "balance_loss_mlp": 1.03229499,
      "epoch": 0.07539229243071004,
      "flos": 21324187537920.0,
      "grad_norm": 1.7718974068641749,
      "language_loss": 0.87736368,
      "learning_rate": 3.978477854813226e-06,
      "loss": 0.89986765,
      "num_input_tokens_seen": 13304150,
      "step": 627,
      "time_per_iteration": 2.8174307346343994
    },
    {
      "auxiliary_loss_clip": 0.01249195,
      "auxiliary_loss_mlp": 0.01042275,
      "balance_loss_clip": 1.06741107,
      "balance_loss_mlp": 1.02927566,
      "epoch": 0.07551253532134912,
      "flos": 13042540920960.0,
      "grad_norm": 1.9194130712628374,
      "language_loss": 0.8265726,
      "learning_rate": 3.97836373413064e-06,
      "loss": 0.8494873,
      "num_input_tokens_seen": 13322205,
      "step": 628,
      "time_per_iteration": 2.634899139404297
    },
    {
      "auxiliary_loss_clip": 0.0126112,
      "auxiliary_loss_mlp": 0.01050038,
      "balance_loss_clip": 1.06530023,
      "balance_loss_mlp": 1.03659666,
      "epoch": 0.07563277821198822,
      "flos": 19208761908480.0,
      "grad_norm": 2.0542968144716673,
      "language_loss": 0.7451176,
      "learning_rate": 3.978249313331848e-06,
      "loss": 0.76822925,
      "num_input_tokens_seen": 13340435,
      "step": 629,
      "time_per_iteration": 2.6369986534118652
    },
    {
      "auxiliary_loss_clip": 0.01251578,
      "auxiliary_loss_mlp": 0.00766174,
      "balance_loss_clip": 1.06532121,
      "balance_loss_mlp": 1.00058961,
      "epoch": 0.07575302110262731,
      "flos": 19537200892800.0,
      "grad_norm": 2.7936885758862715,
      "language_loss": 0.62184578,
      "learning_rate": 3.978134592434208e-06,
      "loss": 0.64202332,
      "num_input_tokens_seen": 13358185,
      "step": 630,
      "time_per_iteration": 2.7721550464630127
    },
    {
      "auxiliary_loss_clip": 0.0111171,
      "auxiliary_loss_mlp": 0.01005466,
      "balance_loss_clip": 1.04026532,
      "balance_loss_mlp": 1.00026894,
      "epoch": 0.0758732639932664,
      "flos": 67961808017280.0,
      "grad_norm": 1.0017116308081735,
      "language_loss": 0.59355319,
      "learning_rate": 3.978019571455123e-06,
      "loss": 0.61472493,
      "num_input_tokens_seen": 13410130,
      "step": 631,
      "time_per_iteration": 3.3719329833984375
    },
    {
      "auxiliary_loss_clip": 0.01263147,
      "auxiliary_loss_mlp": 0.01050222,
      "balance_loss_clip": 1.07086647,
      "balance_loss_mlp": 1.03799105,
      "epoch": 0.07599350688390549,
      "flos": 18989204025600.0,
      "grad_norm": 2.1757507872781843,
      "language_loss": 0.83882713,
      "learning_rate": 3.977904250412042e-06,
      "loss": 0.86196089,
      "num_input_tokens_seen": 13429085,
      "step": 632,
      "time_per_iteration": 2.732921600341797
    },
    {
      "auxiliary_loss_clip": 0.01238489,
      "auxiliary_loss_mlp": 0.01045879,
      "balance_loss_clip": 1.06654298,
      "balance_loss_mlp": 1.03339183,
      "epoch": 0.07611374977454458,
      "flos": 21069006341760.0,
      "grad_norm": 2.6352270921505627,
      "language_loss": 0.8567484,
      "learning_rate": 3.97778862932246e-06,
      "loss": 0.87959212,
      "num_input_tokens_seen": 13446250,
      "step": 633,
      "time_per_iteration": 2.758497953414917
    },
    {
      "auxiliary_loss_clip": 0.01131955,
      "auxiliary_loss_mlp": 0.01043017,
      "balance_loss_clip": 1.04479384,
      "balance_loss_mlp": 1.03061938,
      "epoch": 0.07623399266518367,
      "flos": 18514536773760.0,
      "grad_norm": 2.8179001248391984,
      "language_loss": 0.94381821,
      "learning_rate": 3.9776727082039144e-06,
      "loss": 0.96556795,
      "num_input_tokens_seen": 13463220,
      "step": 634,
      "time_per_iteration": 3.2206778526306152
    },
    {
      "auxiliary_loss_clip": 0.01169657,
      "auxiliary_loss_mlp": 0.0100561,
      "balance_loss_clip": 1.04181182,
      "balance_loss_mlp": 1.0006268,
      "epoch": 0.07635423555582276,
      "flos": 44663036077440.0,
      "grad_norm": 0.8108504480942063,
      "language_loss": 0.55459434,
      "learning_rate": 3.977556487073991e-06,
      "loss": 0.57634699,
      "num_input_tokens_seen": 13517775,
      "step": 635,
      "time_per_iteration": 3.588414430618286
    },
    {
      "auxiliary_loss_clip": 0.01218399,
      "auxiliary_loss_mlp": 0.01050786,
      "balance_loss_clip": 1.05915666,
      "balance_loss_mlp": 1.03822184,
      "epoch": 0.07647447844646185,
      "flos": 21761148487680.0,
      "grad_norm": 1.7114184728097788,
      "language_loss": 0.81469089,
      "learning_rate": 3.97743996595032e-06,
      "loss": 0.83738279,
      "num_input_tokens_seen": 13537815,
      "step": 636,
      "time_per_iteration": 2.7424802780151367
    },
    {
      "auxiliary_loss_clip": 0.01264466,
      "auxiliary_loss_mlp": 0.01041865,
      "balance_loss_clip": 1.06963754,
      "balance_loss_mlp": 1.02801847,
      "epoch": 0.07659472133710095,
      "flos": 23806799948160.0,
      "grad_norm": 1.5562689182378755,
      "language_loss": 0.81803429,
      "learning_rate": 3.9773231448505804e-06,
      "loss": 0.84109759,
      "num_input_tokens_seen": 13559605,
      "step": 637,
      "time_per_iteration": 2.6903228759765625
    },
    {
      "auxiliary_loss_clip": 0.01227134,
      "auxiliary_loss_mlp": 0.00767283,
      "balance_loss_clip": 1.06557965,
      "balance_loss_mlp": 1.00058293,
      "epoch": 0.07671496422774003,
      "flos": 21469984842240.0,
      "grad_norm": 13.499698074999108,
      "language_loss": 0.78413266,
      "learning_rate": 3.977206023792491e-06,
      "loss": 0.80407679,
      "num_input_tokens_seen": 13579495,
      "step": 638,
      "time_per_iteration": 2.7648534774780273
    },
    {
      "auxiliary_loss_clip": 0.01243559,
      "auxiliary_loss_mlp": 0.01044072,
      "balance_loss_clip": 1.06733739,
      "balance_loss_mlp": 1.0308218,
      "epoch": 0.07683520711837913,
      "flos": 16980971558400.0,
      "grad_norm": 2.6399330717275893,
      "language_loss": 0.81288517,
      "learning_rate": 3.97708860279382e-06,
      "loss": 0.83576143,
      "num_input_tokens_seen": 13597605,
      "step": 639,
      "time_per_iteration": 2.8613946437835693
    },
    {
      "auxiliary_loss_clip": 0.01205985,
      "auxiliary_loss_mlp": 0.01045563,
      "balance_loss_clip": 1.06039095,
      "balance_loss_mlp": 1.03230143,
      "epoch": 0.07695545000901821,
      "flos": 23476744851840.0,
      "grad_norm": 1.7793490510944252,
      "language_loss": 0.78112751,
      "learning_rate": 3.97697088187238e-06,
      "loss": 0.80364299,
      "num_input_tokens_seen": 13618120,
      "step": 640,
      "time_per_iteration": 2.7413299083709717
    },
    {
      "auxiliary_loss_clip": 0.01229562,
      "auxiliary_loss_mlp": 0.01047609,
      "balance_loss_clip": 1.06778765,
      "balance_loss_mlp": 1.03506196,
      "epoch": 0.07707569289965731,
      "flos": 17634258167040.0,
      "grad_norm": 2.1095951714488668,
      "language_loss": 0.92218769,
      "learning_rate": 3.976852861046029e-06,
      "loss": 0.9449594,
      "num_input_tokens_seen": 13634735,
      "step": 641,
      "time_per_iteration": 3.655611515045166
    },
    {
      "auxiliary_loss_clip": 0.01196123,
      "auxiliary_loss_mlp": 0.01051685,
      "balance_loss_clip": 1.06461263,
      "balance_loss_mlp": 1.03862596,
      "epoch": 0.0771959357902964,
      "flos": 25775674087680.0,
      "grad_norm": 1.5462469115400466,
      "language_loss": 0.80450118,
      "learning_rate": 3.97673454033267e-06,
      "loss": 0.82697928,
      "num_input_tokens_seen": 13656835,
      "step": 642,
      "time_per_iteration": 2.8135969638824463
    },
    {
      "auxiliary_loss_clip": 0.0122969,
      "auxiliary_loss_mlp": 0.01046167,
      "balance_loss_clip": 1.06342769,
      "balance_loss_mlp": 1.03313136,
      "epoch": 0.07731617868093549,
      "flos": 19828651847040.0,
      "grad_norm": 2.020864726259297,
      "language_loss": 0.82506478,
      "learning_rate": 3.976615919750254e-06,
      "loss": 0.84782338,
      "num_input_tokens_seen": 13674535,
      "step": 643,
      "time_per_iteration": 2.7075448036193848
    },
    {
      "auxiliary_loss_clip": 0.01240993,
      "auxiliary_loss_mlp": 0.01050252,
      "balance_loss_clip": 1.06772649,
      "balance_loss_mlp": 1.03712726,
      "epoch": 0.07743642157157458,
      "flos": 21324654414720.0,
      "grad_norm": 1.8523831467072889,
      "language_loss": 0.8700943,
      "learning_rate": 3.976496999316775e-06,
      "loss": 0.89300674,
      "num_input_tokens_seen": 13693290,
      "step": 644,
      "time_per_iteration": 3.634830951690674
    },
    {
      "auxiliary_loss_clip": 0.01233671,
      "auxiliary_loss_mlp": 0.01046,
      "balance_loss_clip": 1.07120335,
      "balance_loss_mlp": 1.03345287,
      "epoch": 0.07755666446221367,
      "flos": 19969133938560.0,
      "grad_norm": 2.0231060207296725,
      "language_loss": 0.84091461,
      "learning_rate": 3.976377779050271e-06,
      "loss": 0.86371136,
      "num_input_tokens_seen": 13711420,
      "step": 645,
      "time_per_iteration": 3.560788154602051
    },
    {
      "auxiliary_loss_clip": 0.01235197,
      "auxiliary_loss_mlp": 0.01045366,
      "balance_loss_clip": 1.06389594,
      "balance_loss_mlp": 1.03195548,
      "epoch": 0.07767690735285276,
      "flos": 23623224514560.0,
      "grad_norm": 2.8003029888870454,
      "language_loss": 0.84315658,
      "learning_rate": 3.976258258968831e-06,
      "loss": 0.86596215,
      "num_input_tokens_seen": 13729965,
      "step": 646,
      "time_per_iteration": 3.552372455596924
    },
    {
      "auxiliary_loss_clip": 0.01216262,
      "auxiliary_loss_mlp": 0.01043758,
      "balance_loss_clip": 1.06652737,
      "balance_loss_mlp": 1.03090096,
      "epoch": 0.07779715024349185,
      "flos": 22236246702720.0,
      "grad_norm": 2.3474605646442477,
      "language_loss": 0.74300957,
      "learning_rate": 3.976138439090583e-06,
      "loss": 0.76560974,
      "num_input_tokens_seen": 13748045,
      "step": 647,
      "time_per_iteration": 2.750065326690674
    },
    {
      "auxiliary_loss_clip": 0.01212134,
      "auxiliary_loss_mlp": 0.01043747,
      "balance_loss_clip": 1.06496298,
      "balance_loss_mlp": 1.03089595,
      "epoch": 0.07791739313413094,
      "flos": 20955097336320.0,
      "grad_norm": 2.754996001697786,
      "language_loss": 0.84877717,
      "learning_rate": 3.976018319433706e-06,
      "loss": 0.87133604,
      "num_input_tokens_seen": 13765590,
      "step": 648,
      "time_per_iteration": 2.650344133377075
    },
    {
      "auxiliary_loss_clip": 0.01245154,
      "auxiliary_loss_mlp": 0.01045853,
      "balance_loss_clip": 1.06705499,
      "balance_loss_mlp": 1.03294241,
      "epoch": 0.07803763602477004,
      "flos": 19312327797120.0,
      "grad_norm": 2.383402959691909,
      "language_loss": 0.91423553,
      "learning_rate": 3.9758979000164205e-06,
      "loss": 0.93714559,
      "num_input_tokens_seen": 13782410,
      "step": 649,
      "time_per_iteration": 2.746569871902466
    },
    {
      "auxiliary_loss_clip": 0.01218129,
      "auxiliary_loss_mlp": 0.01046679,
      "balance_loss_clip": 1.06154442,
      "balance_loss_mlp": 1.03339958,
      "epoch": 0.07815787891540912,
      "flos": 22710806213760.0,
      "grad_norm": 2.7258674707255675,
      "language_loss": 0.72311056,
      "learning_rate": 3.975777180856995e-06,
      "loss": 0.74575853,
      "num_input_tokens_seen": 13801530,
      "step": 650,
      "time_per_iteration": 2.7542121410369873
    },
    {
      "auxiliary_loss_clip": 0.01263896,
      "auxiliary_loss_mlp": 0.01051339,
      "balance_loss_clip": 1.07055414,
      "balance_loss_mlp": 1.03813648,
      "epoch": 0.07827812180604822,
      "flos": 22711129436160.0,
      "grad_norm": 2.7953816954262187,
      "language_loss": 0.86331278,
      "learning_rate": 3.975656161973742e-06,
      "loss": 0.88646507,
      "num_input_tokens_seen": 13820615,
      "step": 651,
      "time_per_iteration": 2.7876672744750977
    },
    {
      "auxiliary_loss_clip": 0.01258149,
      "auxiliary_loss_mlp": 0.01046523,
      "balance_loss_clip": 1.06506479,
      "balance_loss_mlp": 1.03304625,
      "epoch": 0.0783983646966873,
      "flos": 21725597001600.0,
      "grad_norm": 2.479438875058371,
      "language_loss": 0.88980746,
      "learning_rate": 3.9755348433850194e-06,
      "loss": 0.91285419,
      "num_input_tokens_seen": 13835955,
      "step": 652,
      "time_per_iteration": 2.611888885498047
    },
    {
      "auxiliary_loss_clip": 0.01133718,
      "auxiliary_loss_mlp": 0.01018571,
      "balance_loss_clip": 1.03433418,
      "balance_loss_mlp": 1.01301575,
      "epoch": 0.0785186075873264,
      "flos": 60640877537280.0,
      "grad_norm": 0.9681590403985557,
      "language_loss": 0.63662952,
      "learning_rate": 3.975413225109232e-06,
      "loss": 0.6581524,
      "num_input_tokens_seen": 13896505,
      "step": 653,
      "time_per_iteration": 3.267427682876587
    },
    {
      "auxiliary_loss_clip": 0.01249411,
      "auxiliary_loss_mlp": 0.01046588,
      "balance_loss_clip": 1.0695951,
      "balance_loss_mlp": 1.03254485,
      "epoch": 0.0786388504779655,
      "flos": 23877902920320.0,
      "grad_norm": 4.551158836279269,
      "language_loss": 0.93552053,
      "learning_rate": 3.975291307164829e-06,
      "loss": 0.95848054,
      "num_input_tokens_seen": 13915150,
      "step": 654,
      "time_per_iteration": 2.7345218658447266
    },
    {
      "auxiliary_loss_clip": 0.01203775,
      "auxiliary_loss_mlp": 0.01036142,
      "balance_loss_clip": 1.06079221,
      "balance_loss_mlp": 1.02434051,
      "epoch": 0.07875909336860458,
      "flos": 15158684822400.0,
      "grad_norm": 1.9010958421918984,
      "language_loss": 0.85193247,
      "learning_rate": 3.975169089570306e-06,
      "loss": 0.87433165,
      "num_input_tokens_seen": 13933525,
      "step": 655,
      "time_per_iteration": 2.6870152950286865
    },
    {
      "auxiliary_loss_clip": 0.01235759,
      "auxiliary_loss_mlp": 0.01047479,
      "balance_loss_clip": 1.06658268,
      "balance_loss_mlp": 1.03468156,
      "epoch": 0.07887933625924368,
      "flos": 22236857233920.0,
      "grad_norm": 2.016431709667311,
      "language_loss": 0.91472542,
      "learning_rate": 3.975046572344202e-06,
      "loss": 0.93755782,
      "num_input_tokens_seen": 13949985,
      "step": 656,
      "time_per_iteration": 2.739366054534912
    },
    {
      "auxiliary_loss_clip": 0.01208627,
      "auxiliary_loss_mlp": 0.01041292,
      "balance_loss_clip": 1.05930781,
      "balance_loss_mlp": 1.02878094,
      "epoch": 0.07899957914988276,
      "flos": 20777734955520.0,
      "grad_norm": 1.8479765497574434,
      "language_loss": 0.7103464,
      "learning_rate": 3.974923755505103e-06,
      "loss": 0.73284554,
      "num_input_tokens_seen": 13969215,
      "step": 657,
      "time_per_iteration": 2.723175048828125
    },
    {
      "auxiliary_loss_clip": 0.0120665,
      "auxiliary_loss_mlp": 0.01050382,
      "balance_loss_clip": 1.06406927,
      "balance_loss_mlp": 1.03729939,
      "epoch": 0.07911982204052186,
      "flos": 23003047267200.0,
      "grad_norm": 1.7950925914752998,
      "language_loss": 0.91232836,
      "learning_rate": 3.974800639071641e-06,
      "loss": 0.93489867,
      "num_input_tokens_seen": 13989935,
      "step": 658,
      "time_per_iteration": 2.7492823600769043
    },
    {
      "auxiliary_loss_clip": 0.01166288,
      "auxiliary_loss_mlp": 0.00766439,
      "balance_loss_clip": 1.05419707,
      "balance_loss_mlp": 1.00031745,
      "epoch": 0.07924006493116094,
      "flos": 23111389664640.0,
      "grad_norm": 3.079121098956225,
      "language_loss": 1.00707841,
      "learning_rate": 3.974677223062492e-06,
      "loss": 1.02640557,
      "num_input_tokens_seen": 14007150,
      "step": 659,
      "time_per_iteration": 2.7649238109588623
    },
    {
      "auxiliary_loss_clip": 0.01229075,
      "auxiliary_loss_mlp": 0.01044235,
      "balance_loss_clip": 1.06776404,
      "balance_loss_mlp": 1.03085399,
      "epoch": 0.07936030782180004,
      "flos": 16472153450880.0,
      "grad_norm": 2.8233558904336706,
      "language_loss": 0.74490088,
      "learning_rate": 3.974553507496378e-06,
      "loss": 0.76763403,
      "num_input_tokens_seen": 14025725,
      "step": 660,
      "time_per_iteration": 2.6774539947509766
    },
    {
      "auxiliary_loss_clip": 0.01221873,
      "auxiliary_loss_mlp": 0.01048467,
      "balance_loss_clip": 1.06459475,
      "balance_loss_mlp": 1.0343231,
      "epoch": 0.07948055071243913,
      "flos": 23733290764800.0,
      "grad_norm": 2.3436207497808885,
      "language_loss": 0.89206803,
      "learning_rate": 3.974429492392068e-06,
      "loss": 0.91477144,
      "num_input_tokens_seen": 14045750,
      "step": 661,
      "time_per_iteration": 2.7273857593536377
    },
    {
      "auxiliary_loss_clip": 0.01260816,
      "auxiliary_loss_mlp": 0.00766743,
      "balance_loss_clip": 1.07056594,
      "balance_loss_mlp": 1.0003624,
      "epoch": 0.07960079360307822,
      "flos": 19573326996480.0,
      "grad_norm": 2.5342840785427834,
      "language_loss": 0.91150105,
      "learning_rate": 3.974305177768373e-06,
      "loss": 0.93177664,
      "num_input_tokens_seen": 14063960,
      "step": 662,
      "time_per_iteration": 2.7140934467315674
    },
    {
      "auxiliary_loss_clip": 0.01201469,
      "auxiliary_loss_mlp": 0.01044995,
      "balance_loss_clip": 1.06305552,
      "balance_loss_mlp": 1.03167295,
      "epoch": 0.07972103649371731,
      "flos": 23513409659520.0,
      "grad_norm": 2.2830597569813103,
      "language_loss": 0.86300647,
      "learning_rate": 3.974180563644152e-06,
      "loss": 0.88547111,
      "num_input_tokens_seen": 14082525,
      "step": 663,
      "time_per_iteration": 2.6993300914764404
    },
    {
      "auxiliary_loss_clip": 0.0123263,
      "auxiliary_loss_mlp": 0.01047235,
      "balance_loss_clip": 1.06595027,
      "balance_loss_mlp": 1.03529608,
      "epoch": 0.0798412793843564,
      "flos": 16726867770240.0,
      "grad_norm": 2.155790611079928,
      "language_loss": 0.89190286,
      "learning_rate": 3.97405565003831e-06,
      "loss": 0.91470146,
      "num_input_tokens_seen": 14098610,
      "step": 664,
      "time_per_iteration": 2.747103452682495
    },
    {
      "auxiliary_loss_clip": 0.01215416,
      "auxiliary_loss_mlp": 0.0104512,
      "balance_loss_clip": 1.06268322,
      "balance_loss_mlp": 1.0325911,
      "epoch": 0.07996152227499549,
      "flos": 18223337214720.0,
      "grad_norm": 2.2648918898753427,
      "language_loss": 0.78409982,
      "learning_rate": 3.973930436969794e-06,
      "loss": 0.80670524,
      "num_input_tokens_seen": 14117065,
      "step": 665,
      "time_per_iteration": 2.710326671600342
    },
    {
      "auxiliary_loss_clip": 0.01219266,
      "auxiliary_loss_mlp": 0.01053722,
      "balance_loss_clip": 1.06150174,
      "balance_loss_mlp": 1.03995359,
      "epoch": 0.08008176516563459,
      "flos": 20594877793920.0,
      "grad_norm": 1.7950458805067477,
      "language_loss": 0.85717058,
      "learning_rate": 3.973804924457602e-06,
      "loss": 0.87990052,
      "num_input_tokens_seen": 14135145,
      "step": 666,
      "time_per_iteration": 2.6944193840026855
    },
    {
      "auxiliary_loss_clip": 0.01221342,
      "auxiliary_loss_mlp": 0.01046128,
      "balance_loss_clip": 1.06180918,
      "balance_loss_mlp": 1.03135192,
      "epoch": 0.08020200805627367,
      "flos": 31834306863360.0,
      "grad_norm": 1.7753152305464721,
      "language_loss": 0.8553949,
      "learning_rate": 3.973679112520771e-06,
      "loss": 0.87806964,
      "num_input_tokens_seen": 14156860,
      "step": 667,
      "time_per_iteration": 3.761136531829834
    },
    {
      "auxiliary_loss_clip": 0.01199509,
      "auxiliary_loss_mlp": 0.01046045,
      "balance_loss_clip": 1.056916,
      "balance_loss_mlp": 1.03316426,
      "epoch": 0.08032225094691277,
      "flos": 17783503176960.0,
      "grad_norm": 2.0171366247085034,
      "language_loss": 0.98920327,
      "learning_rate": 3.973553001178389e-06,
      "loss": 1.01165879,
      "num_input_tokens_seen": 14174365,
      "step": 668,
      "time_per_iteration": 2.744783878326416
    },
    {
      "auxiliary_loss_clip": 0.01210438,
      "auxiliary_loss_mlp": 0.01037566,
      "balance_loss_clip": 1.06546521,
      "balance_loss_mlp": 1.02421427,
      "epoch": 0.08044249383755185,
      "flos": 24061693835520.0,
      "grad_norm": 2.724550520006448,
      "language_loss": 0.757469,
      "learning_rate": 3.973426590449585e-06,
      "loss": 0.77994907,
      "num_input_tokens_seen": 14192320,
      "step": 669,
      "time_per_iteration": 2.766023635864258
    },
    {
      "auxiliary_loss_clip": 0.01197543,
      "auxiliary_loss_mlp": 0.01050949,
      "balance_loss_clip": 1.06214976,
      "balance_loss_mlp": 1.03898025,
      "epoch": 0.08056273672819095,
      "flos": 18223624523520.0,
      "grad_norm": 2.2438785365717155,
      "language_loss": 0.7539289,
      "learning_rate": 3.9732998803535364e-06,
      "loss": 0.77641386,
      "num_input_tokens_seen": 14210380,
      "step": 670,
      "time_per_iteration": 3.617702007293701
    },
    {
      "auxiliary_loss_clip": 0.0126021,
      "auxiliary_loss_mlp": 0.01045846,
      "balance_loss_clip": 1.06901252,
      "balance_loss_mlp": 1.03272688,
      "epoch": 0.08068297961883003,
      "flos": 19676856971520.0,
      "grad_norm": 2.4803702379559733,
      "language_loss": 0.85466373,
      "learning_rate": 3.973172870909465e-06,
      "loss": 0.87772429,
      "num_input_tokens_seen": 14225145,
      "step": 671,
      "time_per_iteration": 3.518620252609253
    },
    {
      "auxiliary_loss_clip": 0.01230826,
      "auxiliary_loss_mlp": 0.01056695,
      "balance_loss_clip": 1.06166792,
      "balance_loss_mlp": 1.042974,
      "epoch": 0.08080322250946913,
      "flos": 23148736830720.0,
      "grad_norm": 2.5574045988186542,
      "language_loss": 0.80810928,
      "learning_rate": 3.973045562136638e-06,
      "loss": 0.83098453,
      "num_input_tokens_seen": 14241960,
      "step": 672,
      "time_per_iteration": 3.5936081409454346
    },
    {
      "auxiliary_loss_clip": 0.01247713,
      "auxiliary_loss_mlp": 0.01047933,
      "balance_loss_clip": 1.06666267,
      "balance_loss_mlp": 1.03614914,
      "epoch": 0.08092346540010822,
      "flos": 21763626526080.0,
      "grad_norm": 2.14255213459971,
      "language_loss": 0.91484416,
      "learning_rate": 3.972917954054368e-06,
      "loss": 0.93780059,
      "num_input_tokens_seen": 14260515,
      "step": 673,
      "time_per_iteration": 2.6793365478515625
    },
    {
      "auxiliary_loss_clip": 0.01229853,
      "auxiliary_loss_mlp": 0.01052898,
      "balance_loss_clip": 1.07022917,
      "balance_loss_mlp": 1.03851557,
      "epoch": 0.08104370829074731,
      "flos": 21032485188480.0,
      "grad_norm": 2.659748712115016,
      "language_loss": 0.81934547,
      "learning_rate": 3.972790046682013e-06,
      "loss": 0.84217286,
      "num_input_tokens_seen": 14279190,
      "step": 674,
      "time_per_iteration": 2.7026286125183105
    },
    {
      "auxiliary_loss_clip": 0.01212019,
      "auxiliary_loss_mlp": 0.01042176,
      "balance_loss_clip": 1.06014276,
      "balance_loss_mlp": 1.02945662,
      "epoch": 0.0811639511813864,
      "flos": 20083186598400.0,
      "grad_norm": 1.9966303642096719,
      "language_loss": 0.79499924,
      "learning_rate": 3.972661840038977e-06,
      "loss": 0.81754112,
      "num_input_tokens_seen": 14299480,
      "step": 675,
      "time_per_iteration": 2.7462522983551025
    },
    {
      "auxiliary_loss_clip": 0.01249442,
      "auxiliary_loss_mlp": 0.01052621,
      "balance_loss_clip": 1.07153201,
      "balance_loss_mlp": 1.03936505,
      "epoch": 0.08128419407202549,
      "flos": 16836718538880.0,
      "grad_norm": 2.7810527676196397,
      "language_loss": 0.83501029,
      "learning_rate": 3.972533334144707e-06,
      "loss": 0.85803092,
      "num_input_tokens_seen": 14316405,
      "step": 676,
      "time_per_iteration": 2.6242363452911377
    },
    {
      "auxiliary_loss_clip": 0.01247163,
      "auxiliary_loss_mlp": 0.01044762,
      "balance_loss_clip": 1.06664872,
      "balance_loss_mlp": 1.03051043,
      "epoch": 0.08140443696266458,
      "flos": 23769273214080.0,
      "grad_norm": 2.1328014802487836,
      "language_loss": 0.78496337,
      "learning_rate": 3.972404529018699e-06,
      "loss": 0.80788261,
      "num_input_tokens_seen": 14336265,
      "step": 677,
      "time_per_iteration": 2.7077319622039795
    },
    {
      "auxiliary_loss_clip": 0.01222862,
      "auxiliary_loss_mlp": 0.01047054,
      "balance_loss_clip": 1.06021166,
      "balance_loss_mlp": 1.03386927,
      "epoch": 0.08152467985330367,
      "flos": 24390132819840.0,
      "grad_norm": 1.9413844099161806,
      "language_loss": 0.85340655,
      "learning_rate": 3.972275424680493e-06,
      "loss": 0.87610567,
      "num_input_tokens_seen": 14356375,
      "step": 678,
      "time_per_iteration": 2.7961320877075195
    },
    {
      "auxiliary_loss_clip": 0.01259722,
      "auxiliary_loss_mlp": 0.01042584,
      "balance_loss_clip": 1.06969404,
      "balance_loss_mlp": 1.02979279,
      "epoch": 0.08164492274394276,
      "flos": 19317750750720.0,
      "grad_norm": 2.7103477167499763,
      "language_loss": 0.920012,
      "learning_rate": 3.972146021149673e-06,
      "loss": 0.94303501,
      "num_input_tokens_seen": 14374650,
      "step": 679,
      "time_per_iteration": 2.6265804767608643
    },
    {
      "auxiliary_loss_clip": 0.0121503,
      "auxiliary_loss_mlp": 0.01037906,
      "balance_loss_clip": 1.06470227,
      "balance_loss_mlp": 1.02546704,
      "epoch": 0.08176516563458186,
      "flos": 14830461319680.0,
      "grad_norm": 5.743056017428087,
      "language_loss": 0.78892112,
      "learning_rate": 3.972016318445868e-06,
      "loss": 0.81145048,
      "num_input_tokens_seen": 14392650,
      "step": 680,
      "time_per_iteration": 2.748793840408325
    },
    {
      "auxiliary_loss_clip": 0.01251021,
      "auxiliary_loss_mlp": 0.01045014,
      "balance_loss_clip": 1.06926346,
      "balance_loss_mlp": 1.03152561,
      "epoch": 0.08188540852522094,
      "flos": 22602320161920.0,
      "grad_norm": 2.282037419579402,
      "language_loss": 0.92086124,
      "learning_rate": 3.971886316588757e-06,
      "loss": 0.94382167,
      "num_input_tokens_seen": 14413155,
      "step": 681,
      "time_per_iteration": 2.6780483722686768
    },
    {
      "auxiliary_loss_clip": 0.01203733,
      "auxiliary_loss_mlp": 0.01047748,
      "balance_loss_clip": 1.06427979,
      "balance_loss_mlp": 1.03464651,
      "epoch": 0.08200565141586004,
      "flos": 19463727623040.0,
      "grad_norm": 3.9623611601020374,
      "language_loss": 0.73622012,
      "learning_rate": 3.9717560155980595e-06,
      "loss": 0.75873494,
      "num_input_tokens_seen": 14428805,
      "step": 682,
      "time_per_iteration": 2.6980581283569336
    },
    {
      "auxiliary_loss_clip": 0.01246413,
      "auxiliary_loss_mlp": 0.01051862,
      "balance_loss_clip": 1.06792903,
      "balance_loss_mlp": 1.0387553,
      "epoch": 0.08212589430649912,
      "flos": 20594662312320.0,
      "grad_norm": 2.0517749349244028,
      "language_loss": 0.91840953,
      "learning_rate": 3.971625415493542e-06,
      "loss": 0.9413923,
      "num_input_tokens_seen": 14447125,
      "step": 683,
      "time_per_iteration": 2.656672477722168
    },
    {
      "auxiliary_loss_clip": 0.01212506,
      "auxiliary_loss_mlp": 0.01040253,
      "balance_loss_clip": 1.06385517,
      "balance_loss_mlp": 1.02676463,
      "epoch": 0.08224613719713822,
      "flos": 25953611086080.0,
      "grad_norm": 2.0398826269649515,
      "language_loss": 0.87763906,
      "learning_rate": 3.971494516295017e-06,
      "loss": 0.90016663,
      "num_input_tokens_seen": 14466575,
      "step": 684,
      "time_per_iteration": 2.7585256099700928
    },
    {
      "auxiliary_loss_clip": 0.01214612,
      "auxiliary_loss_mlp": 0.01048093,
      "balance_loss_clip": 1.06122828,
      "balance_loss_mlp": 1.03522408,
      "epoch": 0.08236638008777732,
      "flos": 23768734510080.0,
      "grad_norm": 2.6034344329253427,
      "language_loss": 0.85242552,
      "learning_rate": 3.971363318022341e-06,
      "loss": 0.87505251,
      "num_input_tokens_seen": 14487915,
      "step": 685,
      "time_per_iteration": 2.760298013687134
    },
    {
      "auxiliary_loss_clip": 0.0122892,
      "auxiliary_loss_mlp": 0.01047175,
      "balance_loss_clip": 1.06367302,
      "balance_loss_mlp": 1.03446734,
      "epoch": 0.0824866229784164,
      "flos": 38799144887040.0,
      "grad_norm": 1.9590745625048704,
      "language_loss": 0.68354154,
      "learning_rate": 3.971231820695417e-06,
      "loss": 0.70630252,
      "num_input_tokens_seen": 14511530,
      "step": 686,
      "time_per_iteration": 2.845672845840454
    },
    {
      "auxiliary_loss_clip": 0.01232359,
      "auxiliary_loss_mlp": 0.01050199,
      "balance_loss_clip": 1.06686616,
      "balance_loss_mlp": 1.03744912,
      "epoch": 0.0826068658690555,
      "flos": 23107762391040.0,
      "grad_norm": 2.7992248273957006,
      "language_loss": 0.8119337,
      "learning_rate": 3.971100024334193e-06,
      "loss": 0.83475929,
      "num_input_tokens_seen": 14529050,
      "step": 687,
      "time_per_iteration": 2.6569132804870605
    },
    {
      "auxiliary_loss_clip": 0.01193925,
      "auxiliary_loss_mlp": 0.01042343,
      "balance_loss_clip": 1.05751133,
      "balance_loss_mlp": 1.02958739,
      "epoch": 0.08272710875969458,
      "flos": 21136374299520.0,
      "grad_norm": 2.127174707314435,
      "language_loss": 0.86258209,
      "learning_rate": 3.970967928958663e-06,
      "loss": 0.8849448,
      "num_input_tokens_seen": 14546165,
      "step": 688,
      "time_per_iteration": 2.7673656940460205
    },
    {
      "auxiliary_loss_clip": 0.01203311,
      "auxiliary_loss_mlp": 0.01045625,
      "balance_loss_clip": 1.06268549,
      "balance_loss_mlp": 1.03302431,
      "epoch": 0.08284735165033368,
      "flos": 19063000517760.0,
      "grad_norm": 1.8616118731682587,
      "language_loss": 0.83487213,
      "learning_rate": 3.970835534588865e-06,
      "loss": 0.85736144,
      "num_input_tokens_seen": 14563660,
      "step": 689,
      "time_per_iteration": 2.7525806427001953
    },
    {
      "auxiliary_loss_clip": 0.01233901,
      "auxiliary_loss_mlp": 0.01045432,
      "balance_loss_clip": 1.07063866,
      "balance_loss_mlp": 1.03281987,
      "epoch": 0.08296759454097276,
      "flos": 16727442387840.0,
      "grad_norm": 1.8247371081941353,
      "language_loss": 0.85330856,
      "learning_rate": 3.970702841244883e-06,
      "loss": 0.87610185,
      "num_input_tokens_seen": 14581980,
      "step": 690,
      "time_per_iteration": 2.672231435775757
    },
    {
      "auxiliary_loss_clip": 0.01248408,
      "auxiliary_loss_mlp": 0.01050024,
      "balance_loss_clip": 1.07248449,
      "balance_loss_mlp": 1.03710198,
      "epoch": 0.08308783743161186,
      "flos": 18004928567040.0,
      "grad_norm": 2.073013076449816,
      "language_loss": 0.82841212,
      "learning_rate": 3.970569848946847e-06,
      "loss": 0.85139644,
      "num_input_tokens_seen": 14601795,
      "step": 691,
      "time_per_iteration": 2.5989747047424316
    },
    {
      "auxiliary_loss_clip": 0.01231282,
      "auxiliary_loss_mlp": 0.01045423,
      "balance_loss_clip": 1.06554413,
      "balance_loss_mlp": 1.03340673,
      "epoch": 0.08320808032225095,
      "flos": 15079788599040.0,
      "grad_norm": 2.181946994990189,
      "language_loss": 0.82509315,
      "learning_rate": 3.970436557714932e-06,
      "loss": 0.84786022,
      "num_input_tokens_seen": 14618315,
      "step": 692,
      "time_per_iteration": 2.7314164638519287
    },
    {
      "auxiliary_loss_clip": 0.01221709,
      "auxiliary_loss_mlp": 0.01044254,
      "balance_loss_clip": 1.06309927,
      "balance_loss_mlp": 1.03008056,
      "epoch": 0.08332832321289003,
      "flos": 22383085501440.0,
      "grad_norm": 2.948274446222267,
      "language_loss": 0.86432374,
      "learning_rate": 3.970302967569358e-06,
      "loss": 0.88698339,
      "num_input_tokens_seen": 14636905,
      "step": 693,
      "time_per_iteration": 2.750959634780884
    },
    {
      "auxiliary_loss_clip": 0.01242371,
      "auxiliary_loss_mlp": 0.01053536,
      "balance_loss_clip": 1.06880677,
      "balance_loss_mlp": 1.03972578,
      "epoch": 0.08344856610352913,
      "flos": 24717386655360.0,
      "grad_norm": 10.584755954454883,
      "language_loss": 0.68796128,
      "learning_rate": 3.9701690785303896e-06,
      "loss": 0.71092033,
      "num_input_tokens_seen": 14656100,
      "step": 694,
      "time_per_iteration": 3.595456838607788
    },
    {
      "auxiliary_loss_clip": 0.01246837,
      "auxiliary_loss_mlp": 0.01046018,
      "balance_loss_clip": 1.06837487,
      "balance_loss_mlp": 1.03300595,
      "epoch": 0.08356880899416821,
      "flos": 25370206387200.0,
      "grad_norm": 2.2797396386524014,
      "language_loss": 0.88473594,
      "learning_rate": 3.970034890618339e-06,
      "loss": 0.90766442,
      "num_input_tokens_seen": 14675790,
      "step": 695,
      "time_per_iteration": 2.8468432426452637
    },
    {
      "auxiliary_loss_clip": 0.01228479,
      "auxiliary_loss_mlp": 0.0104265,
      "balance_loss_clip": 1.06473327,
      "balance_loss_mlp": 1.03079474,
      "epoch": 0.08368905188480731,
      "flos": 24353072962560.0,
      "grad_norm": 2.5071696388115656,
      "language_loss": 0.88209796,
      "learning_rate": 3.969900403853562e-06,
      "loss": 0.9048093,
      "num_input_tokens_seen": 14694830,
      "step": 696,
      "time_per_iteration": 3.7635116577148438
    },
    {
      "auxiliary_loss_clip": 0.01262162,
      "auxiliary_loss_mlp": 0.01045396,
      "balance_loss_clip": 1.0730418,
      "balance_loss_mlp": 1.03315878,
      "epoch": 0.08380929477544641,
      "flos": 18037319656320.0,
      "grad_norm": 1.769552686175701,
      "language_loss": 0.78107512,
      "learning_rate": 3.96976561825646e-06,
      "loss": 0.80415064,
      "num_input_tokens_seen": 14711920,
      "step": 697,
      "time_per_iteration": 3.4891507625579834
    },
    {
      "auxiliary_loss_clip": 0.01202907,
      "auxiliary_loss_mlp": 0.01048161,
      "balance_loss_clip": 1.06612742,
      "balance_loss_mlp": 1.03479159,
      "epoch": 0.08392953766608549,
      "flos": 26286287875200.0,
      "grad_norm": 2.0194426955691314,
      "language_loss": 0.87234247,
      "learning_rate": 3.969630533847479e-06,
      "loss": 0.89485317,
      "num_input_tokens_seen": 14730880,
      "step": 698,
      "time_per_iteration": 3.704092502593994
    },
    {
      "auxiliary_loss_clip": 0.01243063,
      "auxiliary_loss_mlp": 0.01036836,
      "balance_loss_clip": 1.06575418,
      "balance_loss_mlp": 1.02510595,
      "epoch": 0.08404978055672459,
      "flos": 22492146170880.0,
      "grad_norm": 2.777024362057275,
      "language_loss": 0.84283358,
      "learning_rate": 3.969495150647113e-06,
      "loss": 0.86563253,
      "num_input_tokens_seen": 14749050,
      "step": 699,
      "time_per_iteration": 2.719754457473755
    },
    {
      "auxiliary_loss_clip": 0.01209372,
      "auxiliary_loss_mlp": 0.01043586,
      "balance_loss_clip": 1.06598949,
      "balance_loss_mlp": 1.03032398,
      "epoch": 0.08417002344736367,
      "flos": 24826878288000.0,
      "grad_norm": 1.7877805385368393,
      "language_loss": 0.76581144,
      "learning_rate": 3.969359468675899e-06,
      "loss": 0.78834099,
      "num_input_tokens_seen": 14769180,
      "step": 700,
      "time_per_iteration": 2.732675075531006
    },
    {
      "auxiliary_loss_clip": 0.01238464,
      "auxiliary_loss_mlp": 0.01044252,
      "balance_loss_clip": 1.06728113,
      "balance_loss_mlp": 1.03196156,
      "epoch": 0.08429026633800277,
      "flos": 16945922862720.0,
      "grad_norm": 2.1459497001476158,
      "language_loss": 0.89544439,
      "learning_rate": 3.969223487954418e-06,
      "loss": 0.91827154,
      "num_input_tokens_seen": 14786640,
      "step": 701,
      "time_per_iteration": 2.6603808403015137
    },
    {
      "auxiliary_loss_clip": 0.01194966,
      "auxiliary_loss_mlp": 0.01038167,
      "balance_loss_clip": 1.06360745,
      "balance_loss_mlp": 1.0261209,
      "epoch": 0.08441050922864185,
      "flos": 23841920471040.0,
      "grad_norm": 2.586219021030874,
      "language_loss": 0.83047974,
      "learning_rate": 3.969087208503301e-06,
      "loss": 0.85281104,
      "num_input_tokens_seen": 14806720,
      "step": 702,
      "time_per_iteration": 2.750436544418335
    },
    {
      "auxiliary_loss_clip": 0.01196682,
      "auxiliary_loss_mlp": 0.01043609,
      "balance_loss_clip": 1.06262255,
      "balance_loss_mlp": 1.03034067,
      "epoch": 0.08453075211928095,
      "flos": 25520205582720.0,
      "grad_norm": 2.4177860894404404,
      "language_loss": 0.84816563,
      "learning_rate": 3.968950630343219e-06,
      "loss": 0.87056857,
      "num_input_tokens_seen": 14823705,
      "step": 703,
      "time_per_iteration": 2.7762889862060547
    },
    {
      "auxiliary_loss_clip": 0.01220587,
      "auxiliary_loss_mlp": 0.01039826,
      "balance_loss_clip": 1.06103706,
      "balance_loss_mlp": 1.02772021,
      "epoch": 0.08465099500992004,
      "flos": 19532496211200.0,
      "grad_norm": 2.1464520325466285,
      "language_loss": 0.93510211,
      "learning_rate": 3.968813753494892e-06,
      "loss": 0.95770609,
      "num_input_tokens_seen": 14841865,
      "step": 704,
      "time_per_iteration": 2.6497962474823
    },
    {
      "auxiliary_loss_clip": 0.01196328,
      "auxiliary_loss_mlp": 0.00766134,
      "balance_loss_clip": 1.05840993,
      "balance_loss_mlp": 1.00028729,
      "epoch": 0.08477123790055913,
      "flos": 29351299403520.0,
      "grad_norm": 4.527949140847245,
      "language_loss": 0.75522226,
      "learning_rate": 3.968676577979084e-06,
      "loss": 0.77484685,
      "num_input_tokens_seen": 14861415,
      "step": 705,
      "time_per_iteration": 2.968214273452759
    },
    {
      "auxiliary_loss_clip": 0.01193664,
      "auxiliary_loss_mlp": 0.01041379,
      "balance_loss_clip": 1.06046247,
      "balance_loss_mlp": 1.02816451,
      "epoch": 0.08489148079119822,
      "flos": 18624495283200.0,
      "grad_norm": 2.353269958503046,
      "language_loss": 0.77990454,
      "learning_rate": 3.968539103816605e-06,
      "loss": 0.80225492,
      "num_input_tokens_seen": 14879215,
      "step": 706,
      "time_per_iteration": 2.7759852409362793
    },
    {
      "auxiliary_loss_clip": 0.01224104,
      "auxiliary_loss_mlp": 0.00765879,
      "balance_loss_clip": 1.06543887,
      "balance_loss_mlp": 1.00039184,
      "epoch": 0.0850117236818373,
      "flos": 23471393725440.0,
      "grad_norm": 1.8655679264177616,
      "language_loss": 0.89063811,
      "learning_rate": 3.9684013310283085e-06,
      "loss": 0.91053796,
      "num_input_tokens_seen": 14897900,
      "step": 707,
      "time_per_iteration": 2.76870059967041
    },
    {
      "auxiliary_loss_clip": 0.01228089,
      "auxiliary_loss_mlp": 0.01049562,
      "balance_loss_clip": 1.06886601,
      "balance_loss_mlp": 1.03786707,
      "epoch": 0.0851319665724764,
      "flos": 40625058896640.0,
      "grad_norm": 3.0941477794634555,
      "language_loss": 0.64072669,
      "learning_rate": 3.9682632596350956e-06,
      "loss": 0.66350317,
      "num_input_tokens_seen": 14919065,
      "step": 708,
      "time_per_iteration": 2.8337628841400146
    },
    {
      "auxiliary_loss_clip": 0.01239035,
      "auxiliary_loss_mlp": 0.01044301,
      "balance_loss_clip": 1.06920099,
      "balance_loss_mlp": 1.03195655,
      "epoch": 0.0852522094631155,
      "flos": 15879554870400.0,
      "grad_norm": 2.0227811858772586,
      "language_loss": 0.78342557,
      "learning_rate": 3.968124889657911e-06,
      "loss": 0.80625892,
      "num_input_tokens_seen": 14934165,
      "step": 709,
      "time_per_iteration": 2.6406357288360596
    },
    {
      "auxiliary_loss_clip": 0.0119353,
      "auxiliary_loss_mlp": 0.01045,
      "balance_loss_clip": 1.05971694,
      "balance_loss_mlp": 1.03212523,
      "epoch": 0.08537245235375458,
      "flos": 14567091822720.0,
      "grad_norm": 2.2968876080717573,
      "language_loss": 0.90613914,
      "learning_rate": 3.967986221117746e-06,
      "loss": 0.92852449,
      "num_input_tokens_seen": 14950105,
      "step": 710,
      "time_per_iteration": 2.7819108963012695
    },
    {
      "auxiliary_loss_clip": 0.01166738,
      "auxiliary_loss_mlp": 0.01051051,
      "balance_loss_clip": 1.05559039,
      "balance_loss_mlp": 1.03814077,
      "epoch": 0.08549269524439368,
      "flos": 26468929555200.0,
      "grad_norm": 4.729345521178511,
      "language_loss": 0.86564821,
      "learning_rate": 3.967847254035635e-06,
      "loss": 0.88782609,
      "num_input_tokens_seen": 14969490,
      "step": 711,
      "time_per_iteration": 3.0751190185546875
    },
    {
      "auxiliary_loss_clip": 0.01212795,
      "auxiliary_loss_mlp": 0.01047025,
      "balance_loss_clip": 1.06493855,
      "balance_loss_mlp": 1.03359568,
      "epoch": 0.08561293813503276,
      "flos": 13590214565760.0,
      "grad_norm": 2.0428458617370207,
      "language_loss": 0.86749256,
      "learning_rate": 3.967707988432661e-06,
      "loss": 0.8900907,
      "num_input_tokens_seen": 14987195,
      "step": 712,
      "time_per_iteration": 2.995281219482422
    },
    {
      "auxiliary_loss_clip": 0.01254467,
      "auxiliary_loss_mlp": 0.01045017,
      "balance_loss_clip": 1.06729472,
      "balance_loss_mlp": 1.0324049,
      "epoch": 0.08573318102567186,
      "flos": 26943524979840.0,
      "grad_norm": 2.491285011873415,
      "language_loss": 0.87602705,
      "learning_rate": 3.967568424329949e-06,
      "loss": 0.89902192,
      "num_input_tokens_seen": 15007620,
      "step": 713,
      "time_per_iteration": 2.659435749053955
    },
    {
      "auxiliary_loss_clip": 0.01152622,
      "auxiliary_loss_mlp": 0.01020732,
      "balance_loss_clip": 1.04647088,
      "balance_loss_mlp": 1.01598775,
      "epoch": 0.08585342391631094,
      "flos": 67302739319040.0,
      "grad_norm": 0.8351390594223864,
      "language_loss": 0.55526352,
      "learning_rate": 3.967428561748671e-06,
      "loss": 0.57699704,
      "num_input_tokens_seen": 15075590,
      "step": 714,
      "time_per_iteration": 3.45459246635437
    },
    {
      "auxiliary_loss_clip": 0.01179118,
      "auxiliary_loss_mlp": 0.01039305,
      "balance_loss_clip": 1.05474567,
      "balance_loss_mlp": 1.0258944,
      "epoch": 0.08597366680695004,
      "flos": 22456594684800.0,
      "grad_norm": 2.5559555314301963,
      "language_loss": 0.87391782,
      "learning_rate": 3.967288400710045e-06,
      "loss": 0.89610195,
      "num_input_tokens_seen": 15095055,
      "step": 715,
      "time_per_iteration": 2.7718629837036133
    },
    {
      "auxiliary_loss_clip": 0.01206893,
      "auxiliary_loss_mlp": 0.01043356,
      "balance_loss_clip": 1.06617701,
      "balance_loss_mlp": 1.03114271,
      "epoch": 0.08609390969758914,
      "flos": 23550505430400.0,
      "grad_norm": 2.358505753506581,
      "language_loss": 0.88715887,
      "learning_rate": 3.9671479412353335e-06,
      "loss": 0.90966135,
      "num_input_tokens_seen": 15113520,
      "step": 716,
      "time_per_iteration": 2.722043514251709
    },
    {
      "auxiliary_loss_clip": 0.01238073,
      "auxiliary_loss_mlp": 0.01048534,
      "balance_loss_clip": 1.06611252,
      "balance_loss_mlp": 1.03521276,
      "epoch": 0.08621415258822822,
      "flos": 25885848078720.0,
      "grad_norm": 2.8229772703437566,
      "language_loss": 0.74151629,
      "learning_rate": 3.967007183345843e-06,
      "loss": 0.76438236,
      "num_input_tokens_seen": 15133375,
      "step": 717,
      "time_per_iteration": 2.749692440032959
    },
    {
      "auxiliary_loss_clip": 0.01236393,
      "auxiliary_loss_mlp": 0.01040357,
      "balance_loss_clip": 1.06610739,
      "balance_loss_mlp": 1.02752423,
      "epoch": 0.08633439547886732,
      "flos": 13589568120960.0,
      "grad_norm": 3.2042491538703217,
      "language_loss": 0.89368844,
      "learning_rate": 3.966866127062927e-06,
      "loss": 0.91645586,
      "num_input_tokens_seen": 15150500,
      "step": 718,
      "time_per_iteration": 2.624999523162842
    },
    {
      "auxiliary_loss_clip": 0.01147985,
      "auxiliary_loss_mlp": 0.01004591,
      "balance_loss_clip": 1.04104447,
      "balance_loss_mlp": 0.99975073,
      "epoch": 0.0864546383695064,
      "flos": 57767342434560.0,
      "grad_norm": 0.8646542641634731,
      "language_loss": 0.62710309,
      "learning_rate": 3.966724772407982e-06,
      "loss": 0.64862883,
      "num_input_tokens_seen": 15208015,
      "step": 719,
      "time_per_iteration": 4.015966176986694
    },
    {
      "auxiliary_loss_clip": 0.01196814,
      "auxiliary_loss_mlp": 0.01050014,
      "balance_loss_clip": 1.06251395,
      "balance_loss_mlp": 1.03809869,
      "epoch": 0.0865748812601455,
      "flos": 20046952753920.0,
      "grad_norm": 2.0642124941767275,
      "language_loss": 0.88866639,
      "learning_rate": 3.966583119402454e-06,
      "loss": 0.91113466,
      "num_input_tokens_seen": 15224780,
      "step": 720,
      "time_per_iteration": 2.722700834274292
    },
    {
      "auxiliary_loss_clip": 0.01234984,
      "auxiliary_loss_mlp": 0.00766024,
      "balance_loss_clip": 1.06594515,
      "balance_loss_mlp": 1.00026035,
      "epoch": 0.08669512415078459,
      "flos": 35262446935680.0,
      "grad_norm": 1.8061645746717536,
      "language_loss": 0.82249022,
      "learning_rate": 3.9664411680678305e-06,
      "loss": 0.84250027,
      "num_input_tokens_seen": 15246535,
      "step": 721,
      "time_per_iteration": 2.8109726905822754
    },
    {
      "auxiliary_loss_clip": 0.01123631,
      "auxiliary_loss_mlp": 0.01006567,
      "balance_loss_clip": 1.03843272,
      "balance_loss_mlp": 1.00182283,
      "epoch": 0.08681536704142367,
      "flos": 65654870048640.0,
      "grad_norm": 0.8484583294587621,
      "language_loss": 0.6143558,
      "learning_rate": 3.966298918425644e-06,
      "loss": 0.63565779,
      "num_input_tokens_seen": 15304025,
      "step": 722,
      "time_per_iteration": 4.096959352493286
    },
    {
      "auxiliary_loss_clip": 0.01237361,
      "auxiliary_loss_mlp": 0.01047522,
      "balance_loss_clip": 1.0638063,
      "balance_loss_mlp": 1.03502929,
      "epoch": 0.08693560993206277,
      "flos": 34529940881280.0,
      "grad_norm": 3.534067229579416,
      "language_loss": 0.83069241,
      "learning_rate": 3.966156370497476e-06,
      "loss": 0.85354125,
      "num_input_tokens_seen": 15327635,
      "step": 723,
      "time_per_iteration": 3.7428741455078125
    },
    {
      "auxiliary_loss_clip": 0.01241142,
      "auxiliary_loss_mlp": 0.01046554,
      "balance_loss_clip": 1.0659306,
      "balance_loss_mlp": 1.03320241,
      "epoch": 0.08705585282270185,
      "flos": 23149419189120.0,
      "grad_norm": 1.7641467326062703,
      "language_loss": 0.88544208,
      "learning_rate": 3.96601352430495e-06,
      "loss": 0.908319,
      "num_input_tokens_seen": 15347405,
      "step": 724,
      "time_per_iteration": 3.5794427394866943
    },
    {
      "auxiliary_loss_clip": 0.01219568,
      "auxiliary_loss_mlp": 0.010426,
      "balance_loss_clip": 1.06602359,
      "balance_loss_mlp": 1.03014898,
      "epoch": 0.08717609571334095,
      "flos": 29497599498240.0,
      "grad_norm": 1.9273464001831033,
      "language_loss": 0.83000052,
      "learning_rate": 3.965870379869735e-06,
      "loss": 0.85262215,
      "num_input_tokens_seen": 15369450,
      "step": 725,
      "time_per_iteration": 2.828470468521118
    },
    {
      "auxiliary_loss_clip": 0.01235903,
      "auxiliary_loss_mlp": 0.01039862,
      "balance_loss_clip": 1.06201875,
      "balance_loss_mlp": 1.02736354,
      "epoch": 0.08729633860398003,
      "flos": 20667489137280.0,
      "grad_norm": 2.170208736759099,
      "language_loss": 0.87050009,
      "learning_rate": 3.965726937213547e-06,
      "loss": 0.89325768,
      "num_input_tokens_seen": 15388085,
      "step": 726,
      "time_per_iteration": 2.7191240787506104
    },
    {
      "auxiliary_loss_clip": 0.01234163,
      "auxiliary_loss_mlp": 0.01036836,
      "balance_loss_clip": 1.06258941,
      "balance_loss_mlp": 1.02457559,
      "epoch": 0.08741658149461913,
      "flos": 18369493655040.0,
      "grad_norm": 2.395705109954899,
      "language_loss": 0.81270868,
      "learning_rate": 3.965583196358144e-06,
      "loss": 0.83541864,
      "num_input_tokens_seen": 15407120,
      "step": 727,
      "time_per_iteration": 2.66363263130188
    },
    {
      "auxiliary_loss_clip": 0.01253028,
      "auxiliary_loss_mlp": 0.01040496,
      "balance_loss_clip": 1.06612349,
      "balance_loss_mlp": 1.02747297,
      "epoch": 0.08753682438525823,
      "flos": 18729677283840.0,
      "grad_norm": 3.674837615685462,
      "language_loss": 0.74490404,
      "learning_rate": 3.965439157325335e-06,
      "loss": 0.76783925,
      "num_input_tokens_seen": 15424485,
      "step": 728,
      "time_per_iteration": 2.6076135635375977
    },
    {
      "auxiliary_loss_clip": 0.0121188,
      "auxiliary_loss_mlp": 0.0104224,
      "balance_loss_clip": 1.05793107,
      "balance_loss_mlp": 1.02933562,
      "epoch": 0.08765706727589731,
      "flos": 27776113303680.0,
      "grad_norm": 2.3433970136250886,
      "language_loss": 0.75825644,
      "learning_rate": 3.965294820136968e-06,
      "loss": 0.7807976,
      "num_input_tokens_seen": 15446285,
      "step": 729,
      "time_per_iteration": 2.7820146083831787
    },
    {
      "auxiliary_loss_clip": 0.01218706,
      "auxiliary_loss_mlp": 0.01046808,
      "balance_loss_clip": 1.06444001,
      "balance_loss_mlp": 1.03421903,
      "epoch": 0.08777731016653641,
      "flos": 24389127239040.0,
      "grad_norm": 2.411264455360739,
      "language_loss": 0.86889023,
      "learning_rate": 3.965150184814938e-06,
      "loss": 0.89154536,
      "num_input_tokens_seen": 15465770,
      "step": 730,
      "time_per_iteration": 2.749265193939209
    },
    {
      "auxiliary_loss_clip": 0.01205231,
      "auxiliary_loss_mlp": 0.01044585,
      "balance_loss_clip": 1.05829072,
      "balance_loss_mlp": 1.03243732,
      "epoch": 0.08789755305717549,
      "flos": 21981855605760.0,
      "grad_norm": 2.0966104766328817,
      "language_loss": 0.76569158,
      "learning_rate": 3.965005251381189e-06,
      "loss": 0.78818977,
      "num_input_tokens_seen": 15483705,
      "step": 731,
      "time_per_iteration": 2.757766008377075
    },
    {
      "auxiliary_loss_clip": 0.01145601,
      "auxiliary_loss_mlp": 0.01004728,
      "balance_loss_clip": 1.0345,
      "balance_loss_mlp": 1.00022149,
      "epoch": 0.08801779594781459,
      "flos": 58360120583040.0,
      "grad_norm": 0.8995173539817212,
      "language_loss": 0.64620185,
      "learning_rate": 3.964860019857705e-06,
      "loss": 0.66770506,
      "num_input_tokens_seen": 15548620,
      "step": 732,
      "time_per_iteration": 3.2875545024871826
    },
    {
      "auxiliary_loss_clip": 0.01251721,
      "auxiliary_loss_mlp": 0.01043533,
      "balance_loss_clip": 1.06899881,
      "balance_loss_mlp": 1.03096807,
      "epoch": 0.08813803883845367,
      "flos": 23294785530240.0,
      "grad_norm": 1.8823011591025562,
      "language_loss": 0.84298432,
      "learning_rate": 3.964714490266518e-06,
      "loss": 0.86593688,
      "num_input_tokens_seen": 15569265,
      "step": 733,
      "time_per_iteration": 2.574460983276367
    },
    {
      "auxiliary_loss_clip": 0.01142264,
      "auxiliary_loss_mlp": 0.01005089,
      "balance_loss_clip": 1.03471923,
      "balance_loss_mlp": 1.00055873,
      "epoch": 0.08825828172909277,
      "flos": 63424924882560.0,
      "grad_norm": 0.9058110251577008,
      "language_loss": 0.64563841,
      "learning_rate": 3.964568662629706e-06,
      "loss": 0.66711193,
      "num_input_tokens_seen": 15630570,
      "step": 734,
      "time_per_iteration": 3.1315927505493164
    },
    {
      "auxiliary_loss_clip": 0.01230072,
      "auxiliary_loss_mlp": 0.01038409,
      "balance_loss_clip": 1.06045103,
      "balance_loss_mlp": 1.02548051,
      "epoch": 0.08837852461973186,
      "flos": 26720986268160.0,
      "grad_norm": 3.1982520278931736,
      "language_loss": 0.84514499,
      "learning_rate": 3.9644225369693895e-06,
      "loss": 0.8678298,
      "num_input_tokens_seen": 15650870,
      "step": 735,
      "time_per_iteration": 2.72280216217041
    },
    {
      "auxiliary_loss_clip": 0.01254193,
      "auxiliary_loss_mlp": 0.010461,
      "balance_loss_clip": 1.07066083,
      "balance_loss_mlp": 1.03453112,
      "epoch": 0.08849876751037095,
      "flos": 27265427688960.0,
      "grad_norm": 2.0383496174859963,
      "language_loss": 0.86618328,
      "learning_rate": 3.964276113307735e-06,
      "loss": 0.88918614,
      "num_input_tokens_seen": 15670835,
      "step": 736,
      "time_per_iteration": 2.6650760173797607
    },
    {
      "auxiliary_loss_clip": 0.01200667,
      "auxiliary_loss_mlp": 0.01042981,
      "balance_loss_clip": 1.06198967,
      "balance_loss_mlp": 1.03038692,
      "epoch": 0.08861901040101004,
      "flos": 19828759587840.0,
      "grad_norm": 1.9183812032945486,
      "language_loss": 0.81013262,
      "learning_rate": 3.9641293916669574e-06,
      "loss": 0.83256906,
      "num_input_tokens_seen": 15689795,
      "step": 737,
      "time_per_iteration": 2.755119562149048
    },
    {
      "auxiliary_loss_clip": 0.01201493,
      "auxiliary_loss_mlp": 0.01041637,
      "balance_loss_clip": 1.06092775,
      "balance_loss_mlp": 1.02878666,
      "epoch": 0.08873925329164913,
      "flos": 23658704173440.0,
      "grad_norm": 2.04386359428358,
      "language_loss": 0.83056951,
      "learning_rate": 3.9639823720693115e-06,
      "loss": 0.85300076,
      "num_input_tokens_seen": 15711650,
      "step": 738,
      "time_per_iteration": 2.812493324279785
    },
    {
      "auxiliary_loss_clip": 0.01127754,
      "auxiliary_loss_mlp": 0.01010225,
      "balance_loss_clip": 1.04175735,
      "balance_loss_mlp": 1.00483692,
      "epoch": 0.08885949618228822,
      "flos": 71831541893760.0,
      "grad_norm": 0.8473432630130484,
      "language_loss": 0.60009986,
      "learning_rate": 3.963835054537102e-06,
      "loss": 0.62147963,
      "num_input_tokens_seen": 15780615,
      "step": 739,
      "time_per_iteration": 3.373335123062134
    },
    {
      "auxiliary_loss_clip": 0.01215948,
      "auxiliary_loss_mlp": 0.0104004,
      "balance_loss_clip": 1.05923653,
      "balance_loss_mlp": 1.02798796,
      "epoch": 0.08897973907292732,
      "flos": 22346169298560.0,
      "grad_norm": 3.0956375336296422,
      "language_loss": 0.60901785,
      "learning_rate": 3.963687439092676e-06,
      "loss": 0.63157767,
      "num_input_tokens_seen": 15801300,
      "step": 740,
      "time_per_iteration": 2.6933131217956543
    },
    {
      "auxiliary_loss_clip": 0.01233963,
      "auxiliary_loss_mlp": 0.01041706,
      "balance_loss_clip": 1.0653615,
      "balance_loss_mlp": 1.02965403,
      "epoch": 0.0890999819635664,
      "flos": 21251827589760.0,
      "grad_norm": 2.0800687241841422,
      "language_loss": 0.80294734,
      "learning_rate": 3.963539525758427e-06,
      "loss": 0.82570398,
      "num_input_tokens_seen": 15820860,
      "step": 741,
      "time_per_iteration": 2.7075846195220947
    },
    {
      "auxiliary_loss_clip": 0.01224671,
      "auxiliary_loss_mlp": 0.01041473,
      "balance_loss_clip": 1.0651325,
      "balance_loss_mlp": 1.02945662,
      "epoch": 0.0892202248542055,
      "flos": 25370888745600.0,
      "grad_norm": 2.5090202074810315,
      "language_loss": 0.67660534,
      "learning_rate": 3.9633913145567925e-06,
      "loss": 0.69926679,
      "num_input_tokens_seen": 15841350,
      "step": 742,
      "time_per_iteration": 2.8041248321533203
    },
    {
      "auxiliary_loss_clip": 0.01225504,
      "auxiliary_loss_mlp": 0.01042978,
      "balance_loss_clip": 1.06542873,
      "balance_loss_mlp": 1.03087258,
      "epoch": 0.08934046774484458,
      "flos": 24457895827200.0,
      "grad_norm": 3.162040117072184,
      "language_loss": 0.81627935,
      "learning_rate": 3.9632428055102575e-06,
      "loss": 0.83896416,
      "num_input_tokens_seen": 15861360,
      "step": 743,
      "time_per_iteration": 2.7083804607391357
    },
    {
      "auxiliary_loss_clip": 0.01243309,
      "auxiliary_loss_mlp": 0.01034553,
      "balance_loss_clip": 1.06892991,
      "balance_loss_mlp": 1.02229202,
      "epoch": 0.08946071063548368,
      "flos": 35772773414400.0,
      "grad_norm": 2.4222512203101583,
      "language_loss": 0.66943353,
      "learning_rate": 3.9630939986413495e-06,
      "loss": 0.69221216,
      "num_input_tokens_seen": 15883160,
      "step": 744,
      "time_per_iteration": 2.7982137203216553
    },
    {
      "auxiliary_loss_clip": 0.01190332,
      "auxiliary_loss_mlp": 0.01047042,
      "balance_loss_clip": 1.05963683,
      "balance_loss_mlp": 1.03528762,
      "epoch": 0.08958095352612276,
      "flos": 14356584167040.0,
      "grad_norm": 1.787743573293069,
      "language_loss": 0.78074259,
      "learning_rate": 3.962944893972643e-06,
      "loss": 0.80311632,
      "num_input_tokens_seen": 15901610,
      "step": 745,
      "time_per_iteration": 3.629729747772217
    },
    {
      "auxiliary_loss_clip": 0.01217828,
      "auxiliary_loss_mlp": 0.01041113,
      "balance_loss_clip": 1.065323,
      "balance_loss_mlp": 1.02915633,
      "epoch": 0.08970119641676186,
      "flos": 17853277345920.0,
      "grad_norm": 2.8514971394322433,
      "language_loss": 0.90785909,
      "learning_rate": 3.962795491526756e-06,
      "loss": 0.93044853,
      "num_input_tokens_seen": 15918770,
      "step": 746,
      "time_per_iteration": 2.6967415809631348
    },
    {
      "auxiliary_loss_clip": 0.01255447,
      "auxiliary_loss_mlp": 0.0104973,
      "balance_loss_clip": 1.06849134,
      "balance_loss_mlp": 1.03721941,
      "epoch": 0.08982143930740095,
      "flos": 20811670329600.0,
      "grad_norm": 2.6480664192225105,
      "language_loss": 0.8881126,
      "learning_rate": 3.962645791326354e-06,
      "loss": 0.9111644,
      "num_input_tokens_seen": 15938025,
      "step": 747,
      "time_per_iteration": 3.612863063812256
    },
    {
      "auxiliary_loss_clip": 0.01233781,
      "auxiliary_loss_mlp": 0.01034143,
      "balance_loss_clip": 1.06807196,
      "balance_loss_mlp": 1.02262104,
      "epoch": 0.08994168219804004,
      "flos": 24097712198400.0,
      "grad_norm": 1.806066618536283,
      "language_loss": 0.83128136,
      "learning_rate": 3.962495793394146e-06,
      "loss": 0.85396057,
      "num_input_tokens_seen": 15957215,
      "step": 748,
      "time_per_iteration": 2.686896800994873
    },
    {
      "auxiliary_loss_clip": 0.01154345,
      "auxiliary_loss_mlp": 0.0100601,
      "balance_loss_clip": 1.03434443,
      "balance_loss_mlp": 1.00159967,
      "epoch": 0.09006192508867913,
      "flos": 57188893812480.0,
      "grad_norm": 0.7371593434349886,
      "language_loss": 0.61128163,
      "learning_rate": 3.9623454977528864e-06,
      "loss": 0.63288522,
      "num_input_tokens_seen": 16015870,
      "step": 749,
      "time_per_iteration": 4.017616271972656
    },
    {
      "auxiliary_loss_clip": 0.01206062,
      "auxiliary_loss_mlp": 0.01053556,
      "balance_loss_clip": 1.0599792,
      "balance_loss_mlp": 1.04044914,
      "epoch": 0.09018216797931822,
      "flos": 20487505063680.0,
      "grad_norm": 1.9758201657783958,
      "language_loss": 0.84978998,
      "learning_rate": 3.962194904425375e-06,
      "loss": 0.87238616,
      "num_input_tokens_seen": 16036500,
      "step": 750,
      "time_per_iteration": 3.651291847229004
    },
    {
      "auxiliary_loss_clip": 0.01232775,
      "auxiliary_loss_mlp": 0.01038304,
      "balance_loss_clip": 1.06524587,
      "balance_loss_mlp": 1.02711594,
      "epoch": 0.09030241086995731,
      "flos": 22638123043200.0,
      "grad_norm": 2.2728895278818304,
      "language_loss": 0.68001461,
      "learning_rate": 3.9620440134344566e-06,
      "loss": 0.70272541,
      "num_input_tokens_seen": 16054655,
      "step": 751,
      "time_per_iteration": 2.768510341644287
    },
    {
      "auxiliary_loss_clip": 0.0120409,
      "auxiliary_loss_mlp": 0.01041261,
      "balance_loss_clip": 1.06400156,
      "balance_loss_mlp": 1.02882123,
      "epoch": 0.09042265376059641,
      "flos": 21871502046720.0,
      "grad_norm": 2.7224802721906887,
      "language_loss": 0.82869315,
      "learning_rate": 3.9618928248030215e-06,
      "loss": 0.85114664,
      "num_input_tokens_seen": 16074165,
      "step": 752,
      "time_per_iteration": 2.7840282917022705
    },
    {
      "auxiliary_loss_clip": 0.01233282,
      "auxiliary_loss_mlp": 0.01039217,
      "balance_loss_clip": 1.06599772,
      "balance_loss_mlp": 1.02734351,
      "epoch": 0.0905428966512355,
      "flos": 24316192673280.0,
      "grad_norm": 2.744090631735611,
      "language_loss": 0.82770991,
      "learning_rate": 3.961741338554005e-06,
      "loss": 0.8504349,
      "num_input_tokens_seen": 16092505,
      "step": 753,
      "time_per_iteration": 2.681762218475342
    },
    {
      "auxiliary_loss_clip": 0.01230282,
      "auxiliary_loss_mlp": 0.01055359,
      "balance_loss_clip": 1.06675577,
      "balance_loss_mlp": 1.04274082,
      "epoch": 0.09066313954187459,
      "flos": 35845061535360.0,
      "grad_norm": 2.209333204443945,
      "language_loss": 0.75877368,
      "learning_rate": 3.9615895547103865e-06,
      "loss": 0.7816301,
      "num_input_tokens_seen": 16116150,
      "step": 754,
      "time_per_iteration": 2.815030574798584
    },
    {
      "auxiliary_loss_clip": 0.01218613,
      "auxiliary_loss_mlp": 0.0103846,
      "balance_loss_clip": 1.06370175,
      "balance_loss_mlp": 1.02630639,
      "epoch": 0.09078338243251367,
      "flos": 29168729550720.0,
      "grad_norm": 2.1719525401467727,
      "language_loss": 0.77766424,
      "learning_rate": 3.961437473295193e-06,
      "loss": 0.80023497,
      "num_input_tokens_seen": 16136295,
      "step": 755,
      "time_per_iteration": 2.8149194717407227
    },
    {
      "auxiliary_loss_clip": 0.01168856,
      "auxiliary_loss_mlp": 0.01047662,
      "balance_loss_clip": 1.05044293,
      "balance_loss_mlp": 1.03588414,
      "epoch": 0.09090362532315277,
      "flos": 21907699977600.0,
      "grad_norm": 3.0984925705992983,
      "language_loss": 0.72449785,
      "learning_rate": 3.961285094331495e-06,
      "loss": 0.74666309,
      "num_input_tokens_seen": 16154210,
      "step": 756,
      "time_per_iteration": 2.9124462604522705
    },
    {
      "auxiliary_loss_clip": 0.01247772,
      "auxiliary_loss_mlp": 0.01043139,
      "balance_loss_clip": 1.06688213,
      "balance_loss_mlp": 1.03135478,
      "epoch": 0.09102386821379185,
      "flos": 27344503480320.0,
      "grad_norm": 3.566729164137194,
      "language_loss": 0.85911244,
      "learning_rate": 3.961132417842406e-06,
      "loss": 0.88202155,
      "num_input_tokens_seen": 16173995,
      "step": 757,
      "time_per_iteration": 2.749868392944336
    },
    {
      "auxiliary_loss_clip": 0.01224946,
      "auxiliary_loss_mlp": 0.01037942,
      "balance_loss_clip": 1.06439745,
      "balance_loss_mlp": 1.02626491,
      "epoch": 0.09114411110443095,
      "flos": 20813501923200.0,
      "grad_norm": 3.0828914787017303,
      "language_loss": 0.75058854,
      "learning_rate": 3.960979443851089e-06,
      "loss": 0.77321744,
      "num_input_tokens_seen": 16191020,
      "step": 758,
      "time_per_iteration": 2.7564122676849365
    },
    {
      "auxiliary_loss_clip": 0.01215906,
      "auxiliary_loss_mlp": 0.01036981,
      "balance_loss_clip": 1.06428134,
      "balance_loss_mlp": 1.02511346,
      "epoch": 0.09126435399507005,
      "flos": 26145949438080.0,
      "grad_norm": 1.6708026147587776,
      "language_loss": 0.79192483,
      "learning_rate": 3.96082617238075e-06,
      "loss": 0.81445366,
      "num_input_tokens_seen": 16213645,
      "step": 759,
      "time_per_iteration": 2.876147985458374
    },
    {
      "auxiliary_loss_clip": 0.0121759,
      "auxiliary_loss_mlp": 0.01038623,
      "balance_loss_clip": 1.06346011,
      "balance_loss_mlp": 1.02710152,
      "epoch": 0.09138459688570913,
      "flos": 24388911757440.0,
      "grad_norm": 2.3309888049117022,
      "language_loss": 0.7968297,
      "learning_rate": 3.960672603454639e-06,
      "loss": 0.81939179,
      "num_input_tokens_seen": 16233625,
      "step": 760,
      "time_per_iteration": 2.8499250411987305
    },
    {
      "auxiliary_loss_clip": 0.01224839,
      "auxiliary_loss_mlp": 0.01039615,
      "balance_loss_clip": 1.06297278,
      "balance_loss_mlp": 1.02799821,
      "epoch": 0.09150483977634823,
      "flos": 21032664756480.0,
      "grad_norm": 4.545074128782622,
      "language_loss": 0.76892692,
      "learning_rate": 3.960518737096054e-06,
      "loss": 0.79157138,
      "num_input_tokens_seen": 16253255,
      "step": 761,
      "time_per_iteration": 2.741894006729126
    },
    {
      "auxiliary_loss_clip": 0.0123448,
      "auxiliary_loss_mlp": 0.01039002,
      "balance_loss_clip": 1.06576443,
      "balance_loss_mlp": 1.02778423,
      "epoch": 0.09162508266698731,
      "flos": 22856998567680.0,
      "grad_norm": 2.6640867552443774,
      "language_loss": 0.73249102,
      "learning_rate": 3.960364573328334e-06,
      "loss": 0.75522578,
      "num_input_tokens_seen": 16272580,
      "step": 762,
      "time_per_iteration": 2.6220500469207764
    },
    {
      "auxiliary_loss_clip": 0.01202868,
      "auxiliary_loss_mlp": 0.01036492,
      "balance_loss_clip": 1.05909216,
      "balance_loss_mlp": 1.02445769,
      "epoch": 0.0917453255576264,
      "flos": 21724411852800.0,
      "grad_norm": 1.9967282260163974,
      "language_loss": 0.88695788,
      "learning_rate": 3.9602101121748675e-06,
      "loss": 0.90935147,
      "num_input_tokens_seen": 16293075,
      "step": 763,
      "time_per_iteration": 2.7659759521484375
    },
    {
      "auxiliary_loss_clip": 0.01215631,
      "auxiliary_loss_mlp": 0.01037158,
      "balance_loss_clip": 1.06386256,
      "balance_loss_mlp": 1.02591634,
      "epoch": 0.0918655684482655,
      "flos": 14609215497600.0,
      "grad_norm": 2.093474670557306,
      "language_loss": 0.72849476,
      "learning_rate": 3.960055353659085e-06,
      "loss": 0.7510227,
      "num_input_tokens_seen": 16310185,
      "step": 764,
      "time_per_iteration": 2.725687026977539
    },
    {
      "auxiliary_loss_clip": 0.01201149,
      "auxiliary_loss_mlp": 0.01042913,
      "balance_loss_clip": 1.06054497,
      "balance_loss_mlp": 1.03124249,
      "epoch": 0.09198581133890459,
      "flos": 23435016226560.0,
      "grad_norm": 2.264598638857187,
      "language_loss": 0.83863842,
      "learning_rate": 3.959900297804465e-06,
      "loss": 0.86107904,
      "num_input_tokens_seen": 16330355,
      "step": 765,
      "time_per_iteration": 2.747556686401367
    },
    {
      "auxiliary_loss_clip": 0.01203923,
      "auxiliary_loss_mlp": 0.01036747,
      "balance_loss_clip": 1.06095529,
      "balance_loss_mlp": 1.02587509,
      "epoch": 0.09210605422954368,
      "flos": 16795887753600.0,
      "grad_norm": 2.1479671236386606,
      "language_loss": 0.77162468,
      "learning_rate": 3.9597449446345276e-06,
      "loss": 0.79403138,
      "num_input_tokens_seen": 16347600,
      "step": 766,
      "time_per_iteration": 2.710190534591675
    },
    {
      "auxiliary_loss_clip": 0.01203307,
      "auxiliary_loss_mlp": 0.01040228,
      "balance_loss_clip": 1.05886936,
      "balance_loss_mlp": 1.0286293,
      "epoch": 0.09222629712018277,
      "flos": 22674249146880.0,
      "grad_norm": 3.4427678081549273,
      "language_loss": 0.83120906,
      "learning_rate": 3.95958929417284e-06,
      "loss": 0.85364437,
      "num_input_tokens_seen": 16365755,
      "step": 767,
      "time_per_iteration": 2.697509527206421
    },
    {
      "auxiliary_loss_clip": 0.01144877,
      "auxiliary_loss_mlp": 0.01007611,
      "balance_loss_clip": 1.03667462,
      "balance_loss_mlp": 1.00312924,
      "epoch": 0.09234654001082186,
      "flos": 69976756327680.0,
      "grad_norm": 0.7338713219237929,
      "language_loss": 0.58783251,
      "learning_rate": 3.9594333464430145e-06,
      "loss": 0.60935742,
      "num_input_tokens_seen": 16435245,
      "step": 768,
      "time_per_iteration": 3.3781991004943848
    },
    {
      "auxiliary_loss_clip": 0.01143158,
      "auxiliary_loss_mlp": 0.01043108,
      "balance_loss_clip": 1.05021071,
      "balance_loss_mlp": 1.03204584,
      "epoch": 0.09246678290146094,
      "flos": 20011437181440.0,
      "grad_norm": 2.004930607317738,
      "language_loss": 0.88396406,
      "learning_rate": 3.959277101468709e-06,
      "loss": 0.90582669,
      "num_input_tokens_seen": 16454795,
      "step": 769,
      "time_per_iteration": 2.810681104660034
    },
    {
      "auxiliary_loss_clip": 0.01202836,
      "auxiliary_loss_mlp": 0.01044076,
      "balance_loss_clip": 1.05888081,
      "balance_loss_mlp": 1.03248894,
      "epoch": 0.09258702579210004,
      "flos": 17747448900480.0,
      "grad_norm": 2.9483833178776804,
      "language_loss": 0.78736341,
      "learning_rate": 3.959120559273624e-06,
      "loss": 0.80983251,
      "num_input_tokens_seen": 16472580,
      "step": 770,
      "time_per_iteration": 2.6587328910827637
    },
    {
      "auxiliary_loss_clip": 0.01203467,
      "auxiliary_loss_mlp": 0.01048943,
      "balance_loss_clip": 1.06299174,
      "balance_loss_mlp": 1.03713489,
      "epoch": 0.09270726868273914,
      "flos": 20886544229760.0,
      "grad_norm": 1.9831195196168443,
      "language_loss": 0.83472824,
      "learning_rate": 3.958963719881509e-06,
      "loss": 0.85725236,
      "num_input_tokens_seen": 16490670,
      "step": 771,
      "time_per_iteration": 3.740633010864258
    },
    {
      "auxiliary_loss_clip": 0.012337,
      "auxiliary_loss_mlp": 0.01046347,
      "balance_loss_clip": 1.06759048,
      "balance_loss_mlp": 1.03483748,
      "epoch": 0.09282751157337822,
      "flos": 17015697031680.0,
      "grad_norm": 2.1214620675136886,
      "language_loss": 0.93888104,
      "learning_rate": 3.958806583316154e-06,
      "loss": 0.96168149,
      "num_input_tokens_seen": 16508640,
      "step": 772,
      "time_per_iteration": 2.6777212619781494
    },
    {
      "auxiliary_loss_clip": 0.01251112,
      "auxiliary_loss_mlp": 0.01038309,
      "balance_loss_clip": 1.06988513,
      "balance_loss_mlp": 1.02678728,
      "epoch": 0.09294775446401732,
      "flos": 32523647748480.0,
      "grad_norm": 1.7787970787071645,
      "language_loss": 0.78773671,
      "learning_rate": 3.9586491496013985e-06,
      "loss": 0.81063092,
      "num_input_tokens_seen": 16531035,
      "step": 773,
      "time_per_iteration": 3.7450931072235107
    },
    {
      "auxiliary_loss_clip": 0.01238858,
      "auxiliary_loss_mlp": 0.01036293,
      "balance_loss_clip": 1.06909454,
      "balance_loss_mlp": 1.0245266,
      "epoch": 0.0930679973546564,
      "flos": 18259750627200.0,
      "grad_norm": 2.2994100251818277,
      "language_loss": 0.82940274,
      "learning_rate": 3.958491418761124e-06,
      "loss": 0.85215425,
      "num_input_tokens_seen": 16548605,
      "step": 774,
      "time_per_iteration": 2.754110097885132
    },
    {
      "auxiliary_loss_clip": 0.01216676,
      "auxiliary_loss_mlp": 0.01039747,
      "balance_loss_clip": 1.06003261,
      "balance_loss_mlp": 1.02847612,
      "epoch": 0.0931882402452955,
      "flos": 21099745405440.0,
      "grad_norm": 7.159060694734685,
      "language_loss": 0.7229321,
      "learning_rate": 3.958333390819258e-06,
      "loss": 0.74549633,
      "num_input_tokens_seen": 16565535,
      "step": 775,
      "time_per_iteration": 3.596853017807007
    },
    {
      "auxiliary_loss_clip": 0.01247353,
      "auxiliary_loss_mlp": 0.01050704,
      "balance_loss_clip": 1.06755412,
      "balance_loss_mlp": 1.03831232,
      "epoch": 0.0933084831359346,
      "flos": 24207275658240.0,
      "grad_norm": 2.43999179784978,
      "language_loss": 0.80247557,
      "learning_rate": 3.9581750657997754e-06,
      "loss": 0.8254562,
      "num_input_tokens_seen": 16584900,
      "step": 776,
      "time_per_iteration": 3.526524782180786
    },
    {
      "auxiliary_loss_clip": 0.01215438,
      "auxiliary_loss_mlp": 0.01040537,
      "balance_loss_clip": 1.06233335,
      "balance_loss_mlp": 1.0284189,
      "epoch": 0.09342872602657368,
      "flos": 25480272637440.0,
      "grad_norm": 1.7134016152099005,
      "language_loss": 0.89543742,
      "learning_rate": 3.95801644372669e-06,
      "loss": 0.91799712,
      "num_input_tokens_seen": 16604805,
      "step": 777,
      "time_per_iteration": 2.749793529510498
    },
    {
      "auxiliary_loss_clip": 0.01219697,
      "auxiliary_loss_mlp": 0.01042278,
      "balance_loss_clip": 1.06006765,
      "balance_loss_mlp": 1.03087544,
      "epoch": 0.09354896891721277,
      "flos": 23149060053120.0,
      "grad_norm": 2.859053736978666,
      "language_loss": 0.84294373,
      "learning_rate": 3.957857524624068e-06,
      "loss": 0.86556351,
      "num_input_tokens_seen": 16623685,
      "step": 778,
      "time_per_iteration": 2.721316337585449
    },
    {
      "auxiliary_loss_clip": 0.0121424,
      "auxiliary_loss_mlp": 0.01042161,
      "balance_loss_clip": 1.06113958,
      "balance_loss_mlp": 1.03113365,
      "epoch": 0.09366921180785186,
      "flos": 24279563779200.0,
      "grad_norm": 1.7705659548303052,
      "language_loss": 0.89853454,
      "learning_rate": 3.957698308516016e-06,
      "loss": 0.92109859,
      "num_input_tokens_seen": 16644985,
      "step": 779,
      "time_per_iteration": 2.7251243591308594
    },
    {
      "auxiliary_loss_clip": 0.01229537,
      "auxiliary_loss_mlp": 0.00764929,
      "balance_loss_clip": 1.06850314,
      "balance_loss_mlp": 1.00071621,
      "epoch": 0.09378945469849095,
      "flos": 18730036419840.0,
      "grad_norm": 2.0163891010333943,
      "language_loss": 0.8255415,
      "learning_rate": 3.957538795426688e-06,
      "loss": 0.84548616,
      "num_input_tokens_seen": 16662410,
      "step": 780,
      "time_per_iteration": 2.678981065750122
    },
    {
      "auxiliary_loss_clip": 0.01218288,
      "auxiliary_loss_mlp": 0.01052171,
      "balance_loss_clip": 1.06231105,
      "balance_loss_mlp": 1.04058409,
      "epoch": 0.09390969758913004,
      "flos": 23218834222080.0,
      "grad_norm": 7.087770472743251,
      "language_loss": 0.77223176,
      "learning_rate": 3.9573789853802804e-06,
      "loss": 0.79493642,
      "num_input_tokens_seen": 16680885,
      "step": 781,
      "time_per_iteration": 2.7364144325256348
    },
    {
      "auxiliary_loss_clip": 0.012151,
      "auxiliary_loss_mlp": 0.00765403,
      "balance_loss_clip": 1.06462812,
      "balance_loss_mlp": 1.00065005,
      "epoch": 0.09402994047976913,
      "flos": 19646728439040.0,
      "grad_norm": 2.0635957591956986,
      "language_loss": 0.74735188,
      "learning_rate": 3.957218878401037e-06,
      "loss": 0.76715696,
      "num_input_tokens_seen": 16699375,
      "step": 782,
      "time_per_iteration": 2.7537899017333984
    },
    {
      "auxiliary_loss_clip": 0.01250834,
      "auxiliary_loss_mlp": 0.01041313,
      "balance_loss_clip": 1.06962872,
      "balance_loss_mlp": 1.0292908,
      "epoch": 0.09415018337040823,
      "flos": 29420463041280.0,
      "grad_norm": 1.8494498680723792,
      "language_loss": 0.89577246,
      "learning_rate": 3.957058474513246e-06,
      "loss": 0.91869396,
      "num_input_tokens_seen": 16719230,
      "step": 783,
      "time_per_iteration": 2.7943217754364014
    },
    {
      "auxiliary_loss_clip": 0.01233295,
      "auxiliary_loss_mlp": 0.01041839,
      "balance_loss_clip": 1.06917238,
      "balance_loss_mlp": 1.03087759,
      "epoch": 0.09427042626104731,
      "flos": 24572092141440.0,
      "grad_norm": 1.898341520435001,
      "language_loss": 0.78466845,
      "learning_rate": 3.956897773741241e-06,
      "loss": 0.80741984,
      "num_input_tokens_seen": 16738220,
      "step": 784,
      "time_per_iteration": 2.7535488605499268
    },
    {
      "auxiliary_loss_clip": 0.01206111,
      "auxiliary_loss_mlp": 0.01036706,
      "balance_loss_clip": 1.0604496,
      "balance_loss_mlp": 1.02547002,
      "epoch": 0.09439066915168641,
      "flos": 26359581576960.0,
      "grad_norm": 1.789354898587762,
      "language_loss": 0.71714216,
      "learning_rate": 3.956736776109398e-06,
      "loss": 0.73957038,
      "num_input_tokens_seen": 16759395,
      "step": 785,
      "time_per_iteration": 2.810006618499756
    },
    {
      "auxiliary_loss_clip": 0.01220177,
      "auxiliary_loss_mlp": 0.00765402,
      "balance_loss_clip": 1.06088638,
      "balance_loss_mlp": 1.00062799,
      "epoch": 0.09451091204232549,
      "flos": 19427278296960.0,
      "grad_norm": 2.258281349081871,
      "language_loss": 0.83675635,
      "learning_rate": 3.956575481642143e-06,
      "loss": 0.85661209,
      "num_input_tokens_seen": 16778285,
      "step": 786,
      "time_per_iteration": 2.830056667327881
    },
    {
      "auxiliary_loss_clip": 0.01183777,
      "auxiliary_loss_mlp": 0.01042451,
      "balance_loss_clip": 1.05558836,
      "balance_loss_mlp": 1.03118622,
      "epoch": 0.09463115493296459,
      "flos": 25368051571200.0,
      "grad_norm": 2.368885794193833,
      "language_loss": 0.7503497,
      "learning_rate": 3.956413890363943e-06,
      "loss": 0.77261198,
      "num_input_tokens_seen": 16795265,
      "step": 787,
      "time_per_iteration": 2.80875563621521
    },
    {
      "auxiliary_loss_clip": 0.01235395,
      "auxiliary_loss_mlp": 0.01043599,
      "balance_loss_clip": 1.06837058,
      "balance_loss_mlp": 1.03161836,
      "epoch": 0.09475139782360369,
      "flos": 10123254869760.0,
      "grad_norm": 2.0645299508988604,
      "language_loss": 0.81528306,
      "learning_rate": 3.956252002299312e-06,
      "loss": 0.83807302,
      "num_input_tokens_seen": 16811165,
      "step": 788,
      "time_per_iteration": 2.703986644744873
    },
    {
      "auxiliary_loss_clip": 0.01245473,
      "auxiliary_loss_mlp": 0.01036705,
      "balance_loss_clip": 1.06653619,
      "balance_loss_mlp": 1.02476072,
      "epoch": 0.09487164071424277,
      "flos": 17231088936960.0,
      "grad_norm": 2.0760257031356666,
      "language_loss": 0.90921116,
      "learning_rate": 3.956089817472807e-06,
      "loss": 0.93203294,
      "num_input_tokens_seen": 16828470,
      "step": 789,
      "time_per_iteration": 2.618983507156372
    },
    {
      "auxiliary_loss_clip": 0.01215494,
      "auxiliary_loss_mlp": 0.01038287,
      "balance_loss_clip": 1.06625104,
      "balance_loss_mlp": 1.02655697,
      "epoch": 0.09499188360488187,
      "flos": 30849564528000.0,
      "grad_norm": 2.9811318860132716,
      "language_loss": 0.85540891,
      "learning_rate": 3.955927335909032e-06,
      "loss": 0.87794673,
      "num_input_tokens_seen": 16851680,
      "step": 790,
      "time_per_iteration": 2.7714617252349854
    },
    {
      "auxiliary_loss_clip": 0.01184649,
      "auxiliary_loss_mlp": 0.01035859,
      "balance_loss_clip": 1.06306624,
      "balance_loss_mlp": 1.02422452,
      "epoch": 0.09511212649552095,
      "flos": 29351694453120.0,
      "grad_norm": 2.2241378870695274,
      "language_loss": 0.76105446,
      "learning_rate": 3.955764557632634e-06,
      "loss": 0.78325951,
      "num_input_tokens_seen": 16871490,
      "step": 791,
      "time_per_iteration": 2.813549518585205
    },
    {
      "auxiliary_loss_clip": 0.01207862,
      "auxiliary_loss_mlp": 0.01047009,
      "balance_loss_clip": 1.05999959,
      "balance_loss_mlp": 1.03525519,
      "epoch": 0.09523236938616005,
      "flos": 10378687461120.0,
      "grad_norm": 4.055142254576654,
      "language_loss": 0.94861662,
      "learning_rate": 3.955601482668309e-06,
      "loss": 0.9711653,
      "num_input_tokens_seen": 16889350,
      "step": 792,
      "time_per_iteration": 2.697558641433716
    },
    {
      "auxiliary_loss_clip": 0.01178652,
      "auxiliary_loss_mlp": 0.0104354,
      "balance_loss_clip": 1.0539726,
      "balance_loss_mlp": 1.03188729,
      "epoch": 0.09535261227679913,
      "flos": 19061815368960.0,
      "grad_norm": 2.6248505363224317,
      "language_loss": 0.88497108,
      "learning_rate": 3.955438111040794e-06,
      "loss": 0.90719301,
      "num_input_tokens_seen": 16907625,
      "step": 793,
      "time_per_iteration": 2.7049691677093506
    },
    {
      "auxiliary_loss_clip": 0.01178684,
      "auxiliary_loss_mlp": 0.0103567,
      "balance_loss_clip": 1.05598283,
      "balance_loss_mlp": 1.02481031,
      "epoch": 0.09547285516743823,
      "flos": 20922993555840.0,
      "grad_norm": 2.164634627617604,
      "language_loss": 0.79994118,
      "learning_rate": 3.955274442774873e-06,
      "loss": 0.82208472,
      "num_input_tokens_seen": 16926205,
      "step": 794,
      "time_per_iteration": 2.78458833694458
    },
    {
      "auxiliary_loss_clip": 0.01230276,
      "auxiliary_loss_mlp": 0.01049868,
      "balance_loss_clip": 1.06192338,
      "balance_loss_mlp": 1.0377264,
      "epoch": 0.09559309805807732,
      "flos": 30154405639680.0,
      "grad_norm": 2.197467611193499,
      "language_loss": 0.70397079,
      "learning_rate": 3.9551104778953725e-06,
      "loss": 0.72677219,
      "num_input_tokens_seen": 16946500,
      "step": 795,
      "time_per_iteration": 2.7491538524627686
    },
    {
      "auxiliary_loss_clip": 0.01202322,
      "auxiliary_loss_mlp": 0.01035862,
      "balance_loss_clip": 1.05913424,
      "balance_loss_mlp": 1.02359509,
      "epoch": 0.0957133409487164,
      "flos": 21066743784960.0,
      "grad_norm": 1.8683531426859925,
      "language_loss": 0.85264462,
      "learning_rate": 3.954946216427167e-06,
      "loss": 0.87502646,
      "num_input_tokens_seen": 16966960,
      "step": 796,
      "time_per_iteration": 2.7400617599487305
    },
    {
      "auxiliary_loss_clip": 0.0111456,
      "auxiliary_loss_mlp": 0.01010103,
      "balance_loss_clip": 1.03376174,
      "balance_loss_mlp": 1.0057162,
      "epoch": 0.0958335838393555,
      "flos": 71297979315840.0,
      "grad_norm": 0.8876874758579827,
      "language_loss": 0.61579883,
      "learning_rate": 3.954781658395176e-06,
      "loss": 0.63704544,
      "num_input_tokens_seen": 17023215,
      "step": 797,
      "time_per_iteration": 4.097166061401367
    },
    {
      "auxiliary_loss_clip": 0.01220723,
      "auxiliary_loss_mlp": 0.01046202,
      "balance_loss_clip": 1.06174695,
      "balance_loss_mlp": 1.03333926,
      "epoch": 0.09595382672999458,
      "flos": 21872974504320.0,
      "grad_norm": 2.078387556535874,
      "language_loss": 0.92263746,
      "learning_rate": 3.95461680382436e-06,
      "loss": 0.94530666,
      "num_input_tokens_seen": 17042140,
      "step": 798,
      "time_per_iteration": 2.8010916709899902
    },
    {
      "auxiliary_loss_clip": 0.01236342,
      "auxiliary_loss_mlp": 0.01046756,
      "balance_loss_clip": 1.06696236,
      "balance_loss_mlp": 1.03447747,
      "epoch": 0.09607406962063368,
      "flos": 18695562341760.0,
      "grad_norm": 73.69063321422855,
      "language_loss": 0.86297911,
      "learning_rate": 3.9544516527397295e-06,
      "loss": 0.88581014,
      "num_input_tokens_seen": 17058490,
      "step": 799,
      "time_per_iteration": 3.8383450508117676
    },
    {
      "auxiliary_loss_clip": 0.01206766,
      "auxiliary_loss_mlp": 0.01043904,
      "balance_loss_clip": 1.06323647,
      "balance_loss_mlp": 1.03234673,
      "epoch": 0.09619431251127276,
      "flos": 22568456615040.0,
      "grad_norm": 1.8339756391731807,
      "language_loss": 0.80704331,
      "learning_rate": 3.954286205166338e-06,
      "loss": 0.82955003,
      "num_input_tokens_seen": 17079655,
      "step": 800,
      "time_per_iteration": 2.8015217781066895
    },
    {
      "auxiliary_loss_clip": 0.0123498,
      "auxiliary_loss_mlp": 0.01040274,
      "balance_loss_clip": 1.06909215,
      "balance_loss_mlp": 1.02797771,
      "epoch": 0.09631455540191186,
      "flos": 14246230608000.0,
      "grad_norm": 2.405197594283106,
      "language_loss": 0.83736694,
      "learning_rate": 3.954120461129282e-06,
      "loss": 0.86011946,
      "num_input_tokens_seen": 17097065,
      "step": 801,
      "time_per_iteration": 4.494358777999878
    },
    {
      "auxiliary_loss_clip": 0.01254462,
      "auxiliary_loss_mlp": 0.0105516,
      "balance_loss_clip": 1.07187188,
      "balance_loss_mlp": 1.04380572,
      "epoch": 0.09643479829255096,
      "flos": 20740387789440.0,
      "grad_norm": 2.148884601645676,
      "language_loss": 0.83681571,
      "learning_rate": 3.953954420653706e-06,
      "loss": 0.85991192,
      "num_input_tokens_seen": 17114090,
      "step": 802,
      "time_per_iteration": 2.81734561920166
    },
    {
      "auxiliary_loss_clip": 0.01230834,
      "auxiliary_loss_mlp": 0.010424,
      "balance_loss_clip": 1.06610632,
      "balance_loss_mlp": 1.03010333,
      "epoch": 0.09655504118319004,
      "flos": 24420476833920.0,
      "grad_norm": 1.985828667944063,
      "language_loss": 0.88169736,
      "learning_rate": 3.953788083764798e-06,
      "loss": 0.90442973,
      "num_input_tokens_seen": 17133325,
      "step": 803,
      "time_per_iteration": 2.73512864112854
    },
    {
      "auxiliary_loss_clip": 0.01181709,
      "auxiliary_loss_mlp": 0.01042086,
      "balance_loss_clip": 1.05767536,
      "balance_loss_mlp": 1.03055251,
      "epoch": 0.09667528407382914,
      "flos": 18441961344000.0,
      "grad_norm": 2.4755695405416045,
      "language_loss": 0.92424572,
      "learning_rate": 3.953621450487792e-06,
      "loss": 0.94648367,
      "num_input_tokens_seen": 17151945,
      "step": 804,
      "time_per_iteration": 2.696316719055176
    },
    {
      "auxiliary_loss_clip": 0.01149461,
      "auxiliary_loss_mlp": 0.0100719,
      "balance_loss_clip": 1.03321505,
      "balance_loss_mlp": 1.0029223,
      "epoch": 0.09679552696446822,
      "flos": 70816455544320.0,
      "grad_norm": 0.8432428610399461,
      "language_loss": 0.61220598,
      "learning_rate": 3.953454520847964e-06,
      "loss": 0.63377249,
      "num_input_tokens_seen": 17216790,
      "step": 805,
      "time_per_iteration": 3.343320846557617
    },
    {
      "auxiliary_loss_clip": 0.0120862,
      "auxiliary_loss_mlp": 0.01044766,
      "balance_loss_clip": 1.06190431,
      "balance_loss_mlp": 1.03257108,
      "epoch": 0.09691576985510732,
      "flos": 21945514020480.0,
      "grad_norm": 3.0302582024622073,
      "language_loss": 0.73809743,
      "learning_rate": 3.9532872948706395e-06,
      "loss": 0.76063132,
      "num_input_tokens_seen": 17236285,
      "step": 806,
      "time_per_iteration": 2.727023124694824
    },
    {
      "auxiliary_loss_clip": 0.01217072,
      "auxiliary_loss_mlp": 0.01044108,
      "balance_loss_clip": 1.06198895,
      "balance_loss_mlp": 1.03218126,
      "epoch": 0.09703601274574641,
      "flos": 17965211103360.0,
      "grad_norm": 5.335472179863121,
      "language_loss": 0.82614923,
      "learning_rate": 3.9531197725811845e-06,
      "loss": 0.84876096,
      "num_input_tokens_seen": 17251670,
      "step": 807,
      "time_per_iteration": 2.717808246612549
    },
    {
      "auxiliary_loss_clip": 0.01246409,
      "auxiliary_loss_mlp": 0.01041001,
      "balance_loss_clip": 1.06781459,
      "balance_loss_mlp": 1.02967644,
      "epoch": 0.0971562556363855,
      "flos": 22162162901760.0,
      "grad_norm": 2.208728774567375,
      "language_loss": 0.87726474,
      "learning_rate": 3.952951954005013e-06,
      "loss": 0.90013891,
      "num_input_tokens_seen": 17271355,
      "step": 808,
      "time_per_iteration": 2.6597845554351807
    },
    {
      "auxiliary_loss_clip": 0.01213956,
      "auxiliary_loss_mlp": 0.01040226,
      "balance_loss_clip": 1.05864966,
      "balance_loss_mlp": 1.02823305,
      "epoch": 0.0972764985270246,
      "flos": 25848716394240.0,
      "grad_norm": 1.76077687542616,
      "language_loss": 0.84736747,
      "learning_rate": 3.952783839167584e-06,
      "loss": 0.86990929,
      "num_input_tokens_seen": 17291400,
      "step": 809,
      "time_per_iteration": 2.811009168624878
    },
    {
      "auxiliary_loss_clip": 0.01229025,
      "auxiliary_loss_mlp": 0.01041037,
      "balance_loss_clip": 1.06275785,
      "balance_loss_mlp": 1.02952099,
      "epoch": 0.09739674141766368,
      "flos": 20339373375360.0,
      "grad_norm": 2.9811076377233054,
      "language_loss": 0.74930573,
      "learning_rate": 3.952615428094398e-06,
      "loss": 0.77200639,
      "num_input_tokens_seen": 17310920,
      "step": 810,
      "time_per_iteration": 2.726468801498413
    },
    {
      "auxiliary_loss_clip": 0.0117841,
      "auxiliary_loss_mlp": 0.01047794,
      "balance_loss_clip": 1.05529451,
      "balance_loss_mlp": 1.03624308,
      "epoch": 0.09751698430830277,
      "flos": 15743059188480.0,
      "grad_norm": 1.9658268896634314,
      "language_loss": 0.73651379,
      "learning_rate": 3.952446720811004e-06,
      "loss": 0.75877583,
      "num_input_tokens_seen": 17329245,
      "step": 811,
      "time_per_iteration": 2.7308080196380615
    },
    {
      "auxiliary_loss_clip": 0.01102781,
      "auxiliary_loss_mlp": 0.01007787,
      "balance_loss_clip": 1.02677464,
      "balance_loss_mlp": 1.0030663,
      "epoch": 0.09763722719894186,
      "flos": 63716806800000.0,
      "grad_norm": 0.8385906141165206,
      "language_loss": 0.63648403,
      "learning_rate": 3.952277717342995e-06,
      "loss": 0.65758967,
      "num_input_tokens_seen": 17395680,
      "step": 812,
      "time_per_iteration": 3.4346439838409424
    },
    {
      "auxiliary_loss_clip": 0.01221149,
      "auxiliary_loss_mlp": 0.01045174,
      "balance_loss_clip": 1.06221056,
      "balance_loss_mlp": 1.03172755,
      "epoch": 0.09775747008958095,
      "flos": 22090916275200.0,
      "grad_norm": 1.9810125951662734,
      "language_loss": 0.85455412,
      "learning_rate": 3.952108417716009e-06,
      "loss": 0.87721735,
      "num_input_tokens_seen": 17415135,
      "step": 813,
      "time_per_iteration": 2.6561295986175537
    },
    {
      "auxiliary_loss_clip": 0.01237565,
      "auxiliary_loss_mlp": 0.01049593,
      "balance_loss_clip": 1.06900501,
      "balance_loss_mlp": 1.03667629,
      "epoch": 0.09787771298022005,
      "flos": 21286050272640.0,
      "grad_norm": 1.8005346821377377,
      "language_loss": 0.84861517,
      "learning_rate": 3.951938821955727e-06,
      "loss": 0.87148678,
      "num_input_tokens_seen": 17434535,
      "step": 814,
      "time_per_iteration": 2.6773321628570557
    },
    {
      "auxiliary_loss_clip": 0.01217216,
      "auxiliary_loss_mlp": 0.01038581,
      "balance_loss_clip": 1.06507885,
      "balance_loss_mlp": 1.0263685,
      "epoch": 0.09799795587085913,
      "flos": 22054574689920.0,
      "grad_norm": 1.6538343901511334,
      "language_loss": 0.76672119,
      "learning_rate": 3.9517689300878786e-06,
      "loss": 0.78927916,
      "num_input_tokens_seen": 17454270,
      "step": 815,
      "time_per_iteration": 2.755664825439453
    },
    {
      "auxiliary_loss_clip": 0.01248974,
      "auxiliary_loss_mlp": 0.0104046,
      "balance_loss_clip": 1.06616068,
      "balance_loss_mlp": 1.02866387,
      "epoch": 0.09811819876149823,
      "flos": 22163743100160.0,
      "grad_norm": 2.8304792893161532,
      "language_loss": 0.78690398,
      "learning_rate": 3.951598742138236e-06,
      "loss": 0.8097983,
      "num_input_tokens_seen": 17472995,
      "step": 816,
      "time_per_iteration": 2.638741970062256
    },
    {
      "auxiliary_loss_clip": 0.0122118,
      "auxiliary_loss_mlp": 0.01040295,
      "balance_loss_clip": 1.05907452,
      "balance_loss_mlp": 1.02880347,
      "epoch": 0.09823844165213731,
      "flos": 22231111057920.0,
      "grad_norm": 2.3505121621351783,
      "language_loss": 0.7980029,
      "learning_rate": 3.951428258132615e-06,
      "loss": 0.82061768,
      "num_input_tokens_seen": 17491115,
      "step": 817,
      "time_per_iteration": 2.700972318649292
    },
    {
      "auxiliary_loss_clip": 0.01219376,
      "auxiliary_loss_mlp": 0.01046193,
      "balance_loss_clip": 1.06356478,
      "balance_loss_mlp": 1.03339612,
      "epoch": 0.09835868454277641,
      "flos": 22487728798080.0,
      "grad_norm": 2.1583629528398163,
      "language_loss": 0.84446859,
      "learning_rate": 3.951257478096879e-06,
      "loss": 0.86712432,
      "num_input_tokens_seen": 17509480,
      "step": 818,
      "time_per_iteration": 2.696350336074829
    },
    {
      "auxiliary_loss_clip": 0.01216348,
      "auxiliary_loss_mlp": 0.00766817,
      "balance_loss_clip": 1.06199932,
      "balance_loss_mlp": 1.00056291,
      "epoch": 0.0984789274334155,
      "flos": 16362554077440.0,
      "grad_norm": 3.0322655562052123,
      "language_loss": 0.68962365,
      "learning_rate": 3.951086402056936e-06,
      "loss": 0.70945525,
      "num_input_tokens_seen": 17524080,
      "step": 819,
      "time_per_iteration": 2.6997785568237305
    },
    {
      "auxiliary_loss_clip": 0.01156669,
      "auxiliary_loss_mlp": 0.00766672,
      "balance_loss_clip": 1.06087875,
      "balance_loss_mlp": 1.0006578,
      "epoch": 0.09859917032405459,
      "flos": 24243545416320.0,
      "grad_norm": 1.6867696278171282,
      "language_loss": 0.83701909,
      "learning_rate": 3.950915030038735e-06,
      "loss": 0.85625255,
      "num_input_tokens_seen": 17543875,
      "step": 820,
      "time_per_iteration": 2.8319664001464844
    },
    {
      "auxiliary_loss_clip": 0.01231705,
      "auxiliary_loss_mlp": 0.01042469,
      "balance_loss_clip": 1.06715667,
      "balance_loss_mlp": 1.03054237,
      "epoch": 0.09871941321469369,
      "flos": 17420195064960.0,
      "grad_norm": 2.57055421419102,
      "language_loss": 0.83507842,
      "learning_rate": 3.9507433620682765e-06,
      "loss": 0.85782015,
      "num_input_tokens_seen": 17560810,
      "step": 821,
      "time_per_iteration": 2.699984073638916
    },
    {
      "auxiliary_loss_clip": 0.01198573,
      "auxiliary_loss_mlp": 0.01051772,
      "balance_loss_clip": 1.0578922,
      "balance_loss_mlp": 1.0386529,
      "epoch": 0.09883965610533277,
      "flos": 28477341590400.0,
      "grad_norm": 1.7482413551045455,
      "language_loss": 0.88257635,
      "learning_rate": 3.9505713981716e-06,
      "loss": 0.90507984,
      "num_input_tokens_seen": 17583640,
      "step": 822,
      "time_per_iteration": 2.754671096801758
    },
    {
      "auxiliary_loss_clip": 0.01216927,
      "auxiliary_loss_mlp": 0.01040434,
      "balance_loss_clip": 1.06484795,
      "balance_loss_mlp": 1.02879381,
      "epoch": 0.09895989899597187,
      "flos": 23693932437120.0,
      "grad_norm": 1.8933121638813315,
      "language_loss": 0.81398875,
      "learning_rate": 3.950399138374795e-06,
      "loss": 0.8365624,
      "num_input_tokens_seen": 17602720,
      "step": 823,
      "time_per_iteration": 3.579130172729492
    },
    {
      "auxiliary_loss_clip": 0.01229542,
      "auxiliary_loss_mlp": 0.0104318,
      "balance_loss_clip": 1.06298971,
      "balance_loss_mlp": 1.03083038,
      "epoch": 0.09908014188661095,
      "flos": 24679608526080.0,
      "grad_norm": 1.7437643297725411,
      "language_loss": 0.74301416,
      "learning_rate": 3.95022658270399e-06,
      "loss": 0.76574141,
      "num_input_tokens_seen": 17623085,
      "step": 824,
      "time_per_iteration": 2.7581188678741455
    },
    {
      "auxiliary_loss_clip": 0.01215478,
      "auxiliary_loss_mlp": 0.01057243,
      "balance_loss_clip": 1.06428027,
      "balance_loss_mlp": 1.04398715,
      "epoch": 0.09920038477725004,
      "flos": 14064307200000.0,
      "grad_norm": 1.919964366209745,
      "language_loss": 0.78227514,
      "learning_rate": 3.9500537311853635e-06,
      "loss": 0.80500233,
      "num_input_tokens_seen": 17641040,
      "step": 825,
      "time_per_iteration": 3.624086618423462
    },
    {
      "auxiliary_loss_clip": 0.01228949,
      "auxiliary_loss_mlp": 0.01044834,
      "balance_loss_clip": 1.06006539,
      "balance_loss_mlp": 1.03249037,
      "epoch": 0.09932062766788914,
      "flos": 13407070095360.0,
      "grad_norm": 3.6825376871893805,
      "language_loss": 0.83098298,
      "learning_rate": 3.949880583845136e-06,
      "loss": 0.85372078,
      "num_input_tokens_seen": 17659115,
      "step": 826,
      "time_per_iteration": 2.7161850929260254
    },
    {
      "auxiliary_loss_clip": 0.01213039,
      "auxiliary_loss_mlp": 0.01046911,
      "balance_loss_clip": 1.05997181,
      "balance_loss_mlp": 1.03409052,
      "epoch": 0.09944087055852822,
      "flos": 19500751566720.0,
      "grad_norm": 2.0513737128805816,
      "language_loss": 0.81416333,
      "learning_rate": 3.949707140709575e-06,
      "loss": 0.83676291,
      "num_input_tokens_seen": 17678845,
      "step": 827,
      "time_per_iteration": 4.519668102264404
    },
    {
      "auxiliary_loss_clip": 0.01232799,
      "auxiliary_loss_mlp": 0.01049576,
      "balance_loss_clip": 1.06239641,
      "balance_loss_mlp": 1.03704143,
      "epoch": 0.09956111344916732,
      "flos": 17749100926080.0,
      "grad_norm": 2.4230774788766247,
      "language_loss": 0.83359945,
      "learning_rate": 3.949533401804991e-06,
      "loss": 0.85642326,
      "num_input_tokens_seen": 17695750,
      "step": 828,
      "time_per_iteration": 2.673691749572754
    },
    {
      "auxiliary_loss_clip": 0.0123112,
      "auxiliary_loss_mlp": 0.00766235,
      "balance_loss_clip": 1.06533718,
      "balance_loss_mlp": 1.00062501,
      "epoch": 0.0996813563398064,
      "flos": 17967581400960.0,
      "grad_norm": 4.1966609025211445,
      "language_loss": 0.91040891,
      "learning_rate": 3.949359367157739e-06,
      "loss": 0.93038237,
      "num_input_tokens_seen": 17714445,
      "step": 829,
      "time_per_iteration": 2.806229591369629
    },
    {
      "auxiliary_loss_clip": 0.01236456,
      "auxiliary_loss_mlp": 0.01045063,
      "balance_loss_clip": 1.06555796,
      "balance_loss_mlp": 1.03184843,
      "epoch": 0.0998015992304455,
      "flos": 17457039440640.0,
      "grad_norm": 3.4249463196305547,
      "language_loss": 0.75415516,
      "learning_rate": 3.949185036794222e-06,
      "loss": 0.77697027,
      "num_input_tokens_seen": 17732455,
      "step": 830,
      "time_per_iteration": 2.819215774536133
    },
    {
      "auxiliary_loss_clip": 0.01245195,
      "auxiliary_loss_mlp": 0.01039655,
      "balance_loss_clip": 1.06576025,
      "balance_loss_mlp": 1.02757907,
      "epoch": 0.0999218421210846,
      "flos": 25888757080320.0,
      "grad_norm": 1.804329408124379,
      "language_loss": 0.78951228,
      "learning_rate": 3.949010410740884e-06,
      "loss": 0.81236076,
      "num_input_tokens_seen": 17755280,
      "step": 831,
      "time_per_iteration": 2.6853208541870117
    },
    {
      "auxiliary_loss_clip": 0.01206823,
      "auxiliary_loss_mlp": 0.00766965,
      "balance_loss_clip": 1.05958438,
      "balance_loss_mlp": 1.00052989,
      "epoch": 0.10004208501172368,
      "flos": 21215916967680.0,
      "grad_norm": 1.8601490531013611,
      "language_loss": 0.86316192,
      "learning_rate": 3.948835489024216e-06,
      "loss": 0.88289976,
      "num_input_tokens_seen": 17775015,
      "step": 832,
      "time_per_iteration": 2.9020628929138184
    },
    {
      "auxiliary_loss_clip": 0.01234539,
      "auxiliary_loss_mlp": 0.01036898,
      "balance_loss_clip": 1.06402206,
      "balance_loss_mlp": 1.02522159,
      "epoch": 0.10016232790236278,
      "flos": 17348409734400.0,
      "grad_norm": 2.2162637357167942,
      "language_loss": 0.90873241,
      "learning_rate": 3.948660271670755e-06,
      "loss": 0.93144679,
      "num_input_tokens_seen": 17792165,
      "step": 833,
      "time_per_iteration": 2.6379408836364746
    },
    {
      "auxiliary_loss_clip": 0.01212123,
      "auxiliary_loss_mlp": 0.010404,
      "balance_loss_clip": 1.06154728,
      "balance_loss_mlp": 1.02763271,
      "epoch": 0.10028257079300186,
      "flos": 25666541591040.0,
      "grad_norm": 2.4131961333562812,
      "language_loss": 0.84368348,
      "learning_rate": 3.948484758707079e-06,
      "loss": 0.86620867,
      "num_input_tokens_seen": 17811765,
      "step": 834,
      "time_per_iteration": 2.772630453109741
    },
    {
      "auxiliary_loss_clip": 0.01192461,
      "auxiliary_loss_mlp": 0.01042359,
      "balance_loss_clip": 1.05893183,
      "balance_loss_mlp": 1.03000855,
      "epoch": 0.10040281368364096,
      "flos": 25156035544320.0,
      "grad_norm": 2.3999315447902525,
      "language_loss": 0.83499879,
      "learning_rate": 3.948308950159815e-06,
      "loss": 0.85734701,
      "num_input_tokens_seen": 17830445,
      "step": 835,
      "time_per_iteration": 2.7213056087493896
    },
    {
      "auxiliary_loss_clip": 0.01194534,
      "auxiliary_loss_mlp": 0.0103827,
      "balance_loss_clip": 1.05565333,
      "balance_loss_mlp": 1.0252409,
      "epoch": 0.10052305657428004,
      "flos": 17603303621760.0,
      "grad_norm": 2.6071503651608645,
      "language_loss": 0.76302922,
      "learning_rate": 3.9481328460556326e-06,
      "loss": 0.7853573,
      "num_input_tokens_seen": 17847665,
      "step": 836,
      "time_per_iteration": 2.7633168697357178
    },
    {
      "auxiliary_loss_clip": 0.01207646,
      "auxiliary_loss_mlp": 0.01045346,
      "balance_loss_clip": 1.0585171,
      "balance_loss_mlp": 1.0316782,
      "epoch": 0.10064329946491914,
      "flos": 18660154510080.0,
      "grad_norm": 2.180934301004454,
      "language_loss": 0.8981961,
      "learning_rate": 3.9479564464212455e-06,
      "loss": 0.92072606,
      "num_input_tokens_seen": 17866825,
      "step": 837,
      "time_per_iteration": 2.7888665199279785
    },
    {
      "auxiliary_loss_clip": 0.0125124,
      "auxiliary_loss_mlp": 0.01043464,
      "balance_loss_clip": 1.06414366,
      "balance_loss_mlp": 1.02960014,
      "epoch": 0.10076354235555823,
      "flos": 17199056983680.0,
      "grad_norm": 3.0295685894645743,
      "language_loss": 0.76489699,
      "learning_rate": 3.947779751283414e-06,
      "loss": 0.78784394,
      "num_input_tokens_seen": 17883995,
      "step": 838,
      "time_per_iteration": 2.672478199005127
    },
    {
      "auxiliary_loss_clip": 0.01235008,
      "auxiliary_loss_mlp": 0.00766695,
      "balance_loss_clip": 1.07028317,
      "balance_loss_mlp": 1.0005393,
      "epoch": 0.10088378524619732,
      "flos": 22962252395520.0,
      "grad_norm": 2.0621387202742034,
      "language_loss": 0.76234341,
      "learning_rate": 3.947602760668944e-06,
      "loss": 0.78236043,
      "num_input_tokens_seen": 17903785,
      "step": 839,
      "time_per_iteration": 2.6979777812957764
    },
    {
      "auxiliary_loss_clip": 0.01235178,
      "auxiliary_loss_mlp": 0.01047862,
      "balance_loss_clip": 1.06732488,
      "balance_loss_mlp": 1.03448641,
      "epoch": 0.10100402813683641,
      "flos": 37885828746240.0,
      "grad_norm": 2.235870548688462,
      "language_loss": 0.71517932,
      "learning_rate": 3.947425474604684e-06,
      "loss": 0.73800969,
      "num_input_tokens_seen": 17927720,
      "step": 840,
      "time_per_iteration": 2.867412567138672
    },
    {
      "auxiliary_loss_clip": 0.01218493,
      "auxiliary_loss_mlp": 0.0104513,
      "balance_loss_clip": 1.06271315,
      "balance_loss_mlp": 1.03249359,
      "epoch": 0.1011242710274755,
      "flos": 21543458112000.0,
      "grad_norm": 2.01283865293059,
      "language_loss": 0.92273867,
      "learning_rate": 3.947247893117528e-06,
      "loss": 0.94537497,
      "num_input_tokens_seen": 17946225,
      "step": 841,
      "time_per_iteration": 2.6880884170532227
    },
    {
      "auxiliary_loss_clip": 0.01225586,
      "auxiliary_loss_mlp": 0.01051025,
      "balance_loss_clip": 1.0623318,
      "balance_loss_mlp": 1.03804302,
      "epoch": 0.10124451391811459,
      "flos": 13621456419840.0,
      "grad_norm": 3.650204804012978,
      "language_loss": 0.69128406,
      "learning_rate": 3.947070016234413e-06,
      "loss": 0.71405011,
      "num_input_tokens_seen": 17962015,
      "step": 842,
      "time_per_iteration": 2.7309603691101074
    },
    {
      "auxiliary_loss_clip": 0.01228004,
      "auxiliary_loss_mlp": 0.01049791,
      "balance_loss_clip": 1.06637359,
      "balance_loss_mlp": 1.03687525,
      "epoch": 0.10136475680875369,
      "flos": 16649228522880.0,
      "grad_norm": 2.5382925418729325,
      "language_loss": 0.75212044,
      "learning_rate": 3.946891843982326e-06,
      "loss": 0.77489841,
      "num_input_tokens_seen": 17979680,
      "step": 843,
      "time_per_iteration": 2.7245867252349854
    },
    {
      "auxiliary_loss_clip": 0.01233496,
      "auxiliary_loss_mlp": 0.01040563,
      "balance_loss_clip": 1.06793272,
      "balance_loss_mlp": 1.02733064,
      "epoch": 0.10148499969939277,
      "flos": 19461034103040.0,
      "grad_norm": 2.0780711712283986,
      "language_loss": 0.745902,
      "learning_rate": 3.9467133763882935e-06,
      "loss": 0.76864254,
      "num_input_tokens_seen": 17998145,
      "step": 844,
      "time_per_iteration": 2.7084572315216064
    },
    {
      "auxiliary_loss_clip": 0.01221951,
      "auxiliary_loss_mlp": 0.01047245,
      "balance_loss_clip": 1.0620538,
      "balance_loss_mlp": 1.03374505,
      "epoch": 0.10160524259003187,
      "flos": 21104988791040.0,
      "grad_norm": 2.0014232400570613,
      "language_loss": 0.85912716,
      "learning_rate": 3.9465346134793905e-06,
      "loss": 0.88181913,
      "num_input_tokens_seen": 18017955,
      "step": 845,
      "time_per_iteration": 2.690216064453125
    },
    {
      "auxiliary_loss_clip": 0.01200027,
      "auxiliary_loss_mlp": 0.01036313,
      "balance_loss_clip": 1.06252456,
      "balance_loss_mlp": 1.02404058,
      "epoch": 0.10172548548067095,
      "flos": 17712687513600.0,
      "grad_norm": 2.0335756715779207,
      "language_loss": 0.79773676,
      "learning_rate": 3.9463555552827335e-06,
      "loss": 0.82010019,
      "num_input_tokens_seen": 18035125,
      "step": 846,
      "time_per_iteration": 2.775200605392456
    },
    {
      "auxiliary_loss_clip": 0.01222065,
      "auxiliary_loss_mlp": 0.01046855,
      "balance_loss_clip": 1.06207514,
      "balance_loss_mlp": 1.03443956,
      "epoch": 0.10184572837131005,
      "flos": 21104845136640.0,
      "grad_norm": 5.183985209479954,
      "language_loss": 0.86436945,
      "learning_rate": 3.946176201825487e-06,
      "loss": 0.88705868,
      "num_input_tokens_seen": 18053160,
      "step": 847,
      "time_per_iteration": 2.7158937454223633
    },
    {
      "auxiliary_loss_clip": 0.01217622,
      "auxiliary_loss_mlp": 0.0104694,
      "balance_loss_clip": 1.06430387,
      "balance_loss_mlp": 1.03382099,
      "epoch": 0.10196597126194913,
      "flos": 26067591918720.0,
      "grad_norm": 2.226895493962201,
      "language_loss": 0.83666468,
      "learning_rate": 3.9459965531348575e-06,
      "loss": 0.85931039,
      "num_input_tokens_seen": 18072815,
      "step": 848,
      "time_per_iteration": 2.797443151473999
    },
    {
      "auxiliary_loss_clip": 0.01214277,
      "auxiliary_loss_mlp": 0.00766284,
      "balance_loss_clip": 1.0604893,
      "balance_loss_mlp": 1.00050855,
      "epoch": 0.10208621415258823,
      "flos": 29314634595840.0,
      "grad_norm": 2.1051017775835636,
      "language_loss": 0.85618114,
      "learning_rate": 3.945816609238098e-06,
      "loss": 0.87598681,
      "num_input_tokens_seen": 18092225,
      "step": 849,
      "time_per_iteration": 3.7540879249572754
    },
    {
      "auxiliary_loss_clip": 0.01183873,
      "auxiliary_loss_mlp": 0.01043565,
      "balance_loss_clip": 1.05978727,
      "balance_loss_mlp": 1.02994514,
      "epoch": 0.10220645704322733,
      "flos": 23805794367360.0,
      "grad_norm": 1.8746519610004226,
      "language_loss": 0.85068768,
      "learning_rate": 3.945636370162507e-06,
      "loss": 0.872962,
      "num_input_tokens_seen": 18112335,
      "step": 850,
      "time_per_iteration": 2.7986133098602295
    },
    {
      "auxiliary_loss_clip": 0.01231125,
      "auxiliary_loss_mlp": 0.01046044,
      "balance_loss_clip": 1.06457877,
      "balance_loss_mlp": 1.03362894,
      "epoch": 0.10232669993386641,
      "flos": 23218546913280.0,
      "grad_norm": 1.9147581704216472,
      "language_loss": 0.79162002,
      "learning_rate": 3.945455835935425e-06,
      "loss": 0.81439173,
      "num_input_tokens_seen": 18131520,
      "step": 851,
      "time_per_iteration": 2.6855945587158203
    },
    {
      "auxiliary_loss_clip": 0.01215669,
      "auxiliary_loss_mlp": 0.01052345,
      "balance_loss_clip": 1.05991721,
      "balance_loss_mlp": 1.0390408,
      "epoch": 0.1024469428245055,
      "flos": 22922929981440.0,
      "grad_norm": 2.3633033551218015,
      "language_loss": 0.75384748,
      "learning_rate": 3.94527500658424e-06,
      "loss": 0.77652764,
      "num_input_tokens_seen": 18149185,
      "step": 852,
      "time_per_iteration": 3.6297285556793213
    },
    {
      "auxiliary_loss_clip": 0.01186232,
      "auxiliary_loss_mlp": 0.01038383,
      "balance_loss_clip": 1.06101906,
      "balance_loss_mlp": 1.02588415,
      "epoch": 0.10256718571514459,
      "flos": 31359495957120.0,
      "grad_norm": 2.1535591545344155,
      "language_loss": 0.81323493,
      "learning_rate": 3.945093882136382e-06,
      "loss": 0.83548105,
      "num_input_tokens_seen": 18172960,
      "step": 853,
      "time_per_iteration": 3.778043270111084
    },
    {
      "auxiliary_loss_clip": 0.01221343,
      "auxiliary_loss_mlp": 0.00766406,
      "balance_loss_clip": 1.06723404,
      "balance_loss_mlp": 1.00042975,
      "epoch": 0.10268742860578368,
      "flos": 23474877344640.0,
      "grad_norm": 1.966549455973609,
      "language_loss": 0.84810567,
      "learning_rate": 3.944912462619329e-06,
      "loss": 0.8679831,
      "num_input_tokens_seen": 18191925,
      "step": 854,
      "time_per_iteration": 2.752828359603882
    },
    {
      "auxiliary_loss_clip": 0.01221739,
      "auxiliary_loss_mlp": 0.01057265,
      "balance_loss_clip": 1.06446409,
      "balance_loss_mlp": 1.04326415,
      "epoch": 0.10280767149642277,
      "flos": 25520313323520.0,
      "grad_norm": 4.54777152088384,
      "language_loss": 0.80602741,
      "learning_rate": 3.9447307480606025e-06,
      "loss": 0.82881743,
      "num_input_tokens_seen": 18212010,
      "step": 855,
      "time_per_iteration": 2.7974417209625244
    },
    {
      "auxiliary_loss_clip": 0.01207324,
      "auxiliary_loss_mlp": 0.01050072,
      "balance_loss_clip": 1.06220841,
      "balance_loss_mlp": 1.03693485,
      "epoch": 0.10292791438706186,
      "flos": 17347691462400.0,
      "grad_norm": 2.1737772003421774,
      "language_loss": 0.90353853,
      "learning_rate": 3.944548738487767e-06,
      "loss": 0.92611241,
      "num_input_tokens_seen": 18229525,
      "step": 856,
      "time_per_iteration": 2.7191638946533203
    },
    {
      "auxiliary_loss_clip": 0.01256104,
      "auxiliary_loss_mlp": 0.01051583,
      "balance_loss_clip": 1.07107818,
      "balance_loss_mlp": 1.0380888,
      "epoch": 0.10304815727770096,
      "flos": 27052693390080.0,
      "grad_norm": 2.652260681960219,
      "language_loss": 0.90757865,
      "learning_rate": 3.944366433928434e-06,
      "loss": 0.93065554,
      "num_input_tokens_seen": 18249505,
      "step": 857,
      "time_per_iteration": 2.7114450931549072
    },
    {
      "auxiliary_loss_clip": 0.01208388,
      "auxiliary_loss_mlp": 0.01038074,
      "balance_loss_clip": 1.05919623,
      "balance_loss_mlp": 1.02617133,
      "epoch": 0.10316840016834004,
      "flos": 22782591544320.0,
      "grad_norm": 2.224065329937135,
      "language_loss": 0.83720666,
      "learning_rate": 3.9441838344102594e-06,
      "loss": 0.85967129,
      "num_input_tokens_seen": 18269230,
      "step": 858,
      "time_per_iteration": 2.75365948677063
    },
    {
      "auxiliary_loss_clip": 0.01223887,
      "auxiliary_loss_mlp": 0.01035486,
      "balance_loss_clip": 1.06539249,
      "balance_loss_mlp": 1.02287972,
      "epoch": 0.10328864305897914,
      "flos": 20704584908160.0,
      "grad_norm": 3.03049128011679,
      "language_loss": 0.67695588,
      "learning_rate": 3.944000939960943e-06,
      "loss": 0.69954956,
      "num_input_tokens_seen": 18287955,
      "step": 859,
      "time_per_iteration": 2.8460476398468018
    },
    {
      "auxiliary_loss_clip": 0.01237835,
      "auxiliary_loss_mlp": 0.01051466,
      "balance_loss_clip": 1.06538081,
      "balance_loss_mlp": 1.0390209,
      "epoch": 0.10340888594961822,
      "flos": 28478814048000.0,
      "grad_norm": 1.6342849119935081,
      "language_loss": 0.79977047,
      "learning_rate": 3.943817750608229e-06,
      "loss": 0.82266343,
      "num_input_tokens_seen": 18310505,
      "step": 860,
      "time_per_iteration": 2.7784371376037598
    },
    {
      "auxiliary_loss_clip": 0.01237584,
      "auxiliary_loss_mlp": 0.01046214,
      "balance_loss_clip": 1.06813419,
      "balance_loss_mlp": 1.03424001,
      "epoch": 0.10352912884025732,
      "flos": 13370333460480.0,
      "grad_norm": 2.870927249681166,
      "language_loss": 0.81979513,
      "learning_rate": 3.943634266379908e-06,
      "loss": 0.84263313,
      "num_input_tokens_seen": 18327400,
      "step": 861,
      "time_per_iteration": 2.6743931770324707
    },
    {
      "auxiliary_loss_clip": 0.01234963,
      "auxiliary_loss_mlp": 0.0104742,
      "balance_loss_clip": 1.06387401,
      "balance_loss_mlp": 1.03484905,
      "epoch": 0.10364937173089642,
      "flos": 25558558329600.0,
      "grad_norm": 8.70984526721795,
      "language_loss": 0.8492893,
      "learning_rate": 3.943450487303815e-06,
      "loss": 0.87211311,
      "num_input_tokens_seen": 18347895,
      "step": 862,
      "time_per_iteration": 2.7945656776428223
    },
    {
      "auxiliary_loss_clip": 0.01232137,
      "auxiliary_loss_mlp": 0.01042822,
      "balance_loss_clip": 1.06670702,
      "balance_loss_mlp": 1.02935171,
      "epoch": 0.1037696146215355,
      "flos": 21215486004480.0,
      "grad_norm": 2.724199041010767,
      "language_loss": 0.85389996,
      "learning_rate": 3.943266413407827e-06,
      "loss": 0.87664956,
      "num_input_tokens_seen": 18367170,
      "step": 863,
      "time_per_iteration": 2.6405787467956543
    },
    {
      "auxiliary_loss_clip": 0.01236893,
      "auxiliary_loss_mlp": 0.01041858,
      "balance_loss_clip": 1.0665375,
      "balance_loss_mlp": 1.03008652,
      "epoch": 0.1038898575121746,
      "flos": 25807382818560.0,
      "grad_norm": 2.0212765000930055,
      "language_loss": 0.85204262,
      "learning_rate": 3.94308204471987e-06,
      "loss": 0.87483019,
      "num_input_tokens_seen": 18386185,
      "step": 864,
      "time_per_iteration": 2.736572504043579
    },
    {
      "auxiliary_loss_clip": 0.01202718,
      "auxiliary_loss_mlp": 0.01045815,
      "balance_loss_clip": 1.06030226,
      "balance_loss_mlp": 1.03366148,
      "epoch": 0.10401010040281368,
      "flos": 19062425900160.0,
      "grad_norm": 2.38618981585464,
      "language_loss": 0.74333978,
      "learning_rate": 3.942897381267912e-06,
      "loss": 0.76582509,
      "num_input_tokens_seen": 18402550,
      "step": 865,
      "time_per_iteration": 2.6789400577545166
    },
    {
      "auxiliary_loss_clip": 0.0123074,
      "auxiliary_loss_mlp": 0.01047364,
      "balance_loss_clip": 1.06444407,
      "balance_loss_mlp": 1.03462625,
      "epoch": 0.10413034329345278,
      "flos": 16355119962240.0,
      "grad_norm": 2.4417871316530664,
      "language_loss": 0.66440886,
      "learning_rate": 3.942712423079965e-06,
      "loss": 0.68718994,
      "num_input_tokens_seen": 18418940,
      "step": 866,
      "time_per_iteration": 2.648188829421997
    },
    {
      "auxiliary_loss_clip": 0.01183109,
      "auxiliary_loss_mlp": 0.01040896,
      "balance_loss_clip": 1.05025387,
      "balance_loss_mlp": 1.02853405,
      "epoch": 0.10425058618409186,
      "flos": 17236511890560.0,
      "grad_norm": 2.4365805615118528,
      "language_loss": 0.90076077,
      "learning_rate": 3.942527170184088e-06,
      "loss": 0.92300081,
      "num_input_tokens_seen": 18435560,
      "step": 867,
      "time_per_iteration": 2.6601991653442383
    },
    {
      "auxiliary_loss_clip": 0.01247976,
      "auxiliary_loss_mlp": 0.01045774,
      "balance_loss_clip": 1.06698036,
      "balance_loss_mlp": 1.03282166,
      "epoch": 0.10437082907473096,
      "flos": 17967365919360.0,
      "grad_norm": 2.6423603573592445,
      "language_loss": 0.78080183,
      "learning_rate": 3.942341622608385e-06,
      "loss": 0.80373931,
      "num_input_tokens_seen": 18452590,
      "step": 868,
      "time_per_iteration": 2.705995559692383
    },
    {
      "auxiliary_loss_clip": 0.01218664,
      "auxiliary_loss_mlp": 0.01041669,
      "balance_loss_clip": 1.06767201,
      "balance_loss_mlp": 1.02925324,
      "epoch": 0.10449107196537005,
      "flos": 36283315374720.0,
      "grad_norm": 1.6496802367269785,
      "language_loss": 0.77794516,
      "learning_rate": 3.942155780381001e-06,
      "loss": 0.80054843,
      "num_input_tokens_seen": 18476325,
      "step": 869,
      "time_per_iteration": 2.814772367477417
    },
    {
      "auxiliary_loss_clip": 0.01217448,
      "auxiliary_loss_mlp": 0.01043533,
      "balance_loss_clip": 1.06173706,
      "balance_loss_mlp": 1.03024697,
      "epoch": 0.10461131485600914,
      "flos": 23802095266560.0,
      "grad_norm": 1.9026002993777102,
      "language_loss": 0.75815982,
      "learning_rate": 3.94196964353013e-06,
      "loss": 0.78076965,
      "num_input_tokens_seen": 18495775,
      "step": 870,
      "time_per_iteration": 2.7935009002685547
    },
    {
      "auxiliary_loss_clip": 0.01211194,
      "auxiliary_loss_mlp": 0.00766155,
      "balance_loss_clip": 1.05718327,
      "balance_loss_mlp": 1.00043547,
      "epoch": 0.10473155774664823,
      "flos": 18405476104320.0,
      "grad_norm": 2.2179909265834206,
      "language_loss": 0.80854946,
      "learning_rate": 3.941783212084008e-06,
      "loss": 0.82832295,
      "num_input_tokens_seen": 18513530,
      "step": 871,
      "time_per_iteration": 2.635150194168091
    },
    {
      "auxiliary_loss_clip": 0.01204605,
      "auxiliary_loss_mlp": 0.01041876,
      "balance_loss_clip": 1.06213641,
      "balance_loss_mlp": 1.02898359,
      "epoch": 0.10485180063728732,
      "flos": 25592637358080.0,
      "grad_norm": 2.801892732330398,
      "language_loss": 0.78991276,
      "learning_rate": 3.941596486070916e-06,
      "loss": 0.81237757,
      "num_input_tokens_seen": 18531575,
      "step": 872,
      "time_per_iteration": 2.809086561203003
    },
    {
      "auxiliary_loss_clip": 0.01179668,
      "auxiliary_loss_mlp": 0.01038859,
      "balance_loss_clip": 1.06103826,
      "balance_loss_mlp": 1.02588856,
      "epoch": 0.10497204352792641,
      "flos": 27088747666560.0,
      "grad_norm": 3.041006339393339,
      "language_loss": 0.58689111,
      "learning_rate": 3.941409465519182e-06,
      "loss": 0.60907638,
      "num_input_tokens_seen": 18552100,
      "step": 873,
      "time_per_iteration": 2.810176372528076
    },
    {
      "auxiliary_loss_clip": 0.01222089,
      "auxiliary_loss_mlp": 0.01038217,
      "balance_loss_clip": 1.06166553,
      "balance_loss_mlp": 1.02528894,
      "epoch": 0.10509228641856551,
      "flos": 32858479353600.0,
      "grad_norm": 1.6296496474046385,
      "language_loss": 0.8528837,
      "learning_rate": 3.941222150457176e-06,
      "loss": 0.87548673,
      "num_input_tokens_seen": 18575355,
      "step": 874,
      "time_per_iteration": 2.801281213760376
    },
    {
      "auxiliary_loss_clip": 0.01236626,
      "auxiliary_loss_mlp": 0.01045921,
      "balance_loss_clip": 1.06380415,
      "balance_loss_mlp": 1.03333282,
      "epoch": 0.10521252930920459,
      "flos": 14319165173760.0,
      "grad_norm": 2.7069906480962205,
      "language_loss": 0.72073054,
      "learning_rate": 3.941034540913311e-06,
      "loss": 0.74355596,
      "num_input_tokens_seen": 18592885,
      "step": 875,
      "time_per_iteration": 3.513859510421753
    },
    {
      "auxiliary_loss_clip": 0.01236293,
      "auxiliary_loss_mlp": 0.00767022,
      "balance_loss_clip": 1.06554723,
      "balance_loss_mlp": 1.0003649,
      "epoch": 0.10533277219984369,
      "flos": 21687028773120.0,
      "grad_norm": 1.6527549340217902,
      "language_loss": 0.8231498,
      "learning_rate": 3.940846636916051e-06,
      "loss": 0.84318292,
      "num_input_tokens_seen": 18612920,
      "step": 876,
      "time_per_iteration": 2.6228067874908447
    },
    {
      "auxiliary_loss_clip": 0.01218215,
      "auxiliary_loss_mlp": 0.01043144,
      "balance_loss_clip": 1.06558788,
      "balance_loss_mlp": 1.03044868,
      "epoch": 0.10545301509048277,
      "flos": 22269787027200.0,
      "grad_norm": 2.1469769928113696,
      "language_loss": 0.86653167,
      "learning_rate": 3.940658438493899e-06,
      "loss": 0.88914526,
      "num_input_tokens_seen": 18630765,
      "step": 877,
      "time_per_iteration": 2.7319178581237793
    },
    {
      "auxiliary_loss_clip": 0.01249226,
      "auxiliary_loss_mlp": 0.01042402,
      "balance_loss_clip": 1.06460428,
      "balance_loss_mlp": 1.02933705,
      "epoch": 0.10557325798112187,
      "flos": 22199725549440.0,
      "grad_norm": 2.866454860164799,
      "language_loss": 0.75677502,
      "learning_rate": 3.940469945675405e-06,
      "loss": 0.77969128,
      "num_input_tokens_seen": 18649150,
      "step": 878,
      "time_per_iteration": 3.5418338775634766
    },
    {
      "auxiliary_loss_clip": 0.01166809,
      "auxiliary_loss_mlp": 0.01048354,
      "balance_loss_clip": 1.05872774,
      "balance_loss_mlp": 1.03643882,
      "epoch": 0.10569350087176095,
      "flos": 25775889569280.0,
      "grad_norm": 2.18972083473854,
      "language_loss": 0.91538298,
      "learning_rate": 3.940281158489163e-06,
      "loss": 0.93753463,
      "num_input_tokens_seen": 18668380,
      "step": 879,
      "time_per_iteration": 4.611927032470703
    },
    {
      "auxiliary_loss_clip": 0.01165435,
      "auxiliary_loss_mlp": 0.01043414,
      "balance_loss_clip": 1.05344748,
      "balance_loss_mlp": 1.03068197,
      "epoch": 0.10581374376240005,
      "flos": 17311385790720.0,
      "grad_norm": 1.8625438836957642,
      "language_loss": 0.82948434,
      "learning_rate": 3.940092076963812e-06,
      "loss": 0.85157275,
      "num_input_tokens_seen": 18685875,
      "step": 880,
      "time_per_iteration": 2.7155847549438477
    },
    {
      "auxiliary_loss_clip": 0.01215662,
      "auxiliary_loss_mlp": 0.01047218,
      "balance_loss_clip": 1.06050062,
      "balance_loss_mlp": 1.03408098,
      "epoch": 0.10593398665303914,
      "flos": 34349454017280.0,
      "grad_norm": 2.1203893087907835,
      "language_loss": 0.79066795,
      "learning_rate": 3.9399027011280355e-06,
      "loss": 0.81329674,
      "num_input_tokens_seen": 18707970,
      "step": 881,
      "time_per_iteration": 2.8114402294158936
    },
    {
      "auxiliary_loss_clip": 0.01220132,
      "auxiliary_loss_mlp": 0.01040243,
      "balance_loss_clip": 1.06882739,
      "balance_loss_mlp": 1.02670681,
      "epoch": 0.10605422954367823,
      "flos": 23257977068160.0,
      "grad_norm": 2.013315060915039,
      "language_loss": 0.77519238,
      "learning_rate": 3.939713031010561e-06,
      "loss": 0.79779613,
      "num_input_tokens_seen": 18726335,
      "step": 882,
      "time_per_iteration": 2.6769256591796875
    },
    {
      "auxiliary_loss_clip": 0.01200588,
      "auxiliary_loss_mlp": 0.01040384,
      "balance_loss_clip": 1.06201875,
      "balance_loss_mlp": 1.02829576,
      "epoch": 0.10617447243431732,
      "flos": 22820118278400.0,
      "grad_norm": 1.958399565876181,
      "language_loss": 0.77936989,
      "learning_rate": 3.939523066640163e-06,
      "loss": 0.80177963,
      "num_input_tokens_seen": 18745230,
      "step": 883,
      "time_per_iteration": 2.7463083267211914
    },
    {
      "auxiliary_loss_clip": 0.01236378,
      "auxiliary_loss_mlp": 0.01035566,
      "balance_loss_clip": 1.06654358,
      "balance_loss_mlp": 1.02310896,
      "epoch": 0.10629471532495641,
      "flos": 24386577373440.0,
      "grad_norm": 1.9980486137166447,
      "language_loss": 0.8122673,
      "learning_rate": 3.939332808045657e-06,
      "loss": 0.83498675,
      "num_input_tokens_seen": 18764880,
      "step": 884,
      "time_per_iteration": 2.7339131832122803
    },
    {
      "auxiliary_loss_clip": 0.01201727,
      "auxiliary_loss_mlp": 0.01036475,
      "balance_loss_clip": 1.06278765,
      "balance_loss_mlp": 1.02330828,
      "epoch": 0.1064149582155955,
      "flos": 21105491581440.0,
      "grad_norm": 1.808275246480227,
      "language_loss": 0.84684676,
      "learning_rate": 3.939142255255906e-06,
      "loss": 0.86922872,
      "num_input_tokens_seen": 18785765,
      "step": 885,
      "time_per_iteration": 2.7635855674743652
    },
    {
      "auxiliary_loss_clip": 0.01229297,
      "auxiliary_loss_mlp": 0.0105358,
      "balance_loss_clip": 1.06376374,
      "balance_loss_mlp": 1.04108059,
      "epoch": 0.1065352011062346,
      "flos": 20702035042560.0,
      "grad_norm": 2.1588754332130313,
      "language_loss": 0.86845803,
      "learning_rate": 3.938951408299817e-06,
      "loss": 0.89128685,
      "num_input_tokens_seen": 18804605,
      "step": 886,
      "time_per_iteration": 2.6421570777893066
    },
    {
      "auxiliary_loss_clip": 0.0111838,
      "auxiliary_loss_mlp": 0.01033337,
      "balance_loss_clip": 1.05633283,
      "balance_loss_mlp": 1.02909327,
      "epoch": 0.10665544399687368,
      "flos": 62659632689280.0,
      "grad_norm": 0.8080971349885397,
      "language_loss": 0.54477358,
      "learning_rate": 3.938760267206342e-06,
      "loss": 0.56629074,
      "num_input_tokens_seen": 18866425,
      "step": 887,
      "time_per_iteration": 3.50337290763855
    },
    {
      "auxiliary_loss_clip": 0.01246517,
      "auxiliary_loss_mlp": 0.0104142,
      "balance_loss_clip": 1.06672513,
      "balance_loss_mlp": 1.02877152,
      "epoch": 0.10677568688751278,
      "flos": 26140382830080.0,
      "grad_norm": 2.286074217334298,
      "language_loss": 0.78923786,
      "learning_rate": 3.938568832004475e-06,
      "loss": 0.81211722,
      "num_input_tokens_seen": 18885130,
      "step": 888,
      "time_per_iteration": 2.8928821086883545
    },
    {
      "auxiliary_loss_clip": 0.01205005,
      "auxiliary_loss_mlp": 0.0103782,
      "balance_loss_clip": 1.06015253,
      "balance_loss_mlp": 1.02467132,
      "epoch": 0.10689592977815186,
      "flos": 12786533712000.0,
      "grad_norm": 1.9285321127176778,
      "language_loss": 0.75394833,
      "learning_rate": 3.938377102723257e-06,
      "loss": 0.77637655,
      "num_input_tokens_seen": 18902265,
      "step": 889,
      "time_per_iteration": 2.669562578201294
    },
    {
      "auxiliary_loss_clip": 0.01169904,
      "auxiliary_loss_mlp": 0.01051975,
      "balance_loss_clip": 1.05648482,
      "balance_loss_mlp": 1.03873694,
      "epoch": 0.10701617266879096,
      "flos": 22126683242880.0,
      "grad_norm": 4.175362228161861,
      "language_loss": 0.83613646,
      "learning_rate": 3.938185079391774e-06,
      "loss": 0.85835522,
      "num_input_tokens_seen": 18919310,
      "step": 890,
      "time_per_iteration": 2.758146286010742
    },
    {
      "auxiliary_loss_clip": 0.01247529,
      "auxiliary_loss_mlp": 0.01044105,
      "balance_loss_clip": 1.06683469,
      "balance_loss_mlp": 1.03139114,
      "epoch": 0.10713641555943004,
      "flos": 19745625559680.0,
      "grad_norm": 2.4461098603454707,
      "language_loss": 1.06419182,
      "learning_rate": 3.937992762039157e-06,
      "loss": 1.08710814,
      "num_input_tokens_seen": 18932635,
      "step": 891,
      "time_per_iteration": 2.5805094242095947
    },
    {
      "auxiliary_loss_clip": 0.01233391,
      "auxiliary_loss_mlp": 0.01046433,
      "balance_loss_clip": 1.06718135,
      "balance_loss_mlp": 1.03498936,
      "epoch": 0.10725665845006914,
      "flos": 23952992302080.0,
      "grad_norm": 1.6299187110477094,
      "language_loss": 0.80522287,
      "learning_rate": 3.937800150694577e-06,
      "loss": 0.82802105,
      "num_input_tokens_seen": 18953810,
      "step": 892,
      "time_per_iteration": 2.689061403274536
    },
    {
      "auxiliary_loss_clip": 0.01189344,
      "auxiliary_loss_mlp": 0.01039296,
      "balance_loss_clip": 1.06128716,
      "balance_loss_mlp": 1.02690446,
      "epoch": 0.10737690134070824,
      "flos": 18551704371840.0,
      "grad_norm": 2.304122150003733,
      "language_loss": 0.76276076,
      "learning_rate": 3.937607245387255e-06,
      "loss": 0.78504717,
      "num_input_tokens_seen": 18973175,
      "step": 893,
      "time_per_iteration": 2.816016435623169
    },
    {
      "auxiliary_loss_clip": 0.01217855,
      "auxiliary_loss_mlp": 0.01040004,
      "balance_loss_clip": 1.06032777,
      "balance_loss_mlp": 1.02763629,
      "epoch": 0.10749714423134732,
      "flos": 22707609903360.0,
      "grad_norm": 2.104445150080955,
      "language_loss": 0.72159636,
      "learning_rate": 3.937414046146455e-06,
      "loss": 0.74417496,
      "num_input_tokens_seen": 18991130,
      "step": 894,
      "time_per_iteration": 2.7783203125
    },
    {
      "auxiliary_loss_clip": 0.01250052,
      "auxiliary_loss_mlp": 0.01051367,
      "balance_loss_clip": 1.06977296,
      "balance_loss_mlp": 1.03799152,
      "epoch": 0.10761738712198642,
      "flos": 21106066199040.0,
      "grad_norm": 3.602496869310269,
      "language_loss": 0.75437546,
      "learning_rate": 3.9372205530014845e-06,
      "loss": 0.77738965,
      "num_input_tokens_seen": 19009610,
      "step": 895,
      "time_per_iteration": 2.649681568145752
    },
    {
      "auxiliary_loss_clip": 0.01248769,
      "auxiliary_loss_mlp": 0.01041265,
      "balance_loss_clip": 1.06555653,
      "balance_loss_mlp": 1.02809262,
      "epoch": 0.1077376300126255,
      "flos": 23766723348480.0,
      "grad_norm": 2.6405859602332677,
      "language_loss": 0.7135148,
      "learning_rate": 3.937026765981696e-06,
      "loss": 0.73641521,
      "num_input_tokens_seen": 19029680,
      "step": 896,
      "time_per_iteration": 2.71081280708313
    },
    {
      "auxiliary_loss_clip": 0.01202369,
      "auxiliary_loss_mlp": 0.01045481,
      "balance_loss_clip": 1.06374383,
      "balance_loss_mlp": 1.0323379,
      "epoch": 0.1078578729032646,
      "flos": 20919581763840.0,
      "grad_norm": 1.7956908413202568,
      "language_loss": 0.79649657,
      "learning_rate": 3.936832685116488e-06,
      "loss": 0.81897503,
      "num_input_tokens_seen": 19047775,
      "step": 897,
      "time_per_iteration": 2.727457284927368
    },
    {
      "auxiliary_loss_clip": 0.01247253,
      "auxiliary_loss_mlp": 0.01040304,
      "balance_loss_clip": 1.06677735,
      "balance_loss_mlp": 1.02733409,
      "epoch": 0.10797811579390369,
      "flos": 14829886702080.0,
      "grad_norm": 2.400150538612429,
      "language_loss": 0.9014008,
      "learning_rate": 3.936638310435301e-06,
      "loss": 0.92427635,
      "num_input_tokens_seen": 19065640,
      "step": 898,
      "time_per_iteration": 2.640552282333374
    },
    {
      "auxiliary_loss_clip": 0.0123803,
      "auxiliary_loss_mlp": 0.01048388,
      "balance_loss_clip": 1.06714976,
      "balance_loss_mlp": 1.03563881,
      "epoch": 0.10809835868454278,
      "flos": 19536985411200.0,
      "grad_norm": 1.9963498496713716,
      "language_loss": 0.81481671,
      "learning_rate": 3.936443641967623e-06,
      "loss": 0.83768094,
      "num_input_tokens_seen": 19084470,
      "step": 899,
      "time_per_iteration": 2.6696507930755615
    },
    {
      "auxiliary_loss_clip": 0.01224742,
      "auxiliary_loss_mlp": 0.01041858,
      "balance_loss_clip": 1.06626463,
      "balance_loss_mlp": 1.02828622,
      "epoch": 0.10821860157518187,
      "flos": 18442320480000.0,
      "grad_norm": 2.057852574468916,
      "language_loss": 0.83245605,
      "learning_rate": 3.936248679742983e-06,
      "loss": 0.85512209,
      "num_input_tokens_seen": 19102965,
      "step": 900,
      "time_per_iteration": 2.6768975257873535
    },
    {
      "auxiliary_loss_clip": 0.01115453,
      "auxiliary_loss_mlp": 0.01015895,
      "balance_loss_clip": 1.04207575,
      "balance_loss_mlp": 1.01126981,
      "epoch": 0.10833884446582095,
      "flos": 49359468447360.0,
      "grad_norm": 1.0533330000125987,
      "language_loss": 0.70152789,
      "learning_rate": 3.936053423790959e-06,
      "loss": 0.72284138,
      "num_input_tokens_seen": 19151285,
      "step": 901,
      "time_per_iteration": 3.991771697998047
    },
    {
      "auxiliary_loss_clip": 0.01249106,
      "auxiliary_loss_mlp": 0.01038083,
      "balance_loss_clip": 1.0698365,
      "balance_loss_mlp": 1.02579248,
      "epoch": 0.10845908735646005,
      "flos": 20411912891520.0,
      "grad_norm": 2.7753995416901582,
      "language_loss": 0.7761265,
      "learning_rate": 3.935857874141168e-06,
      "loss": 0.79899836,
      "num_input_tokens_seen": 19170120,
      "step": 902,
      "time_per_iteration": 2.6928069591522217
    },
    {
      "auxiliary_loss_clip": 0.01216789,
      "auxiliary_loss_mlp": 0.0104358,
      "balance_loss_clip": 1.06579399,
      "balance_loss_mlp": 1.03042555,
      "epoch": 0.10857933024709913,
      "flos": 14027750133120.0,
      "grad_norm": 2.212637860243819,
      "language_loss": 0.8404665,
      "learning_rate": 3.935662030823279e-06,
      "loss": 0.86307013,
      "num_input_tokens_seen": 19186305,
      "step": 903,
      "time_per_iteration": 2.7537600994110107
    },
    {
      "auxiliary_loss_clip": 0.01232429,
      "auxiliary_loss_mlp": 0.0105149,
      "balance_loss_clip": 1.06293535,
      "balance_loss_mlp": 1.03882456,
      "epoch": 0.10869957313773823,
      "flos": 13369004657280.0,
      "grad_norm": 3.4176534045519698,
      "language_loss": 0.7266472,
      "learning_rate": 3.935465893866998e-06,
      "loss": 0.74948639,
      "num_input_tokens_seen": 19204530,
      "step": 904,
      "time_per_iteration": 3.567657232284546
    },
    {
      "auxiliary_loss_clip": 0.01214739,
      "auxiliary_loss_mlp": 0.01052053,
      "balance_loss_clip": 1.06431055,
      "balance_loss_mlp": 1.03884459,
      "epoch": 0.10881981602837733,
      "flos": 25807095509760.0,
      "grad_norm": 2.092687207648666,
      "language_loss": 0.80228299,
      "learning_rate": 3.935269463302079e-06,
      "loss": 0.82495093,
      "num_input_tokens_seen": 19222735,
      "step": 905,
      "time_per_iteration": 3.640103340148926
    },
    {
      "auxiliary_loss_clip": 0.01239282,
      "auxiliary_loss_mlp": 0.01051159,
      "balance_loss_clip": 1.06821966,
      "balance_loss_mlp": 1.03746235,
      "epoch": 0.10894005891901641,
      "flos": 20777555387520.0,
      "grad_norm": 1.7890963574337118,
      "language_loss": 0.77071738,
      "learning_rate": 3.935072739158322e-06,
      "loss": 0.79362178,
      "num_input_tokens_seen": 19242445,
      "step": 906,
      "time_per_iteration": 2.691774606704712
    },
    {
      "auxiliary_loss_clip": 0.01215313,
      "auxiliary_loss_mlp": 0.01048702,
      "balance_loss_clip": 1.06044984,
      "balance_loss_mlp": 1.03625035,
      "epoch": 0.10906030180965551,
      "flos": 26649883296000.0,
      "grad_norm": 2.024298022238146,
      "language_loss": 0.80061764,
      "learning_rate": 3.934875721465569e-06,
      "loss": 0.8232578,
      "num_input_tokens_seen": 19262865,
      "step": 907,
      "time_per_iteration": 2.7293708324432373
    },
    {
      "auxiliary_loss_clip": 0.01213286,
      "auxiliary_loss_mlp": 0.01045216,
      "balance_loss_clip": 1.06000066,
      "balance_loss_mlp": 1.03142905,
      "epoch": 0.10918054470029459,
      "flos": 36534402420480.0,
      "grad_norm": 2.1699830991752207,
      "language_loss": 0.71643782,
      "learning_rate": 3.9346784102537076e-06,
      "loss": 0.73902285,
      "num_input_tokens_seen": 19285000,
      "step": 908,
      "time_per_iteration": 2.8297483921051025
    },
    {
      "auxiliary_loss_clip": 0.01248014,
      "auxiliary_loss_mlp": 0.01041137,
      "balance_loss_clip": 1.06628454,
      "balance_loss_mlp": 1.02910304,
      "epoch": 0.10930078759093369,
      "flos": 21762549118080.0,
      "grad_norm": 1.8101754864628627,
      "language_loss": 0.78425217,
      "learning_rate": 3.934480805552669e-06,
      "loss": 0.80714369,
      "num_input_tokens_seen": 19306010,
      "step": 909,
      "time_per_iteration": 2.6531052589416504
    },
    {
      "auxiliary_loss_clip": 0.01243554,
      "auxiliary_loss_mlp": 0.00766678,
      "balance_loss_clip": 1.06356514,
      "balance_loss_mlp": 1.00029337,
      "epoch": 0.10942103048157277,
      "flos": 22601781457920.0,
      "grad_norm": 2.0949817134062494,
      "language_loss": 0.88251209,
      "learning_rate": 3.93428290739243e-06,
      "loss": 0.90261441,
      "num_input_tokens_seen": 19325380,
      "step": 910,
      "time_per_iteration": 2.61702823638916
    },
    {
      "auxiliary_loss_clip": 0.01219529,
      "auxiliary_loss_mlp": 0.01041906,
      "balance_loss_clip": 1.06306159,
      "balance_loss_mlp": 1.02987766,
      "epoch": 0.10954127337221187,
      "flos": 15045781397760.0,
      "grad_norm": 2.4045280418354027,
      "language_loss": 0.80006766,
      "learning_rate": 3.9340847158030125e-06,
      "loss": 0.82268202,
      "num_input_tokens_seen": 19338960,
      "step": 911,
      "time_per_iteration": 2.666435480117798
    },
    {
      "auxiliary_loss_clip": 0.01237052,
      "auxiliary_loss_mlp": 0.01043158,
      "balance_loss_clip": 1.06685925,
      "balance_loss_mlp": 1.03102255,
      "epoch": 0.10966151626285096,
      "flos": 21650974496640.0,
      "grad_norm": 1.804493007429792,
      "language_loss": 0.75487876,
      "learning_rate": 3.9338862308144814e-06,
      "loss": 0.77768087,
      "num_input_tokens_seen": 19357780,
      "step": 912,
      "time_per_iteration": 2.6516332626342773
    },
    {
      "auxiliary_loss_clip": 0.01249054,
      "auxiliary_loss_mlp": 0.01037409,
      "balance_loss_clip": 1.06906915,
      "balance_loss_mlp": 1.025756,
      "epoch": 0.10978175915349005,
      "flos": 20121359777280.0,
      "grad_norm": 1.6696219718298633,
      "language_loss": 0.84225106,
      "learning_rate": 3.933687452456946e-06,
      "loss": 0.86511564,
      "num_input_tokens_seen": 19377680,
      "step": 913,
      "time_per_iteration": 2.65940523147583
    },
    {
      "auxiliary_loss_clip": 0.01200463,
      "auxiliary_loss_mlp": 0.01040995,
      "balance_loss_clip": 1.0598619,
      "balance_loss_mlp": 1.02810299,
      "epoch": 0.10990200204412914,
      "flos": 20412667077120.0,
      "grad_norm": 2.7558152884865494,
      "language_loss": 0.86751771,
      "learning_rate": 3.933488380760562e-06,
      "loss": 0.88993227,
      "num_input_tokens_seen": 19397040,
      "step": 914,
      "time_per_iteration": 2.735445022583008
    },
    {
      "auxiliary_loss_clip": 0.0124561,
      "auxiliary_loss_mlp": 0.00767108,
      "balance_loss_clip": 1.06500518,
      "balance_loss_mlp": 1.0002178,
      "epoch": 0.11002224493476823,
      "flos": 17530117660800.0,
      "grad_norm": 1.976580054584242,
      "language_loss": 0.87228578,
      "learning_rate": 3.9332890157555286e-06,
      "loss": 0.89241302,
      "num_input_tokens_seen": 19413975,
      "step": 915,
      "time_per_iteration": 2.6587424278259277
    },
    {
      "auxiliary_loss_clip": 0.01218549,
      "auxiliary_loss_mlp": 0.01041519,
      "balance_loss_clip": 1.06381798,
      "balance_loss_mlp": 1.02860856,
      "epoch": 0.11014248782540732,
      "flos": 12203093099520.0,
      "grad_norm": 1.8954252434749164,
      "language_loss": 0.7646786,
      "learning_rate": 3.933089357472088e-06,
      "loss": 0.78727925,
      "num_input_tokens_seen": 19432005,
      "step": 916,
      "time_per_iteration": 2.6733617782592773
    },
    {
      "auxiliary_loss_clip": 0.01246705,
      "auxiliary_loss_mlp": 0.01042062,
      "balance_loss_clip": 1.06915808,
      "balance_loss_mlp": 1.02972412,
      "epoch": 0.11026273071604642,
      "flos": 22382977760640.0,
      "grad_norm": 1.8198640056544872,
      "language_loss": 0.86086285,
      "learning_rate": 3.932889405940529e-06,
      "loss": 0.88375056,
      "num_input_tokens_seen": 19450100,
      "step": 917,
      "time_per_iteration": 2.6456074714660645
    },
    {
      "auxiliary_loss_clip": 0.01216563,
      "auxiliary_loss_mlp": 0.01044116,
      "balance_loss_clip": 1.06689,
      "balance_loss_mlp": 1.03125954,
      "epoch": 0.1103829736066855,
      "flos": 19829046896640.0,
      "grad_norm": 2.207391567752706,
      "language_loss": 0.80012709,
      "learning_rate": 3.932689161191184e-06,
      "loss": 0.82273388,
      "num_input_tokens_seen": 19467805,
      "step": 918,
      "time_per_iteration": 2.759584665298462
    },
    {
      "auxiliary_loss_clip": 0.01229231,
      "auxiliary_loss_mlp": 0.01041287,
      "balance_loss_clip": 1.06298268,
      "balance_loss_mlp": 1.02829289,
      "epoch": 0.1105032164973246,
      "flos": 22669616292480.0,
      "grad_norm": 2.089991588550204,
      "language_loss": 0.88282669,
      "learning_rate": 3.93248862325443e-06,
      "loss": 0.90553188,
      "num_input_tokens_seen": 19486710,
      "step": 919,
      "time_per_iteration": 2.6891262531280518
    },
    {
      "auxiliary_loss_clip": 0.01143171,
      "auxiliary_loss_mlp": 0.01005942,
      "balance_loss_clip": 1.04195523,
      "balance_loss_mlp": 1.00153112,
      "epoch": 0.11062345938796368,
      "flos": 66483507876480.0,
      "grad_norm": 0.9356238967383186,
      "language_loss": 0.6436373,
      "learning_rate": 3.932287792160688e-06,
      "loss": 0.66512835,
      "num_input_tokens_seen": 19545170,
      "step": 920,
      "time_per_iteration": 3.168898344039917
    },
    {
      "auxiliary_loss_clip": 0.01232107,
      "auxiliary_loss_mlp": 0.01039851,
      "balance_loss_clip": 1.06333661,
      "balance_loss_mlp": 1.02704203,
      "epoch": 0.11074370227860278,
      "flos": 21907771804800.0,
      "grad_norm": 2.5904337070281085,
      "language_loss": 0.80551583,
      "learning_rate": 3.932086667940424e-06,
      "loss": 0.82823539,
      "num_input_tokens_seen": 19561875,
      "step": 921,
      "time_per_iteration": 2.661914825439453
    },
    {
      "auxiliary_loss_clip": 0.01225628,
      "auxiliary_loss_mlp": 0.00766505,
      "balance_loss_clip": 1.06499016,
      "balance_loss_mlp": 1.00023973,
      "epoch": 0.11086394516924186,
      "flos": 28658115763200.0,
      "grad_norm": 2.0189925312360812,
      "language_loss": 0.82029617,
      "learning_rate": 3.93188525062415e-06,
      "loss": 0.84021753,
      "num_input_tokens_seen": 19582340,
      "step": 922,
      "time_per_iteration": 2.7445971965789795
    },
    {
      "auxiliary_loss_clip": 0.01230313,
      "auxiliary_loss_mlp": 0.01040618,
      "balance_loss_clip": 1.06485081,
      "balance_loss_mlp": 1.02708733,
      "epoch": 0.11098418805988096,
      "flos": 24535247765760.0,
      "grad_norm": 2.4779742144676358,
      "language_loss": 0.86209947,
      "learning_rate": 3.931683540242418e-06,
      "loss": 0.88480878,
      "num_input_tokens_seen": 19603405,
      "step": 923,
      "time_per_iteration": 2.666851282119751
    },
    {
      "auxiliary_loss_clip": 0.01220921,
      "auxiliary_loss_mlp": 0.01047538,
      "balance_loss_clip": 1.06030118,
      "balance_loss_mlp": 1.03468776,
      "epoch": 0.11110443095052006,
      "flos": 22960384888320.0,
      "grad_norm": 2.9617159815421537,
      "language_loss": 0.91108,
      "learning_rate": 3.9314815368258295e-06,
      "loss": 0.93376458,
      "num_input_tokens_seen": 19619885,
      "step": 924,
      "time_per_iteration": 2.738511085510254
    },
    {
      "auxiliary_loss_clip": 0.01233552,
      "auxiliary_loss_mlp": 0.01039946,
      "balance_loss_clip": 1.06717813,
      "balance_loss_mlp": 1.0283525,
      "epoch": 0.11122467384115914,
      "flos": 18950025265920.0,
      "grad_norm": 1.744489379875936,
      "language_loss": 0.78978252,
      "learning_rate": 3.9312792404050275e-06,
      "loss": 0.81251746,
      "num_input_tokens_seen": 19637940,
      "step": 925,
      "time_per_iteration": 2.615124464035034
    },
    {
      "auxiliary_loss_clip": 0.01241466,
      "auxiliary_loss_mlp": 0.0104485,
      "balance_loss_clip": 1.06591105,
      "balance_loss_mlp": 1.03347111,
      "epoch": 0.11134491673179824,
      "flos": 25082957324160.0,
      "grad_norm": 2.0333945926790253,
      "language_loss": 0.77662396,
      "learning_rate": 3.9310766510107e-06,
      "loss": 0.79948711,
      "num_input_tokens_seen": 19657115,
      "step": 926,
      "time_per_iteration": 2.6691548824310303
    },
    {
      "auxiliary_loss_clip": 0.01204878,
      "auxiliary_loss_mlp": 0.01044171,
      "balance_loss_clip": 1.06116569,
      "balance_loss_mlp": 1.03167772,
      "epoch": 0.11146515962243732,
      "flos": 24499121662080.0,
      "grad_norm": 2.0510050306855874,
      "language_loss": 0.9190051,
      "learning_rate": 3.9308737686735806e-06,
      "loss": 0.94149554,
      "num_input_tokens_seen": 19677075,
      "step": 927,
      "time_per_iteration": 3.668632984161377
    },
    {
      "auxiliary_loss_clip": 0.01245128,
      "auxiliary_loss_mlp": 0.01046252,
      "balance_loss_clip": 1.06597066,
      "balance_loss_mlp": 1.03475976,
      "epoch": 0.11158540251307641,
      "flos": 22343763087360.0,
      "grad_norm": 2.314189862331838,
      "language_loss": 0.82781088,
      "learning_rate": 3.9306705934244455e-06,
      "loss": 0.85072464,
      "num_input_tokens_seen": 19697155,
      "step": 928,
      "time_per_iteration": 2.637141466140747
    },
    {
      "auxiliary_loss_clip": 0.01196225,
      "auxiliary_loss_mlp": 0.01036468,
      "balance_loss_clip": 1.05799985,
      "balance_loss_mlp": 1.02480948,
      "epoch": 0.11170564540371551,
      "flos": 19902304684800.0,
      "grad_norm": 2.0344790231946113,
      "language_loss": 0.88410747,
      "learning_rate": 3.930467125294116e-06,
      "loss": 0.90643442,
      "num_input_tokens_seen": 19716705,
      "step": 929,
      "time_per_iteration": 2.741306781768799
    },
    {
      "auxiliary_loss_clip": 0.01075586,
      "auxiliary_loss_mlp": 0.0101398,
      "balance_loss_clip": 1.02908206,
      "balance_loss_mlp": 1.00954556,
      "epoch": 0.1118258882943546,
      "flos": 64586239499520.0,
      "grad_norm": 0.9619706542878402,
      "language_loss": 0.6049825,
      "learning_rate": 3.930263364313458e-06,
      "loss": 0.62587816,
      "num_input_tokens_seen": 19767275,
      "step": 930,
      "time_per_iteration": 4.053828716278076
    },
    {
      "auxiliary_loss_clip": 0.01195583,
      "auxiliary_loss_mlp": 0.01048109,
      "balance_loss_clip": 1.05755901,
      "balance_loss_mlp": 1.03541338,
      "epoch": 0.11194613118499369,
      "flos": 17201965985280.0,
      "grad_norm": 1.9678652485269101,
      "language_loss": 0.83169824,
      "learning_rate": 3.930059310513384e-06,
      "loss": 0.85413522,
      "num_input_tokens_seen": 19786315,
      "step": 931,
      "time_per_iteration": 3.678701877593994
    },
    {
      "auxiliary_loss_clip": 0.0118279,
      "auxiliary_loss_mlp": 0.00766164,
      "balance_loss_clip": 1.05664051,
      "balance_loss_mlp": 1.00023556,
      "epoch": 0.11206637407563277,
      "flos": 31863465728640.0,
      "grad_norm": 1.7894914356257638,
      "language_loss": 0.8412717,
      "learning_rate": 3.929854963924846e-06,
      "loss": 0.86076123,
      "num_input_tokens_seen": 19806580,
      "step": 932,
      "time_per_iteration": 3.7827935218811035
    },
    {
      "auxiliary_loss_clip": 0.01197657,
      "auxiliary_loss_mlp": 0.01043243,
      "balance_loss_clip": 1.05710959,
      "balance_loss_mlp": 1.03179884,
      "epoch": 0.11218661696627187,
      "flos": 21945621761280.0,
      "grad_norm": 2.442619687463289,
      "language_loss": 0.77553833,
      "learning_rate": 3.929650324578845e-06,
      "loss": 0.79794729,
      "num_input_tokens_seen": 19826045,
      "step": 933,
      "time_per_iteration": 2.7438747882843018
    },
    {
      "auxiliary_loss_clip": 0.0121345,
      "auxiliary_loss_mlp": 0.01047755,
      "balance_loss_clip": 1.06033564,
      "balance_loss_mlp": 1.03448749,
      "epoch": 0.11230685985691095,
      "flos": 25878198481920.0,
      "grad_norm": 2.4748029237980855,
      "language_loss": 0.82257754,
      "learning_rate": 3.929445392506423e-06,
      "loss": 0.84518957,
      "num_input_tokens_seen": 19843985,
      "step": 934,
      "time_per_iteration": 2.769374132156372
    },
    {
      "auxiliary_loss_clip": 0.01226924,
      "auxiliary_loss_mlp": 0.01042588,
      "balance_loss_clip": 1.0663228,
      "balance_loss_mlp": 1.03107274,
      "epoch": 0.11242710274755005,
      "flos": 22231506107520.0,
      "grad_norm": 1.8630698329724482,
      "language_loss": 0.75717986,
      "learning_rate": 3.92924016773867e-06,
      "loss": 0.77987498,
      "num_input_tokens_seen": 19860480,
      "step": 935,
      "time_per_iteration": 2.684752941131592
    },
    {
      "auxiliary_loss_clip": 0.01208588,
      "auxiliary_loss_mlp": 0.0076555,
      "balance_loss_clip": 1.05764651,
      "balance_loss_mlp": 1.00021172,
      "epoch": 0.11254734563818915,
      "flos": 17712184723200.0,
      "grad_norm": 2.4708820206276396,
      "language_loss": 0.73527056,
      "learning_rate": 3.9290346503067175e-06,
      "loss": 0.75501192,
      "num_input_tokens_seen": 19877145,
      "step": 936,
      "time_per_iteration": 2.6843154430389404
    },
    {
      "auxiliary_loss_clip": 0.0122702,
      "auxiliary_loss_mlp": 0.01045479,
      "balance_loss_clip": 1.05984712,
      "balance_loss_mlp": 1.03398156,
      "epoch": 0.11266758852882823,
      "flos": 54930397334400.0,
      "grad_norm": 1.8824483813893804,
      "language_loss": 0.79062212,
      "learning_rate": 3.9288288402417415e-06,
      "loss": 0.8133471,
      "num_input_tokens_seen": 19903405,
      "step": 937,
      "time_per_iteration": 2.996154546737671
    },
    {
      "auxiliary_loss_clip": 0.01228938,
      "auxiliary_loss_mlp": 0.01048263,
      "balance_loss_clip": 1.06383896,
      "balance_loss_mlp": 1.03577638,
      "epoch": 0.11278783141946733,
      "flos": 18878132194560.0,
      "grad_norm": 2.418799948569045,
      "language_loss": 0.70665085,
      "learning_rate": 3.928622737574964e-06,
      "loss": 0.72942287,
      "num_input_tokens_seen": 19918740,
      "step": 938,
      "time_per_iteration": 2.589139938354492
    },
    {
      "auxiliary_loss_clip": 0.01209412,
      "auxiliary_loss_mlp": 0.01041237,
      "balance_loss_clip": 1.05730867,
      "balance_loss_mlp": 1.02933967,
      "epoch": 0.11290807431010641,
      "flos": 26469252777600.0,
      "grad_norm": 1.90500818354009,
      "language_loss": 0.91213286,
      "learning_rate": 3.928416342337652e-06,
      "loss": 0.93463939,
      "num_input_tokens_seen": 19938475,
      "step": 939,
      "time_per_iteration": 2.8485541343688965
    },
    {
      "auxiliary_loss_clip": 0.01210972,
      "auxiliary_loss_mlp": 0.01042714,
      "balance_loss_clip": 1.06076241,
      "balance_loss_mlp": 1.03037548,
      "epoch": 0.1130283172007455,
      "flos": 22710590732160.0,
      "grad_norm": 1.8356567422733645,
      "language_loss": 0.82697129,
      "learning_rate": 3.928209654561113e-06,
      "loss": 0.84950811,
      "num_input_tokens_seen": 19959310,
      "step": 940,
      "time_per_iteration": 2.6778650283813477
    },
    {
      "auxiliary_loss_clip": 0.01204254,
      "auxiliary_loss_mlp": 0.01040828,
      "balance_loss_clip": 1.06117082,
      "balance_loss_mlp": 1.02959824,
      "epoch": 0.1131485600913846,
      "flos": 23219911630080.0,
      "grad_norm": 2.7090578420998614,
      "language_loss": 0.81434512,
      "learning_rate": 3.928002674276703e-06,
      "loss": 0.83679593,
      "num_input_tokens_seen": 19978700,
      "step": 941,
      "time_per_iteration": 2.724036693572998
    },
    {
      "auxiliary_loss_clip": 0.01159386,
      "auxiliary_loss_mlp": 0.01038365,
      "balance_loss_clip": 1.05062127,
      "balance_loss_mlp": 1.0259254,
      "epoch": 0.11326880298202369,
      "flos": 14064271286400.0,
      "grad_norm": 2.5148930335213078,
      "language_loss": 0.76159537,
      "learning_rate": 3.92779540151582e-06,
      "loss": 0.78357291,
      "num_input_tokens_seen": 19995785,
      "step": 942,
      "time_per_iteration": 2.6909029483795166
    },
    {
      "auxiliary_loss_clip": 0.01214356,
      "auxiliary_loss_mlp": 0.01033661,
      "balance_loss_clip": 1.06281066,
      "balance_loss_mlp": 1.02162063,
      "epoch": 0.11338904587266278,
      "flos": 16325386479360.0,
      "grad_norm": 1.8727610944021158,
      "language_loss": 0.85762382,
      "learning_rate": 3.927587836309907e-06,
      "loss": 0.88010401,
      "num_input_tokens_seen": 20013615,
      "step": 943,
      "time_per_iteration": 2.709599733352661
    },
    {
      "auxiliary_loss_clip": 0.01205138,
      "auxiliary_loss_mlp": 0.01044077,
      "balance_loss_clip": 1.05908132,
      "balance_loss_mlp": 1.03213835,
      "epoch": 0.11350928876330187,
      "flos": 24426258923520.0,
      "grad_norm": 2.0838301047652243,
      "language_loss": 0.78213561,
      "learning_rate": 3.927379978690452e-06,
      "loss": 0.80462772,
      "num_input_tokens_seen": 20032880,
      "step": 944,
      "time_per_iteration": 2.7202539443969727
    },
    {
      "auxiliary_loss_clip": 0.01179112,
      "auxiliary_loss_mlp": 0.01039572,
      "balance_loss_clip": 1.05057192,
      "balance_loss_mlp": 1.027812,
      "epoch": 0.11362953165394096,
      "flos": 24497074586880.0,
      "grad_norm": 2.251789963989746,
      "language_loss": 0.87207854,
      "learning_rate": 3.927171828688987e-06,
      "loss": 0.89426535,
      "num_input_tokens_seen": 20052405,
      "step": 945,
      "time_per_iteration": 2.74395751953125
    },
    {
      "auxiliary_loss_clip": 0.01239403,
      "auxiliary_loss_mlp": 0.01044228,
      "balance_loss_clip": 1.06358266,
      "balance_loss_mlp": 1.03308821,
      "epoch": 0.11374977454458005,
      "flos": 24060831909120.0,
      "grad_norm": 2.713441190846801,
      "language_loss": 0.8200044,
      "learning_rate": 3.926963386337088e-06,
      "loss": 0.84284067,
      "num_input_tokens_seen": 20070635,
      "step": 946,
      "time_per_iteration": 2.6787290573120117
    },
    {
      "auxiliary_loss_clip": 0.01244004,
      "auxiliary_loss_mlp": 0.01039332,
      "balance_loss_clip": 1.0654825,
      "balance_loss_mlp": 1.02702355,
      "epoch": 0.11387001743521914,
      "flos": 39457638967680.0,
      "grad_norm": 2.66043057144668,
      "language_loss": 0.6982249,
      "learning_rate": 3.926754651666375e-06,
      "loss": 0.72105825,
      "num_input_tokens_seen": 20091195,
      "step": 947,
      "time_per_iteration": 2.7977957725524902
    },
    {
      "auxiliary_loss_clip": 0.0119403,
      "auxiliary_loss_mlp": 0.0103976,
      "balance_loss_clip": 1.05890822,
      "balance_loss_mlp": 1.0280838,
      "epoch": 0.11399026032585824,
      "flos": 25082454533760.0,
      "grad_norm": 3.1230107214901137,
      "language_loss": 0.78485614,
      "learning_rate": 3.926545624708513e-06,
      "loss": 0.80719399,
      "num_input_tokens_seen": 20110435,
      "step": 948,
      "time_per_iteration": 2.7633402347564697
    },
    {
      "auxiliary_loss_clip": 0.0118835,
      "auxiliary_loss_mlp": 0.01043386,
      "balance_loss_clip": 1.05540562,
      "balance_loss_mlp": 1.03183484,
      "epoch": 0.11411050321649732,
      "flos": 17961835224960.0,
      "grad_norm": 2.7802415678317396,
      "language_loss": 0.85803127,
      "learning_rate": 3.926336305495213e-06,
      "loss": 0.88034868,
      "num_input_tokens_seen": 20128995,
      "step": 949,
      "time_per_iteration": 2.6712121963500977
    },
    {
      "auxiliary_loss_clip": 0.01181103,
      "auxiliary_loss_mlp": 0.01039434,
      "balance_loss_clip": 1.05337596,
      "balance_loss_mlp": 1.02756119,
      "epoch": 0.11423074610713642,
      "flos": 22455409536000.0,
      "grad_norm": 3.203545537145531,
      "language_loss": 0.88745785,
      "learning_rate": 3.926126694058226e-06,
      "loss": 0.9096632,
      "num_input_tokens_seen": 20148145,
      "step": 950,
      "time_per_iteration": 2.7590508460998535
    },
    {
      "auxiliary_loss_clip": 0.01174442,
      "auxiliary_loss_mlp": 0.01038335,
      "balance_loss_clip": 1.05846429,
      "balance_loss_mlp": 1.02693856,
      "epoch": 0.1143509889977755,
      "flos": 19717687756800.0,
      "grad_norm": 1.4023193683428083,
      "language_loss": 0.8210789,
      "learning_rate": 3.92591679042935e-06,
      "loss": 0.8432067,
      "num_input_tokens_seen": 20168035,
      "step": 951,
      "time_per_iteration": 2.7586982250213623
    },
    {
      "auxiliary_loss_clip": 0.01227502,
      "auxiliary_loss_mlp": 0.01047092,
      "balance_loss_clip": 1.06656897,
      "balance_loss_mlp": 1.03474772,
      "epoch": 0.1144712318884146,
      "flos": 19822869757440.0,
      "grad_norm": 1.9381444018132117,
      "language_loss": 0.82432806,
      "learning_rate": 3.92570659464043e-06,
      "loss": 0.84707403,
      "num_input_tokens_seen": 20186095,
      "step": 952,
      "time_per_iteration": 2.6687779426574707
    },
    {
      "auxiliary_loss_clip": 0.01223601,
      "auxiliary_loss_mlp": 0.0076702,
      "balance_loss_clip": 1.06467128,
      "balance_loss_mlp": 1.0002538,
      "epoch": 0.1145914747790537,
      "flos": 14939198766720.0,
      "grad_norm": 2.02107728223693,
      "language_loss": 0.80192882,
      "learning_rate": 3.925496106723349e-06,
      "loss": 0.82183498,
      "num_input_tokens_seen": 20203535,
      "step": 953,
      "time_per_iteration": 3.585479259490967
    },
    {
      "auxiliary_loss_clip": 0.01226048,
      "auxiliary_loss_mlp": 0.0104189,
      "balance_loss_clip": 1.06188786,
      "balance_loss_mlp": 1.02976036,
      "epoch": 0.11471171766969278,
      "flos": 19865029345920.0,
      "grad_norm": 2.4363253936646863,
      "language_loss": 0.83883834,
      "learning_rate": 3.9252853267100405e-06,
      "loss": 0.86151773,
      "num_input_tokens_seen": 20222780,
      "step": 954,
      "time_per_iteration": 2.6228597164154053
    },
    {
      "auxiliary_loss_clip": 0.01186543,
      "auxiliary_loss_mlp": 0.01039105,
      "balance_loss_clip": 1.05905259,
      "balance_loss_mlp": 1.02729714,
      "epoch": 0.11483196056033187,
      "flos": 22526476594560.0,
      "grad_norm": 2.3855024245819303,
      "language_loss": 0.83867049,
      "learning_rate": 3.9250742546324786e-06,
      "loss": 0.86092699,
      "num_input_tokens_seen": 20243015,
      "step": 955,
      "time_per_iteration": 2.7445526123046875
    },
    {
      "auxiliary_loss_clip": 0.01207852,
      "auxiliary_loss_mlp": 0.01037793,
      "balance_loss_clip": 1.0602169,
      "balance_loss_mlp": 1.0262115,
      "epoch": 0.11495220345097096,
      "flos": 28220292887040.0,
      "grad_norm": 1.9286240264953656,
      "language_loss": 0.86832291,
      "learning_rate": 3.924862890522683e-06,
      "loss": 0.89077938,
      "num_input_tokens_seen": 20263025,
      "step": 956,
      "time_per_iteration": 3.6156067848205566
    },
    {
      "auxiliary_loss_clip": 0.01223611,
      "auxiliary_loss_mlp": 0.01042058,
      "balance_loss_clip": 1.06044948,
      "balance_loss_mlp": 1.02960014,
      "epoch": 0.11507244634161005,
      "flos": 17492267704320.0,
      "grad_norm": 3.8476838524434,
      "language_loss": 0.86394691,
      "learning_rate": 3.9246512344127174e-06,
      "loss": 0.88660359,
      "num_input_tokens_seen": 20280685,
      "step": 957,
      "time_per_iteration": 4.452392101287842
    },
    {
      "auxiliary_loss_clip": 0.01148771,
      "auxiliary_loss_mlp": 0.01039029,
      "balance_loss_clip": 1.05381346,
      "balance_loss_mlp": 1.02762115,
      "epoch": 0.11519268923224914,
      "flos": 22564937082240.0,
      "grad_norm": 2.357301816905252,
      "language_loss": 0.81979465,
      "learning_rate": 3.9244392863346895e-06,
      "loss": 0.84167266,
      "num_input_tokens_seen": 20300090,
      "step": 958,
      "time_per_iteration": 2.839357376098633
    },
    {
      "auxiliary_loss_clip": 0.01218342,
      "auxiliary_loss_mlp": 0.01046092,
      "balance_loss_clip": 1.06786919,
      "balance_loss_mlp": 1.03230548,
      "epoch": 0.11531293212288823,
      "flos": 16982839065600.0,
      "grad_norm": 1.921215739868516,
      "language_loss": 0.92433584,
      "learning_rate": 3.9242270463207524e-06,
      "loss": 0.94698018,
      "num_input_tokens_seen": 20318480,
      "step": 959,
      "time_per_iteration": 2.6328580379486084
    },
    {
      "auxiliary_loss_clip": 0.01169025,
      "auxiliary_loss_mlp": 0.01033907,
      "balance_loss_clip": 1.05587888,
      "balance_loss_mlp": 1.02188516,
      "epoch": 0.11543317501352733,
      "flos": 12422004537600.0,
      "grad_norm": 3.680768908868924,
      "language_loss": 0.85114092,
      "learning_rate": 3.924014514403102e-06,
      "loss": 0.87317026,
      "num_input_tokens_seen": 20334635,
      "step": 960,
      "time_per_iteration": 2.782628297805786
    },
    {
      "auxiliary_loss_clip": 0.01175066,
      "auxiliary_loss_mlp": 0.01036467,
      "balance_loss_clip": 1.05417514,
      "balance_loss_mlp": 1.02456963,
      "epoch": 0.11555341790416641,
      "flos": 19821648695040.0,
      "grad_norm": 2.280670672565852,
      "language_loss": 0.91109574,
      "learning_rate": 3.92380169061398e-06,
      "loss": 0.93321109,
      "num_input_tokens_seen": 20352415,
      "step": 961,
      "time_per_iteration": 2.796318531036377
    },
    {
      "auxiliary_loss_clip": 0.01186945,
      "auxiliary_loss_mlp": 0.00766728,
      "balance_loss_clip": 1.0534786,
      "balance_loss_mlp": 1.00023937,
      "epoch": 0.11567366079480551,
      "flos": 25738865625600.0,
      "grad_norm": 2.2182044948234854,
      "language_loss": 0.83695763,
      "learning_rate": 3.9235885749856705e-06,
      "loss": 0.85649431,
      "num_input_tokens_seen": 20371095,
      "step": 962,
      "time_per_iteration": 2.735335111618042
    },
    {
      "auxiliary_loss_clip": 0.01212266,
      "auxiliary_loss_mlp": 0.01048401,
      "balance_loss_clip": 1.06498218,
      "balance_loss_mlp": 1.03653395,
      "epoch": 0.1157939036854446,
      "flos": 18223301301120.0,
      "grad_norm": 2.1172917927199033,
      "language_loss": 0.82457185,
      "learning_rate": 3.9233751675505035e-06,
      "loss": 0.84717852,
      "num_input_tokens_seen": 20389805,
      "step": 963,
      "time_per_iteration": 2.7368810176849365
    },
    {
      "auxiliary_loss_clip": 0.01208606,
      "auxiliary_loss_mlp": 0.01050231,
      "balance_loss_clip": 1.06327796,
      "balance_loss_mlp": 1.03797066,
      "epoch": 0.11591414657608369,
      "flos": 23073755189760.0,
      "grad_norm": 2.831350012665959,
      "language_loss": 0.85114169,
      "learning_rate": 3.923161468340853e-06,
      "loss": 0.87373006,
      "num_input_tokens_seen": 20409640,
      "step": 964,
      "time_per_iteration": 2.6985890865325928
    },
    {
      "auxiliary_loss_clip": 0.01164901,
      "auxiliary_loss_mlp": 0.01051439,
      "balance_loss_clip": 1.05065608,
      "balance_loss_mlp": 1.03969646,
      "epoch": 0.11603438946672277,
      "flos": 19461716461440.0,
      "grad_norm": 1.8287065880319882,
      "language_loss": 0.81426215,
      "learning_rate": 3.9229474773891374e-06,
      "loss": 0.83642554,
      "num_input_tokens_seen": 20428180,
      "step": 965,
      "time_per_iteration": 2.735017776489258
    },
    {
      "auxiliary_loss_clip": 0.01197022,
      "auxiliary_loss_mlp": 0.01041269,
      "balance_loss_clip": 1.05257916,
      "balance_loss_mlp": 1.02877593,
      "epoch": 0.11615463235736187,
      "flos": 26831986272000.0,
      "grad_norm": 11.018900659824112,
      "language_loss": 0.84009349,
      "learning_rate": 3.922733194727818e-06,
      "loss": 0.86247647,
      "num_input_tokens_seen": 20447975,
      "step": 966,
      "time_per_iteration": 2.722153902053833
    },
    {
      "auxiliary_loss_clip": 0.01232885,
      "auxiliary_loss_mlp": 0.01043351,
      "balance_loss_clip": 1.06558871,
      "balance_loss_mlp": 1.03029168,
      "epoch": 0.11627487524800097,
      "flos": 18580324533120.0,
      "grad_norm": 2.2477593052344944,
      "language_loss": 0.87119257,
      "learning_rate": 3.922518620389402e-06,
      "loss": 0.89395499,
      "num_input_tokens_seen": 20464840,
      "step": 967,
      "time_per_iteration": 2.640848159790039
    },
    {
      "auxiliary_loss_clip": 0.01120612,
      "auxiliary_loss_mlp": 0.0104354,
      "balance_loss_clip": 1.04912531,
      "balance_loss_mlp": 1.03221548,
      "epoch": 0.11639511813864005,
      "flos": 18150474476160.0,
      "grad_norm": 2.1908432728590013,
      "language_loss": 0.89534372,
      "learning_rate": 3.922303754406439e-06,
      "loss": 0.91698527,
      "num_input_tokens_seen": 20482680,
      "step": 968,
      "time_per_iteration": 2.917951822280884
    },
    {
      "auxiliary_loss_clip": 0.01176585,
      "auxiliary_loss_mlp": 0.01046399,
      "balance_loss_clip": 1.05419278,
      "balance_loss_mlp": 1.03430557,
      "epoch": 0.11651536102927915,
      "flos": 20922023888640.0,
      "grad_norm": 2.0891838082258176,
      "language_loss": 0.79058111,
      "learning_rate": 3.922088596811526e-06,
      "loss": 0.81281096,
      "num_input_tokens_seen": 20501810,
      "step": 969,
      "time_per_iteration": 3.018881320953369
    },
    {
      "auxiliary_loss_clip": 0.01216814,
      "auxiliary_loss_mlp": 0.01044417,
      "balance_loss_clip": 1.06240034,
      "balance_loss_mlp": 1.0320673,
      "epoch": 0.11663560391991823,
      "flos": 16508602776960.0,
      "grad_norm": 2.761817269000313,
      "language_loss": 0.86915678,
      "learning_rate": 3.9218731476373e-06,
      "loss": 0.89176911,
      "num_input_tokens_seen": 20517995,
      "step": 970,
      "time_per_iteration": 2.578216552734375
    },
    {
      "auxiliary_loss_clip": 0.01230007,
      "auxiliary_loss_mlp": 0.01045348,
      "balance_loss_clip": 1.06567621,
      "balance_loss_mlp": 1.03213394,
      "epoch": 0.11675584681055733,
      "flos": 19865029345920.0,
      "grad_norm": 2.21898532467825,
      "language_loss": 0.84510791,
      "learning_rate": 3.9216574069164455e-06,
      "loss": 0.86786151,
      "num_input_tokens_seen": 20536970,
      "step": 971,
      "time_per_iteration": 2.662442207336426
    },
    {
      "auxiliary_loss_clip": 0.01236273,
      "auxiliary_loss_mlp": 0.01049932,
      "balance_loss_clip": 1.06203866,
      "balance_loss_mlp": 1.03840458,
      "epoch": 0.11687608970119642,
      "flos": 21944364785280.0,
      "grad_norm": 1.601132191449602,
      "language_loss": 0.8026607,
      "learning_rate": 3.921441374681691e-06,
      "loss": 0.82552278,
      "num_input_tokens_seen": 20557030,
      "step": 972,
      "time_per_iteration": 2.6640870571136475
    },
    {
      "auxiliary_loss_clip": 0.0120301,
      "auxiliary_loss_mlp": 0.01047886,
      "balance_loss_clip": 1.05925512,
      "balance_loss_mlp": 1.03559542,
      "epoch": 0.1169963325918355,
      "flos": 24061155131520.0,
      "grad_norm": 1.7536400200938245,
      "language_loss": 0.64747274,
      "learning_rate": 3.921225050965808e-06,
      "loss": 0.66998172,
      "num_input_tokens_seen": 20576915,
      "step": 973,
      "time_per_iteration": 2.7196614742279053
    },
    {
      "auxiliary_loss_clip": 0.01194518,
      "auxiliary_loss_mlp": 0.01043712,
      "balance_loss_clip": 1.06012273,
      "balance_loss_mlp": 1.03163612,
      "epoch": 0.1171165754824746,
      "flos": 23368151059200.0,
      "grad_norm": 2.3956614799887364,
      "language_loss": 0.74985856,
      "learning_rate": 3.921008435801612e-06,
      "loss": 0.77224088,
      "num_input_tokens_seen": 20596000,
      "step": 974,
      "time_per_iteration": 2.765462636947632
    },
    {
      "auxiliary_loss_clip": 0.01214333,
      "auxiliary_loss_mlp": 0.01036882,
      "balance_loss_clip": 1.05909729,
      "balance_loss_mlp": 1.02458584,
      "epoch": 0.11723681837311369,
      "flos": 18552243075840.0,
      "grad_norm": 2.0090630832286593,
      "language_loss": 0.75966966,
      "learning_rate": 3.920791529221963e-06,
      "loss": 0.78218174,
      "num_input_tokens_seen": 20614675,
      "step": 975,
      "time_per_iteration": 2.6394972801208496
    },
    {
      "auxiliary_loss_clip": 0.01208587,
      "auxiliary_loss_mlp": 0.00766352,
      "balance_loss_clip": 1.05880678,
      "balance_loss_mlp": 1.00027275,
      "epoch": 0.11735706126375278,
      "flos": 23550541344000.0,
      "grad_norm": 2.2279314327012347,
      "language_loss": 0.76726687,
      "learning_rate": 3.920574331259768e-06,
      "loss": 0.78701627,
      "num_input_tokens_seen": 20635875,
      "step": 976,
      "time_per_iteration": 2.747483253479004
    },
    {
      "auxiliary_loss_clip": 0.01197717,
      "auxiliary_loss_mlp": 0.01042687,
      "balance_loss_clip": 1.05875754,
      "balance_loss_mlp": 1.02947283,
      "epoch": 0.11747730415439187,
      "flos": 22381541216640.0,
      "grad_norm": 4.241316545385091,
      "language_loss": 0.79744089,
      "learning_rate": 3.9203568419479716e-06,
      "loss": 0.81984496,
      "num_input_tokens_seen": 20656430,
      "step": 977,
      "time_per_iteration": 2.6808676719665527
    },
    {
      "auxiliary_loss_clip": 0.01207752,
      "auxiliary_loss_mlp": 0.01042121,
      "balance_loss_clip": 1.05998313,
      "balance_loss_mlp": 1.03015256,
      "epoch": 0.11759754704503096,
      "flos": 22200731130240.0,
      "grad_norm": 1.8784960756479705,
      "language_loss": 0.75267178,
      "learning_rate": 3.92013906131957e-06,
      "loss": 0.77517051,
      "num_input_tokens_seen": 20675360,
      "step": 978,
      "time_per_iteration": 2.7283780574798584
    },
    {
      "auxiliary_loss_clip": 0.01194735,
      "auxiliary_loss_mlp": 0.01048382,
      "balance_loss_clip": 1.05908442,
      "balance_loss_mlp": 1.03676546,
      "epoch": 0.11771778993567006,
      "flos": 22309755886080.0,
      "grad_norm": 1.594555645243806,
      "language_loss": 0.82723701,
      "learning_rate": 3.9199209894076e-06,
      "loss": 0.8496682,
      "num_input_tokens_seen": 20695675,
      "step": 979,
      "time_per_iteration": 3.673367500305176
    },
    {
      "auxiliary_loss_clip": 0.01238869,
      "auxiliary_loss_mlp": 0.01039675,
      "balance_loss_clip": 1.06095433,
      "balance_loss_mlp": 1.02675843,
      "epoch": 0.11783803282630914,
      "flos": 21288169175040.0,
      "grad_norm": 2.028395915344561,
      "language_loss": 0.90016335,
      "learning_rate": 3.919702626245142e-06,
      "loss": 0.92294878,
      "num_input_tokens_seen": 20715330,
      "step": 980,
      "time_per_iteration": 2.627700090408325
    },
    {
      "auxiliary_loss_clip": 0.01194115,
      "auxiliary_loss_mlp": 0.01040816,
      "balance_loss_clip": 1.05565047,
      "balance_loss_mlp": 1.02766156,
      "epoch": 0.11795827571694824,
      "flos": 25371535190400.0,
      "grad_norm": 2.516723385917264,
      "language_loss": 0.66517174,
      "learning_rate": 3.919483971865322e-06,
      "loss": 0.68752098,
      "num_input_tokens_seen": 20735325,
      "step": 981,
      "time_per_iteration": 2.7805824279785156
    },
    {
      "auxiliary_loss_clip": 0.01210565,
      "auxiliary_loss_mlp": 0.0104357,
      "balance_loss_clip": 1.06297064,
      "balance_loss_mlp": 1.0314467,
      "epoch": 0.11807851860758732,
      "flos": 23622218933760.0,
      "grad_norm": 3.010496536546041,
      "language_loss": 0.87977231,
      "learning_rate": 3.91926502630131e-06,
      "loss": 0.90231359,
      "num_input_tokens_seen": 20755940,
      "step": 982,
      "time_per_iteration": 4.591506004333496
    },
    {
      "auxiliary_loss_clip": 0.01232182,
      "auxiliary_loss_mlp": 0.01046582,
      "balance_loss_clip": 1.06879413,
      "balance_loss_mlp": 1.03469634,
      "epoch": 0.11819876149822642,
      "flos": 24972496024320.0,
      "grad_norm": 1.853811005970903,
      "language_loss": 0.72170103,
      "learning_rate": 3.91904578958632e-06,
      "loss": 0.74448866,
      "num_input_tokens_seen": 20775355,
      "step": 983,
      "time_per_iteration": 3.6544196605682373
    },
    {
      "auxiliary_loss_clip": 0.01242216,
      "auxiliary_loss_mlp": 0.01037051,
      "balance_loss_clip": 1.06588817,
      "balance_loss_mlp": 1.02527916,
      "epoch": 0.11831900438886551,
      "flos": 23003226835200.0,
      "grad_norm": 2.361953154635545,
      "language_loss": 0.83964217,
      "learning_rate": 3.918826261753608e-06,
      "loss": 0.86243486,
      "num_input_tokens_seen": 20794935,
      "step": 984,
      "time_per_iteration": 2.6990838050842285
    },
    {
      "auxiliary_loss_clip": 0.01208057,
      "auxiliary_loss_mlp": 0.01043314,
      "balance_loss_clip": 1.05932236,
      "balance_loss_mlp": 1.03069019,
      "epoch": 0.1184392472795046,
      "flos": 27965147604480.0,
      "grad_norm": 3.018648225488855,
      "language_loss": 0.7081753,
      "learning_rate": 3.918606442836478e-06,
      "loss": 0.73068893,
      "num_input_tokens_seen": 20817155,
      "step": 985,
      "time_per_iteration": 2.7193636894226074
    },
    {
      "auxiliary_loss_clip": 0.01223472,
      "auxiliary_loss_mlp": 0.01039511,
      "balance_loss_clip": 1.06475222,
      "balance_loss_mlp": 1.02745318,
      "epoch": 0.1185594901701437,
      "flos": 19898497843200.0,
      "grad_norm": 1.8282481166300326,
      "language_loss": 0.77451992,
      "learning_rate": 3.918386332868277e-06,
      "loss": 0.79714978,
      "num_input_tokens_seen": 20835125,
      "step": 986,
      "time_per_iteration": 2.672311544418335
    },
    {
      "auxiliary_loss_clip": 0.01217269,
      "auxiliary_loss_mlp": 0.01035429,
      "balance_loss_clip": 1.06138325,
      "balance_loss_mlp": 1.023669,
      "epoch": 0.11867973306078278,
      "flos": 18912354877440.0,
      "grad_norm": 1.6806511150770254,
      "language_loss": 0.94136679,
      "learning_rate": 3.918165931882394e-06,
      "loss": 0.96389377,
      "num_input_tokens_seen": 20853525,
      "step": 987,
      "time_per_iteration": 2.6779942512512207
    },
    {
      "auxiliary_loss_clip": 0.01149473,
      "auxiliary_loss_mlp": 0.01045635,
      "balance_loss_clip": 1.04991853,
      "balance_loss_mlp": 1.03426874,
      "epoch": 0.11879997595142187,
      "flos": 16982803152000.0,
      "grad_norm": 2.606474351763251,
      "language_loss": 0.75148934,
      "learning_rate": 3.917945239912264e-06,
      "loss": 0.77344036,
      "num_input_tokens_seen": 20871000,
      "step": 988,
      "time_per_iteration": 2.821244955062866
    },
    {
      "auxiliary_loss_clip": 0.01178148,
      "auxiliary_loss_mlp": 0.01042161,
      "balance_loss_clip": 1.05618382,
      "balance_loss_mlp": 1.03101516,
      "epoch": 0.11892021884206096,
      "flos": 17530369056000.0,
      "grad_norm": 2.0452572511216562,
      "language_loss": 0.7549057,
      "learning_rate": 3.917724256991367e-06,
      "loss": 0.77710873,
      "num_input_tokens_seen": 20889745,
      "step": 989,
      "time_per_iteration": 2.781541347503662
    },
    {
      "auxiliary_loss_clip": 0.01196474,
      "auxiliary_loss_mlp": 0.01043677,
      "balance_loss_clip": 1.05832624,
      "balance_loss_mlp": 1.03133297,
      "epoch": 0.11904046173270005,
      "flos": 30955895763840.0,
      "grad_norm": 2.37185533473021,
      "language_loss": 0.81819278,
      "learning_rate": 3.9175029831532245e-06,
      "loss": 0.84059429,
      "num_input_tokens_seen": 20909260,
      "step": 990,
      "time_per_iteration": 2.807943344116211
    },
    {
      "auxiliary_loss_clip": 0.01194762,
      "auxiliary_loss_mlp": 0.01037252,
      "balance_loss_clip": 1.06024683,
      "balance_loss_mlp": 1.02527118,
      "epoch": 0.11916070462333915,
      "flos": 20157234485760.0,
      "grad_norm": 2.335386215766892,
      "language_loss": 0.88629401,
      "learning_rate": 3.917281418431404e-06,
      "loss": 0.90861416,
      "num_input_tokens_seen": 20928305,
      "step": 991,
      "time_per_iteration": 2.8357701301574707
    },
    {
      "auxiliary_loss_clip": 0.01209864,
      "auxiliary_loss_mlp": 0.01040948,
      "balance_loss_clip": 1.06176877,
      "balance_loss_mlp": 1.02884209,
      "epoch": 0.11928094751397823,
      "flos": 23551115961600.0,
      "grad_norm": 4.627295551758001,
      "language_loss": 0.76924664,
      "learning_rate": 3.917059562859516e-06,
      "loss": 0.79175472,
      "num_input_tokens_seen": 20947630,
      "step": 992,
      "time_per_iteration": 2.7211391925811768
    },
    {
      "auxiliary_loss_clip": 0.01196209,
      "auxiliary_loss_mlp": 0.01046774,
      "balance_loss_clip": 1.05791759,
      "balance_loss_mlp": 1.03545523,
      "epoch": 0.11940119040461733,
      "flos": 23908426502400.0,
      "grad_norm": 2.103340825448616,
      "language_loss": 0.88911939,
      "learning_rate": 3.916837416471218e-06,
      "loss": 0.91154921,
      "num_input_tokens_seen": 20964250,
      "step": 993,
      "time_per_iteration": 2.7415590286254883
    },
    {
      "auxiliary_loss_clip": 0.01212488,
      "auxiliary_loss_mlp": 0.01041683,
      "balance_loss_clip": 1.055179,
      "balance_loss_mlp": 1.02973223,
      "epoch": 0.11952143329525641,
      "flos": 13844533835520.0,
      "grad_norm": 2.7099815192463628,
      "language_loss": 0.72377706,
      "learning_rate": 3.916614979300207e-06,
      "loss": 0.7463187,
      "num_input_tokens_seen": 20979095,
      "step": 994,
      "time_per_iteration": 2.6360268592834473
    },
    {
      "auxiliary_loss_clip": 0.01169415,
      "auxiliary_loss_mlp": 0.01040863,
      "balance_loss_clip": 1.05651307,
      "balance_loss_mlp": 1.02904916,
      "epoch": 0.11964167618589551,
      "flos": 27015525792000.0,
      "grad_norm": 1.7423434347040332,
      "language_loss": 0.79016989,
      "learning_rate": 3.9163922513802274e-06,
      "loss": 0.81227267,
      "num_input_tokens_seen": 21001430,
      "step": 995,
      "time_per_iteration": 2.7856686115264893
    },
    {
      "auxiliary_loss_clip": 0.01239784,
      "auxiliary_loss_mlp": 0.01046106,
      "balance_loss_clip": 1.06280887,
      "balance_loss_mlp": 1.03411973,
      "epoch": 0.1197619190765346,
      "flos": 12567622273920.0,
      "grad_norm": 3.5008829695044756,
      "language_loss": 0.82648677,
      "learning_rate": 3.916169232745067e-06,
      "loss": 0.84934568,
      "num_input_tokens_seen": 21019105,
      "step": 996,
      "time_per_iteration": 2.5894672870635986
    },
    {
      "auxiliary_loss_clip": 0.01197132,
      "auxiliary_loss_mlp": 0.01041151,
      "balance_loss_clip": 1.05834794,
      "balance_loss_mlp": 1.0289979,
      "epoch": 0.11988216196717369,
      "flos": 16909437623040.0,
      "grad_norm": 3.463031328335616,
      "language_loss": 0.9199627,
      "learning_rate": 3.915945923428559e-06,
      "loss": 0.94234556,
      "num_input_tokens_seen": 21035630,
      "step": 997,
      "time_per_iteration": 2.6703150272369385
    },
    {
      "auxiliary_loss_clip": 0.01216926,
      "auxiliary_loss_mlp": 0.01036467,
      "balance_loss_clip": 1.05744815,
      "balance_loss_mlp": 1.02417016,
      "epoch": 0.12000240485781279,
      "flos": 16216577205120.0,
      "grad_norm": 2.223741481512645,
      "language_loss": 0.82723713,
      "learning_rate": 3.915722323464577e-06,
      "loss": 0.84977108,
      "num_input_tokens_seen": 21054235,
      "step": 998,
      "time_per_iteration": 2.5862603187561035
    },
    {
      "auxiliary_loss_clip": 0.01223153,
      "auxiliary_loss_mlp": 0.01045563,
      "balance_loss_clip": 1.06189752,
      "balance_loss_mlp": 1.03320646,
      "epoch": 0.12012264774845187,
      "flos": 49344887525760.0,
      "grad_norm": 2.654948446917869,
      "language_loss": 0.7049132,
      "learning_rate": 3.91549843288704e-06,
      "loss": 0.72760034,
      "num_input_tokens_seen": 21077915,
      "step": 999,
      "time_per_iteration": 2.8897671699523926
    },
    {
      "auxiliary_loss_clip": 0.01187739,
      "auxiliary_loss_mlp": 0.00766025,
      "balance_loss_clip": 1.05170202,
      "balance_loss_mlp": 1.00029778,
      "epoch": 0.12024289063909097,
      "flos": 26979435601920.0,
      "grad_norm": 6.836598332948917,
      "language_loss": 0.79306066,
      "learning_rate": 3.915274251729916e-06,
      "loss": 0.81259835,
      "num_input_tokens_seen": 21099205,
      "step": 1000,
      "time_per_iteration": 2.7679402828216553
    },
    {
      "auxiliary_loss_clip": 0.01199068,
      "auxiliary_loss_mlp": 0.01040181,
      "balance_loss_clip": 1.05937397,
      "balance_loss_mlp": 1.02802742,
      "epoch": 0.12036313352973005,
      "flos": 19537308633600.0,
      "grad_norm": 2.100508337181235,
      "language_loss": 0.90401828,
      "learning_rate": 3.91504978002721e-06,
      "loss": 0.92641079,
      "num_input_tokens_seen": 21118260,
      "step": 1001,
      "time_per_iteration": 2.6632964611053467
    },
    {
      "auxiliary_loss_clip": 0.01205623,
      "auxiliary_loss_mlp": 0.00765827,
      "balance_loss_clip": 1.0555104,
      "balance_loss_mlp": 1.00025058,
      "epoch": 0.12048337642036915,
      "flos": 17268256535040.0,
      "grad_norm": 2.139760483327339,
      "language_loss": 0.7625075,
      "learning_rate": 3.914825017812974e-06,
      "loss": 0.78222197,
      "num_input_tokens_seen": 21134910,
      "step": 1002,
      "time_per_iteration": 2.669376850128174
    },
    {
      "auxiliary_loss_clip": 0.01208367,
      "auxiliary_loss_mlp": 0.01040331,
      "balance_loss_clip": 1.06180906,
      "balance_loss_mlp": 1.02870786,
      "epoch": 0.12060361931100824,
      "flos": 22856962654080.0,
      "grad_norm": 3.1723476437260074,
      "language_loss": 0.72827929,
      "learning_rate": 3.9145999651213065e-06,
      "loss": 0.75076628,
      "num_input_tokens_seen": 21154150,
      "step": 1003,
      "time_per_iteration": 2.6851589679718018
    },
    {
      "auxiliary_loss_clip": 0.01219851,
      "auxiliary_loss_mlp": 0.01039249,
      "balance_loss_clip": 1.05988896,
      "balance_loss_mlp": 1.02698219,
      "epoch": 0.12072386220164733,
      "flos": 16726795943040.0,
      "grad_norm": 2.9322866067130313,
      "language_loss": 0.87743139,
      "learning_rate": 3.9143746219863465e-06,
      "loss": 0.90002239,
      "num_input_tokens_seen": 21171255,
      "step": 1004,
      "time_per_iteration": 2.6655683517456055
    },
    {
      "auxiliary_loss_clip": 0.01128134,
      "auxiliary_loss_mlp": 0.01015774,
      "balance_loss_clip": 1.03513479,
      "balance_loss_mlp": 1.01172054,
      "epoch": 0.12084410509228642,
      "flos": 55144176105600.0,
      "grad_norm": 0.9642825807684384,
      "language_loss": 0.64759457,
      "learning_rate": 3.914148988442278e-06,
      "loss": 0.66903365,
      "num_input_tokens_seen": 21227045,
      "step": 1005,
      "time_per_iteration": 4.149083375930786
    },
    {
      "auxiliary_loss_clip": 0.01199904,
      "auxiliary_loss_mlp": 0.01044434,
      "balance_loss_clip": 1.06069803,
      "balance_loss_mlp": 1.03080845,
      "epoch": 0.1209643479829255,
      "flos": 26760236855040.0,
      "grad_norm": 3.3712035042437263,
      "language_loss": 0.95385063,
      "learning_rate": 3.91392306452333e-06,
      "loss": 0.97629404,
      "num_input_tokens_seen": 21244120,
      "step": 1006,
      "time_per_iteration": 2.6977760791778564
    },
    {
      "auxiliary_loss_clip": 0.01241587,
      "auxiliary_loss_mlp": 0.01039246,
      "balance_loss_clip": 1.06577992,
      "balance_loss_mlp": 1.02599573,
      "epoch": 0.1210845908735646,
      "flos": 11035026725760.0,
      "grad_norm": 3.0433924067305123,
      "language_loss": 0.66414505,
      "learning_rate": 3.913696850263774e-06,
      "loss": 0.68695337,
      "num_input_tokens_seen": 21258485,
      "step": 1007,
      "time_per_iteration": 2.5928385257720947
    },
    {
      "auxiliary_loss_clip": 0.01224049,
      "auxiliary_loss_mlp": 0.01034809,
      "balance_loss_clip": 1.06162,
      "balance_loss_mlp": 1.02187455,
      "epoch": 0.1212048337642037,
      "flos": 20484631975680.0,
      "grad_norm": 2.714896387572321,
      "language_loss": 0.79220355,
      "learning_rate": 3.913470345697929e-06,
      "loss": 0.8147921,
      "num_input_tokens_seen": 21277115,
      "step": 1008,
      "time_per_iteration": 2.6702680587768555
    },
    {
      "auxiliary_loss_clip": 0.01180781,
      "auxiliary_loss_mlp": 0.01055174,
      "balance_loss_clip": 1.0581758,
      "balance_loss_mlp": 1.04278183,
      "epoch": 0.12132507665484278,
      "flos": 22346061557760.0,
      "grad_norm": 2.1612957399340558,
      "language_loss": 0.85538167,
      "learning_rate": 3.913243550860153e-06,
      "loss": 0.87774122,
      "num_input_tokens_seen": 21294880,
      "step": 1009,
      "time_per_iteration": 4.556337833404541
    },
    {
      "auxiliary_loss_clip": 0.01229511,
      "auxiliary_loss_mlp": 0.01038687,
      "balance_loss_clip": 1.06807673,
      "balance_loss_mlp": 1.02620614,
      "epoch": 0.12144531954548188,
      "flos": 29314957818240.0,
      "grad_norm": 1.9318389329214432,
      "language_loss": 0.76182044,
      "learning_rate": 3.913016465784852e-06,
      "loss": 0.78450239,
      "num_input_tokens_seen": 21315555,
      "step": 1010,
      "time_per_iteration": 2.760122060775757
    },
    {
      "auxiliary_loss_clip": 0.01177105,
      "auxiliary_loss_mlp": 0.01043451,
      "balance_loss_clip": 1.05449259,
      "balance_loss_mlp": 1.03123784,
      "epoch": 0.12156556243612096,
      "flos": 20485242506880.0,
      "grad_norm": 3.152718434235606,
      "language_loss": 0.7217055,
      "learning_rate": 3.912789090506474e-06,
      "loss": 0.74391103,
      "num_input_tokens_seen": 21334815,
      "step": 1011,
      "time_per_iteration": 2.7064361572265625
    },
    {
      "auxiliary_loss_clip": 0.01196555,
      "auxiliary_loss_mlp": 0.01042131,
      "balance_loss_clip": 1.05560446,
      "balance_loss_mlp": 1.02916718,
      "epoch": 0.12168580532676006,
      "flos": 16472009796480.0,
      "grad_norm": 2.2598393338023715,
      "language_loss": 0.71455395,
      "learning_rate": 3.9125614250595114e-06,
      "loss": 0.73694086,
      "num_input_tokens_seen": 21351025,
      "step": 1012,
      "time_per_iteration": 2.720088243484497
    },
    {
      "auxiliary_loss_clip": 0.01224939,
      "auxiliary_loss_mlp": 0.01041069,
      "balance_loss_clip": 1.06164598,
      "balance_loss_mlp": 1.02883244,
      "epoch": 0.12180604821739914,
      "flos": 15341290588800.0,
      "grad_norm": 3.610529489626667,
      "language_loss": 0.8918128,
      "learning_rate": 3.912333469478502e-06,
      "loss": 0.91447282,
      "num_input_tokens_seen": 21368990,
      "step": 1013,
      "time_per_iteration": 2.6277966499328613
    },
    {
      "auxiliary_loss_clip": 0.01208199,
      "auxiliary_loss_mlp": 0.01046439,
      "balance_loss_clip": 1.05979753,
      "balance_loss_mlp": 1.03442907,
      "epoch": 0.12192629110803824,
      "flos": 19318038059520.0,
      "grad_norm": 2.0282263993499448,
      "language_loss": 0.78075641,
      "learning_rate": 3.912105223798025e-06,
      "loss": 0.80330276,
      "num_input_tokens_seen": 21388410,
      "step": 1014,
      "time_per_iteration": 2.7380383014678955
    },
    {
      "auxiliary_loss_clip": 0.01111117,
      "auxiliary_loss_mlp": 0.01008344,
      "balance_loss_clip": 1.03049302,
      "balance_loss_mlp": 1.00438595,
      "epoch": 0.12204653399867733,
      "flos": 47725354085760.0,
      "grad_norm": 0.997267695907163,
      "language_loss": 0.67599809,
      "learning_rate": 3.9118766880527065e-06,
      "loss": 0.69719279,
      "num_input_tokens_seen": 21442845,
      "step": 1015,
      "time_per_iteration": 3.139171838760376
    },
    {
      "auxiliary_loss_clip": 0.01165538,
      "auxiliary_loss_mlp": 0.01035787,
      "balance_loss_clip": 1.0551095,
      "balance_loss_mlp": 1.02442598,
      "epoch": 0.12216677688931642,
      "flos": 18221936584320.0,
      "grad_norm": 2.6617455559319594,
      "language_loss": 0.73838609,
      "learning_rate": 3.9116478622772145e-06,
      "loss": 0.76039934,
      "num_input_tokens_seen": 21461420,
      "step": 1016,
      "time_per_iteration": 2.842247724533081
    },
    {
      "auxiliary_loss_clip": 0.01220746,
      "auxiliary_loss_mlp": 0.01041676,
      "balance_loss_clip": 1.06475544,
      "balance_loss_mlp": 1.03017187,
      "epoch": 0.12228701977995551,
      "flos": 27525636789120.0,
      "grad_norm": 1.6797797903490534,
      "language_loss": 0.87949681,
      "learning_rate": 3.911418746506261e-06,
      "loss": 0.90212107,
      "num_input_tokens_seen": 21481550,
      "step": 1017,
      "time_per_iteration": 2.766968011856079
    },
    {
      "auxiliary_loss_clip": 0.01231613,
      "auxiliary_loss_mlp": 0.01039303,
      "balance_loss_clip": 1.06980133,
      "balance_loss_mlp": 1.02760887,
      "epoch": 0.1224072626705946,
      "flos": 21798136517760.0,
      "grad_norm": 2.1052424541637014,
      "language_loss": 0.78587574,
      "learning_rate": 3.911189340774604e-06,
      "loss": 0.80858493,
      "num_input_tokens_seen": 21501680,
      "step": 1018,
      "time_per_iteration": 2.852184295654297
    },
    {
      "auxiliary_loss_clip": 0.01215152,
      "auxiliary_loss_mlp": 0.01040848,
      "balance_loss_clip": 1.05927789,
      "balance_loss_mlp": 1.02868855,
      "epoch": 0.1225275055612337,
      "flos": 20703758895360.0,
      "grad_norm": 1.8352759341157254,
      "language_loss": 0.79429638,
      "learning_rate": 3.910959645117043e-06,
      "loss": 0.81685638,
      "num_input_tokens_seen": 21521015,
      "step": 1019,
      "time_per_iteration": 2.707390069961548
    },
    {
      "auxiliary_loss_clip": 0.01116646,
      "auxiliary_loss_mlp": 0.00757178,
      "balance_loss_clip": 1.03242874,
      "balance_loss_mlp": 1.00058997,
      "epoch": 0.12264774845187278,
      "flos": 57745294462080.0,
      "grad_norm": 0.8192346531021951,
      "language_loss": 0.56712556,
      "learning_rate": 3.910729659568423e-06,
      "loss": 0.58586377,
      "num_input_tokens_seen": 21578200,
      "step": 1020,
      "time_per_iteration": 3.213798999786377
    },
    {
      "auxiliary_loss_clip": 0.01207365,
      "auxiliary_loss_mlp": 0.01038297,
      "balance_loss_clip": 1.06127453,
      "balance_loss_mlp": 1.02760363,
      "epoch": 0.12276799134251187,
      "flos": 26396282298240.0,
      "grad_norm": 1.8117810529576872,
      "language_loss": 0.82636088,
      "learning_rate": 3.9104993841636344e-06,
      "loss": 0.84881747,
      "num_input_tokens_seen": 21598770,
      "step": 1021,
      "time_per_iteration": 2.7447478771209717
    },
    {
      "auxiliary_loss_clip": 0.01203224,
      "auxiliary_loss_mlp": 0.0076578,
      "balance_loss_clip": 1.0625968,
      "balance_loss_mlp": 1.00021994,
      "epoch": 0.12288823423315097,
      "flos": 21064193919360.0,
      "grad_norm": 1.92550574409633,
      "language_loss": 0.80797172,
      "learning_rate": 3.910268818937608e-06,
      "loss": 0.82766175,
      "num_input_tokens_seen": 21616925,
      "step": 1022,
      "time_per_iteration": 2.702984571456909
    },
    {
      "auxiliary_loss_clip": 0.01179876,
      "auxiliary_loss_mlp": 0.01050083,
      "balance_loss_clip": 1.06073713,
      "balance_loss_mlp": 1.03819799,
      "epoch": 0.12300847712379005,
      "flos": 12312441077760.0,
      "grad_norm": 2.6334926811970325,
      "language_loss": 0.87809557,
      "learning_rate": 3.9100379639253196e-06,
      "loss": 0.90039515,
      "num_input_tokens_seen": 21633645,
      "step": 1023,
      "time_per_iteration": 2.7395734786987305
    },
    {
      "auxiliary_loss_clip": 0.01204328,
      "auxiliary_loss_mlp": 0.01038534,
      "balance_loss_clip": 1.05612648,
      "balance_loss_mlp": 1.02721548,
      "epoch": 0.12312872001442915,
      "flos": 16762239688320.0,
      "grad_norm": 2.4865386583179783,
      "language_loss": 0.86495757,
      "learning_rate": 3.909806819161791e-06,
      "loss": 0.8873862,
      "num_input_tokens_seen": 21649120,
      "step": 1024,
      "time_per_iteration": 2.6661536693573
    },
    {
      "auxiliary_loss_clip": 0.01197366,
      "auxiliary_loss_mlp": 0.01043399,
      "balance_loss_clip": 1.05812716,
      "balance_loss_mlp": 1.0316267,
      "epoch": 0.12324896290506823,
      "flos": 18404937400320.0,
      "grad_norm": 2.7099821135562743,
      "language_loss": 0.86673242,
      "learning_rate": 3.909575384682086e-06,
      "loss": 0.88914007,
      "num_input_tokens_seen": 21668000,
      "step": 1025,
      "time_per_iteration": 2.6912689208984375
    },
    {
      "auxiliary_loss_clip": 0.01222144,
      "auxiliary_loss_mlp": 0.01034944,
      "balance_loss_clip": 1.0599184,
      "balance_loss_mlp": 1.02367234,
      "epoch": 0.12336920579570733,
      "flos": 18915407533440.0,
      "grad_norm": 1.8457799696127073,
      "language_loss": 0.69466639,
      "learning_rate": 3.9093436605213144e-06,
      "loss": 0.71723723,
      "num_input_tokens_seen": 21688500,
      "step": 1026,
      "time_per_iteration": 2.6372241973876953
    },
    {
      "auxiliary_loss_clip": 0.01206877,
      "auxiliary_loss_mlp": 0.010489,
      "balance_loss_clip": 1.06031322,
      "balance_loss_mlp": 1.03750992,
      "epoch": 0.12348944868634643,
      "flos": 23878369797120.0,
      "grad_norm": 1.8186932862759035,
      "language_loss": 0.79729408,
      "learning_rate": 3.909111646714627e-06,
      "loss": 0.81985188,
      "num_input_tokens_seen": 21709345,
      "step": 1027,
      "time_per_iteration": 2.6928157806396484
    },
    {
      "auxiliary_loss_clip": 0.01237321,
      "auxiliary_loss_mlp": 0.01034235,
      "balance_loss_clip": 1.06371367,
      "balance_loss_mlp": 1.02283895,
      "epoch": 0.12360969157698551,
      "flos": 19026084314880.0,
      "grad_norm": 2.066197549482594,
      "language_loss": 0.72243047,
      "learning_rate": 3.9088793432972206e-06,
      "loss": 0.74514604,
      "num_input_tokens_seen": 21728165,
      "step": 1028,
      "time_per_iteration": 2.6527819633483887
    },
    {
      "auxiliary_loss_clip": 0.01181583,
      "auxiliary_loss_mlp": 0.01044987,
      "balance_loss_clip": 1.05958152,
      "balance_loss_mlp": 1.03407979,
      "epoch": 0.1237299344676246,
      "flos": 13224607983360.0,
      "grad_norm": 2.0138793474776984,
      "language_loss": 0.8236081,
      "learning_rate": 3.908646750304336e-06,
      "loss": 0.84587377,
      "num_input_tokens_seen": 21745850,
      "step": 1029,
      "time_per_iteration": 2.7002947330474854
    },
    {
      "auxiliary_loss_clip": 0.01211712,
      "auxiliary_loss_mlp": 0.01055454,
      "balance_loss_clip": 1.06326294,
      "balance_loss_mlp": 1.0434792,
      "epoch": 0.12385017735826369,
      "flos": 20485673470080.0,
      "grad_norm": 1.6877638081388653,
      "language_loss": 0.87112778,
      "learning_rate": 3.908413867771257e-06,
      "loss": 0.89379942,
      "num_input_tokens_seen": 21764760,
      "step": 1030,
      "time_per_iteration": 2.7035481929779053
    },
    {
      "auxiliary_loss_clip": 0.01222017,
      "auxiliary_loss_mlp": 0.01043375,
      "balance_loss_clip": 1.06317127,
      "balance_loss_mlp": 1.03201437,
      "epoch": 0.12397042024890279,
      "flos": 17347835116800.0,
      "grad_norm": 1.6668902048060372,
      "language_loss": 0.80655205,
      "learning_rate": 3.908180695733311e-06,
      "loss": 0.82920605,
      "num_input_tokens_seen": 21784250,
      "step": 1031,
      "time_per_iteration": 3.73565936088562
    },
    {
      "auxiliary_loss_clip": 0.01151903,
      "auxiliary_loss_mlp": 0.01039364,
      "balance_loss_clip": 1.05042434,
      "balance_loss_mlp": 1.02719843,
      "epoch": 0.12409066313954187,
      "flos": 20412343854720.0,
      "grad_norm": 1.9692088083761699,
      "language_loss": 0.82474846,
      "learning_rate": 3.907947234225871e-06,
      "loss": 0.84666115,
      "num_input_tokens_seen": 21803260,
      "step": 1032,
      "time_per_iteration": 2.770810842514038
    },
    {
      "auxiliary_loss_clip": 0.01154066,
      "auxiliary_loss_mlp": 0.01045037,
      "balance_loss_clip": 1.05274415,
      "balance_loss_mlp": 1.03344965,
      "epoch": 0.12421090603018096,
      "flos": 20736688688640.0,
      "grad_norm": 2.1144716535253747,
      "language_loss": 0.8690263,
      "learning_rate": 3.907713483284352e-06,
      "loss": 0.89101732,
      "num_input_tokens_seen": 21822735,
      "step": 1033,
      "time_per_iteration": 2.7912769317626953
    },
    {
      "auxiliary_loss_clip": 0.01135078,
      "auxiliary_loss_mlp": 0.01050138,
      "balance_loss_clip": 1.04941773,
      "balance_loss_mlp": 1.03854489,
      "epoch": 0.12433114892082006,
      "flos": 24498834353280.0,
      "grad_norm": 2.4920298503669795,
      "language_loss": 0.97180855,
      "learning_rate": 3.907479442944216e-06,
      "loss": 0.99366069,
      "num_input_tokens_seen": 21841140,
      "step": 1034,
      "time_per_iteration": 3.0524399280548096
    },
    {
      "auxiliary_loss_clip": 0.01222183,
      "auxiliary_loss_mlp": 0.01042276,
      "balance_loss_clip": 1.06443906,
      "balance_loss_mlp": 1.03123713,
      "epoch": 0.12445139181145914,
      "flos": 19682315838720.0,
      "grad_norm": 2.2645677634719936,
      "language_loss": 0.92359698,
      "learning_rate": 3.907245113240963e-06,
      "loss": 0.94624162,
      "num_input_tokens_seen": 21859260,
      "step": 1035,
      "time_per_iteration": 3.8346590995788574
    },
    {
      "auxiliary_loss_clip": 0.01186464,
      "auxiliary_loss_mlp": 0.01049417,
      "balance_loss_clip": 1.05460072,
      "balance_loss_mlp": 1.0370611,
      "epoch": 0.12457163470209824,
      "flos": 46423087522560.0,
      "grad_norm": 1.7959883945230293,
      "language_loss": 0.73786026,
      "learning_rate": 3.907010494210144e-06,
      "loss": 0.7602191,
      "num_input_tokens_seen": 21881920,
      "step": 1036,
      "time_per_iteration": 2.892209768295288
    },
    {
      "auxiliary_loss_clip": 0.01227789,
      "auxiliary_loss_mlp": 0.01035422,
      "balance_loss_clip": 1.06530428,
      "balance_loss_mlp": 1.02364993,
      "epoch": 0.12469187759273732,
      "flos": 20376289578240.0,
      "grad_norm": 2.2480390252644105,
      "language_loss": 0.91834772,
      "learning_rate": 3.9067755858873495e-06,
      "loss": 0.94097984,
      "num_input_tokens_seen": 21898720,
      "step": 1037,
      "time_per_iteration": 2.655383825302124
    },
    {
      "auxiliary_loss_clip": 0.01102238,
      "auxiliary_loss_mlp": 0.0101771,
      "balance_loss_clip": 1.02810752,
      "balance_loss_mlp": 1.01387143,
      "epoch": 0.12481212048337642,
      "flos": 69224641447680.0,
      "grad_norm": 0.866247275807406,
      "language_loss": 0.62776899,
      "learning_rate": 3.906540388308214e-06,
      "loss": 0.64896846,
      "num_input_tokens_seen": 21958305,
      "step": 1038,
      "time_per_iteration": 3.3002307415008545
    },
    {
      "auxiliary_loss_clip": 0.01160858,
      "auxiliary_loss_mlp": 0.01036866,
      "balance_loss_clip": 1.05657113,
      "balance_loss_mlp": 1.02579701,
      "epoch": 0.12493236337401552,
      "flos": 18223696350720.0,
      "grad_norm": 2.028309569430993,
      "language_loss": 0.81432915,
      "learning_rate": 3.906304901508417e-06,
      "loss": 0.83630633,
      "num_input_tokens_seen": 21977205,
      "step": 1039,
      "time_per_iteration": 2.698611259460449
    },
    {
      "auxiliary_loss_clip": 0.01226121,
      "auxiliary_loss_mlp": 0.01045218,
      "balance_loss_clip": 1.06595254,
      "balance_loss_mlp": 1.03415537,
      "epoch": 0.12505260626465461,
      "flos": 30044375303040.0,
      "grad_norm": 2.199682275757682,
      "language_loss": 0.75662541,
      "learning_rate": 3.9060691255236835e-06,
      "loss": 0.77933884,
      "num_input_tokens_seen": 21997770,
      "step": 1040,
      "time_per_iteration": 2.743642807006836
    },
    {
      "auxiliary_loss_clip": 0.01218266,
      "auxiliary_loss_mlp": 0.01050941,
      "balance_loss_clip": 1.0598948,
      "balance_loss_mlp": 1.03973496,
      "epoch": 0.1251728491552937,
      "flos": 24433980347520.0,
      "grad_norm": 1.7495995931455124,
      "language_loss": 0.80553901,
      "learning_rate": 3.905833060389778e-06,
      "loss": 0.8282311,
      "num_input_tokens_seen": 22021890,
      "step": 1041,
      "time_per_iteration": 2.711275815963745
    },
    {
      "auxiliary_loss_clip": 0.01240483,
      "auxiliary_loss_mlp": 0.00765839,
      "balance_loss_clip": 1.06730604,
      "balance_loss_mlp": 1.00025833,
      "epoch": 0.12529309204593278,
      "flos": 27119809952640.0,
      "grad_norm": 2.057146825837037,
      "language_loss": 0.78258127,
      "learning_rate": 3.905596706142513e-06,
      "loss": 0.80264449,
      "num_input_tokens_seen": 22043300,
      "step": 1042,
      "time_per_iteration": 2.7108614444732666
    },
    {
      "auxiliary_loss_clip": 0.01187402,
      "auxiliary_loss_mlp": 0.01039451,
      "balance_loss_clip": 1.0571909,
      "balance_loss_mlp": 1.02823925,
      "epoch": 0.12541333493657186,
      "flos": 30774151923840.0,
      "grad_norm": 2.0307786263864736,
      "language_loss": 0.85716456,
      "learning_rate": 3.9053600628177435e-06,
      "loss": 0.8794331,
      "num_input_tokens_seen": 22062910,
      "step": 1043,
      "time_per_iteration": 2.827369213104248
    },
    {
      "auxiliary_loss_clip": 0.01235281,
      "auxiliary_loss_mlp": 0.01035742,
      "balance_loss_clip": 1.06225777,
      "balance_loss_mlp": 1.02415442,
      "epoch": 0.12553357782721097,
      "flos": 23659566099840.0,
      "grad_norm": 1.9416982344498939,
      "language_loss": 0.84755492,
      "learning_rate": 3.905123130451367e-06,
      "loss": 0.87026525,
      "num_input_tokens_seen": 22084010,
      "step": 1044,
      "time_per_iteration": 2.6386501789093018
    },
    {
      "auxiliary_loss_clip": 0.01235743,
      "auxiliary_loss_mlp": 0.01042301,
      "balance_loss_clip": 1.06345475,
      "balance_loss_mlp": 1.03059435,
      "epoch": 0.12565382071785006,
      "flos": 24863758577280.0,
      "grad_norm": 2.8232633924260337,
      "language_loss": 0.78900093,
      "learning_rate": 3.904885909079326e-06,
      "loss": 0.81178141,
      "num_input_tokens_seen": 22102795,
      "step": 1045,
      "time_per_iteration": 2.615800142288208
    },
    {
      "auxiliary_loss_clip": 0.01222249,
      "auxiliary_loss_mlp": 0.01048381,
      "balance_loss_clip": 1.0627712,
      "balance_loss_mlp": 1.03635907,
      "epoch": 0.12577406360848914,
      "flos": 21360780518400.0,
      "grad_norm": 2.9764906229022428,
      "language_loss": 0.77890193,
      "learning_rate": 3.904648398737607e-06,
      "loss": 0.80160826,
      "num_input_tokens_seen": 22121360,
      "step": 1046,
      "time_per_iteration": 2.857059955596924
    },
    {
      "auxiliary_loss_clip": 0.01234725,
      "auxiliary_loss_mlp": 0.01036989,
      "balance_loss_clip": 1.06298566,
      "balance_loss_mlp": 1.02581882,
      "epoch": 0.12589430649912825,
      "flos": 36138056774400.0,
      "grad_norm": 1.8871588868468498,
      "language_loss": 0.78104854,
      "learning_rate": 3.9044105994622406e-06,
      "loss": 0.80376571,
      "num_input_tokens_seen": 22142505,
      "step": 1047,
      "time_per_iteration": 2.8072240352630615
    },
    {
      "auxiliary_loss_clip": 0.01211216,
      "auxiliary_loss_mlp": 0.00765871,
      "balance_loss_clip": 1.06199956,
      "balance_loss_mlp": 1.00024652,
      "epoch": 0.12601454938976733,
      "flos": 25337671643520.0,
      "grad_norm": 2.192955793430223,
      "language_loss": 0.81682098,
      "learning_rate": 3.9041725112893005e-06,
      "loss": 0.83659184,
      "num_input_tokens_seen": 22163730,
      "step": 1048,
      "time_per_iteration": 2.744117259979248
    },
    {
      "auxiliary_loss_clip": 0.01190946,
      "auxiliary_loss_mlp": 0.01043072,
      "balance_loss_clip": 1.06274557,
      "balance_loss_mlp": 1.03195572,
      "epoch": 0.12613479228040642,
      "flos": 15560094286080.0,
      "grad_norm": 2.6765530802726034,
      "language_loss": 0.75256962,
      "learning_rate": 3.903934134254904e-06,
      "loss": 0.77490985,
      "num_input_tokens_seen": 22181520,
      "step": 1049,
      "time_per_iteration": 2.733605146408081
    },
    {
      "auxiliary_loss_clip": 0.0122719,
      "auxiliary_loss_mlp": 0.01035764,
      "balance_loss_clip": 1.06268108,
      "balance_loss_mlp": 1.02390289,
      "epoch": 0.1262550351710455,
      "flos": 21470595373440.0,
      "grad_norm": 2.7045057625861,
      "language_loss": 0.85582596,
      "learning_rate": 3.903695468395213e-06,
      "loss": 0.87845552,
      "num_input_tokens_seen": 22199390,
      "step": 1050,
      "time_per_iteration": 2.7244279384613037
    },
    {
      "auxiliary_loss_clip": 0.01208956,
      "auxiliary_loss_mlp": 0.01042426,
      "balance_loss_clip": 1.05702412,
      "balance_loss_mlp": 1.0305171,
      "epoch": 0.1263752780616846,
      "flos": 31576719456000.0,
      "grad_norm": 3.6590693709016753,
      "language_loss": 0.56098348,
      "learning_rate": 3.903456513746434e-06,
      "loss": 0.58349729,
      "num_input_tokens_seen": 22220365,
      "step": 1051,
      "time_per_iteration": 2.882082462310791
    },
    {
      "auxiliary_loss_clip": 0.01236676,
      "auxiliary_loss_mlp": 0.01036846,
      "balance_loss_clip": 1.06408322,
      "balance_loss_mlp": 1.02667701,
      "epoch": 0.1264955209523237,
      "flos": 28768217927040.0,
      "grad_norm": 2.137058736443117,
      "language_loss": 0.87869394,
      "learning_rate": 3.903217270344815e-06,
      "loss": 0.90142918,
      "num_input_tokens_seen": 22240615,
      "step": 1052,
      "time_per_iteration": 2.7169365882873535
    },
    {
      "auxiliary_loss_clip": 0.01183918,
      "auxiliary_loss_mlp": 0.01041025,
      "balance_loss_clip": 1.05434382,
      "balance_loss_mlp": 1.02958727,
      "epoch": 0.12661576384296278,
      "flos": 29241125412480.0,
      "grad_norm": 3.2102904495582836,
      "language_loss": 0.82816863,
      "learning_rate": 3.902977738226648e-06,
      "loss": 0.85041803,
      "num_input_tokens_seen": 22261350,
      "step": 1053,
      "time_per_iteration": 2.761601686477661
    },
    {
      "auxiliary_loss_clip": 0.01222237,
      "auxiliary_loss_mlp": 0.01036744,
      "balance_loss_clip": 1.06330395,
      "balance_loss_mlp": 1.02572322,
      "epoch": 0.12673600673360189,
      "flos": 20850346298880.0,
      "grad_norm": 1.820013316374067,
      "language_loss": 0.91271973,
      "learning_rate": 3.902737917428273e-06,
      "loss": 0.93530953,
      "num_input_tokens_seen": 22279515,
      "step": 1054,
      "time_per_iteration": 2.636953592300415
    },
    {
      "auxiliary_loss_clip": 0.01236978,
      "auxiliary_loss_mlp": 0.01035642,
      "balance_loss_clip": 1.06408501,
      "balance_loss_mlp": 1.02413273,
      "epoch": 0.12685624962424097,
      "flos": 25263695583360.0,
      "grad_norm": 2.1988515504703687,
      "language_loss": 0.84212232,
      "learning_rate": 3.902497807986068e-06,
      "loss": 0.86484855,
      "num_input_tokens_seen": 22299535,
      "step": 1055,
      "time_per_iteration": 2.653538465499878
    },
    {
      "auxiliary_loss_clip": 0.011917,
      "auxiliary_loss_mlp": 0.01036745,
      "balance_loss_clip": 1.05470276,
      "balance_loss_mlp": 1.02516365,
      "epoch": 0.12697649251488005,
      "flos": 27527109246720.0,
      "grad_norm": 1.9942215181809144,
      "language_loss": 0.83942252,
      "learning_rate": 3.902257409936458e-06,
      "loss": 0.86170697,
      "num_input_tokens_seen": 22320300,
      "step": 1056,
      "time_per_iteration": 2.786736011505127
    },
    {
      "auxiliary_loss_clip": 0.01203202,
      "auxiliary_loss_mlp": 0.01035728,
      "balance_loss_clip": 1.06046379,
      "balance_loss_mlp": 1.02507615,
      "epoch": 0.12709673540551916,
      "flos": 21251863503360.0,
      "grad_norm": 2.3064929944173604,
      "language_loss": 0.84262311,
      "learning_rate": 3.902016723315912e-06,
      "loss": 0.86501241,
      "num_input_tokens_seen": 22338240,
      "step": 1057,
      "time_per_iteration": 3.560701847076416
    },
    {
      "auxiliary_loss_clip": 0.01216642,
      "auxiliary_loss_mlp": 0.01037329,
      "balance_loss_clip": 1.05772209,
      "balance_loss_mlp": 1.0265944,
      "epoch": 0.12721697829615825,
      "flos": 25337707557120.0,
      "grad_norm": 2.556425578173309,
      "language_loss": 0.69383103,
      "learning_rate": 3.901775748160941e-06,
      "loss": 0.71637076,
      "num_input_tokens_seen": 22357420,
      "step": 1058,
      "time_per_iteration": 2.6972365379333496
    },
    {
      "auxiliary_loss_clip": 0.01113968,
      "auxiliary_loss_mlp": 0.01016283,
      "balance_loss_clip": 1.03210139,
      "balance_loss_mlp": 1.01304078,
      "epoch": 0.12733722118679733,
      "flos": 61943287754880.0,
      "grad_norm": 0.798876760146273,
      "language_loss": 0.60854244,
      "learning_rate": 3.901534484508101e-06,
      "loss": 0.62984502,
      "num_input_tokens_seen": 22420095,
      "step": 1059,
      "time_per_iteration": 3.2554681301116943
    },
    {
      "auxiliary_loss_clip": 0.01192901,
      "auxiliary_loss_mlp": 0.01035464,
      "balance_loss_clip": 1.05592012,
      "balance_loss_mlp": 1.0240432,
      "epoch": 0.1274574640774364,
      "flos": 26976742081920.0,
      "grad_norm": 1.766505421912866,
      "language_loss": 0.74543524,
      "learning_rate": 3.901292932393991e-06,
      "loss": 0.76771891,
      "num_input_tokens_seen": 22438975,
      "step": 1060,
      "time_per_iteration": 2.769479990005493
    },
    {
      "auxiliary_loss_clip": 0.01237651,
      "auxiliary_loss_mlp": 0.01045327,
      "balance_loss_clip": 1.06697166,
      "balance_loss_mlp": 1.03396678,
      "epoch": 0.12757770696807552,
      "flos": 22236318529920.0,
      "grad_norm": 2.9230540707977415,
      "language_loss": 0.85324538,
      "learning_rate": 3.9010510918552555e-06,
      "loss": 0.87607515,
      "num_input_tokens_seen": 22458050,
      "step": 1061,
      "time_per_iteration": 4.85273289680481
    },
    {
      "auxiliary_loss_clip": 0.01197541,
      "auxiliary_loss_mlp": 0.01044727,
      "balance_loss_clip": 1.05429363,
      "balance_loss_mlp": 1.03251982,
      "epoch": 0.1276979498587146,
      "flos": 28547905858560.0,
      "grad_norm": 2.4842629684057704,
      "language_loss": 0.74904287,
      "learning_rate": 3.900808962928581e-06,
      "loss": 0.7714656,
      "num_input_tokens_seen": 22475665,
      "step": 1062,
      "time_per_iteration": 2.7578938007354736
    },
    {
      "auxiliary_loss_clip": 0.0123426,
      "auxiliary_loss_mlp": 0.01040378,
      "balance_loss_clip": 1.06349015,
      "balance_loss_mlp": 1.02972043,
      "epoch": 0.1278181927493537,
      "flos": 17420338719360.0,
      "grad_norm": 2.3770196167955473,
      "language_loss": 0.89829278,
      "learning_rate": 3.900566545650698e-06,
      "loss": 0.92103916,
      "num_input_tokens_seen": 22493335,
      "step": 1063,
      "time_per_iteration": 2.6138265132904053
    },
    {
      "auxiliary_loss_clip": 0.01219617,
      "auxiliary_loss_mlp": 0.01039962,
      "balance_loss_clip": 1.06359696,
      "balance_loss_mlp": 1.0277853,
      "epoch": 0.1279384356399928,
      "flos": 21138636856320.0,
      "grad_norm": 2.674173751967933,
      "language_loss": 0.81823325,
      "learning_rate": 3.900323840058381e-06,
      "loss": 0.84082901,
      "num_input_tokens_seen": 22511045,
      "step": 1064,
      "time_per_iteration": 2.707716703414917
    },
    {
      "auxiliary_loss_clip": 0.01219082,
      "auxiliary_loss_mlp": 0.0103505,
      "balance_loss_clip": 1.05900753,
      "balance_loss_mlp": 1.02430308,
      "epoch": 0.12805867853063188,
      "flos": 26576733248640.0,
      "grad_norm": 1.8350418296813011,
      "language_loss": 0.81643486,
      "learning_rate": 3.900080846188449e-06,
      "loss": 0.8389762,
      "num_input_tokens_seen": 22529635,
      "step": 1065,
      "time_per_iteration": 2.669092893600464
    },
    {
      "auxiliary_loss_clip": 0.01233326,
      "auxiliary_loss_mlp": 0.01040927,
      "balance_loss_clip": 1.06184471,
      "balance_loss_mlp": 1.02911937,
      "epoch": 0.12817892142127096,
      "flos": 16436206915200.0,
      "grad_norm": 1.9865116618456167,
      "language_loss": 0.81413066,
      "learning_rate": 3.8998375640777625e-06,
      "loss": 0.83687317,
      "num_input_tokens_seen": 22547505,
      "step": 1066,
      "time_per_iteration": 2.6514127254486084
    },
    {
      "auxiliary_loss_clip": 0.01130801,
      "auxiliary_loss_mlp": 0.01007337,
      "balance_loss_clip": 1.04696548,
      "balance_loss_mlp": 1.00376117,
      "epoch": 0.12829916431191005,
      "flos": 60757049099520.0,
      "grad_norm": 0.8071888930315945,
      "language_loss": 0.52612066,
      "learning_rate": 3.899593993763229e-06,
      "loss": 0.54750204,
      "num_input_tokens_seen": 22608465,
      "step": 1067,
      "time_per_iteration": 3.2835206985473633
    },
    {
      "auxiliary_loss_clip": 0.01181333,
      "auxiliary_loss_mlp": 0.01042632,
      "balance_loss_clip": 1.05670452,
      "balance_loss_mlp": 1.03140831,
      "epoch": 0.12841940720254916,
      "flos": 29786895636480.0,
      "grad_norm": 2.160559809724421,
      "language_loss": 0.81916666,
      "learning_rate": 3.899350135281796e-06,
      "loss": 0.84140635,
      "num_input_tokens_seen": 22629465,
      "step": 1068,
      "time_per_iteration": 2.830223560333252
    },
    {
      "auxiliary_loss_clip": 0.01193624,
      "auxiliary_loss_mlp": 0.01039066,
      "balance_loss_clip": 1.05892146,
      "balance_loss_mlp": 1.02828896,
      "epoch": 0.12853965009318824,
      "flos": 25951851319680.0,
      "grad_norm": 2.014659319365307,
      "language_loss": 0.79925734,
      "learning_rate": 3.8991059886704585e-06,
      "loss": 0.82158422,
      "num_input_tokens_seen": 22648970,
      "step": 1069,
      "time_per_iteration": 2.8045356273651123
    },
    {
      "auxiliary_loss_clip": 0.0118362,
      "auxiliary_loss_mlp": 0.01040102,
      "balance_loss_clip": 1.05872977,
      "balance_loss_mlp": 1.02861023,
      "epoch": 0.12865989298382732,
      "flos": 30846871008000.0,
      "grad_norm": 2.150975240117665,
      "language_loss": 0.82911611,
      "learning_rate": 3.898861553966252e-06,
      "loss": 0.85135329,
      "num_input_tokens_seen": 22668620,
      "step": 1070,
      "time_per_iteration": 2.8105504512786865
    },
    {
      "auxiliary_loss_clip": 0.01142522,
      "auxiliary_loss_mlp": 0.01039841,
      "balance_loss_clip": 1.05012536,
      "balance_loss_mlp": 1.02904654,
      "epoch": 0.12878013587446643,
      "flos": 25885776251520.0,
      "grad_norm": 1.620895140793095,
      "language_loss": 0.88055265,
      "learning_rate": 3.898616831206257e-06,
      "loss": 0.90237629,
      "num_input_tokens_seen": 22689045,
      "step": 1071,
      "time_per_iteration": 2.937422275543213
    },
    {
      "auxiliary_loss_clip": 0.01181227,
      "auxiliary_loss_mlp": 0.01045001,
      "balance_loss_clip": 1.05263722,
      "balance_loss_mlp": 1.03314543,
      "epoch": 0.12890037876510552,
      "flos": 23333138277120.0,
      "grad_norm": 1.9696164750286227,
      "language_loss": 0.76916087,
      "learning_rate": 3.8983718204276e-06,
      "loss": 0.7914232,
      "num_input_tokens_seen": 22711265,
      "step": 1072,
      "time_per_iteration": 2.799363613128662
    },
    {
      "auxiliary_loss_clip": 0.01204179,
      "auxiliary_loss_mlp": 0.01040309,
      "balance_loss_clip": 1.05744338,
      "balance_loss_mlp": 1.0291748,
      "epoch": 0.1290206216557446,
      "flos": 23587242065280.0,
      "grad_norm": 3.9786945919320647,
      "language_loss": 0.82347155,
      "learning_rate": 3.898126521667446e-06,
      "loss": 0.84591645,
      "num_input_tokens_seen": 22731420,
      "step": 1073,
      "time_per_iteration": 2.7921648025512695
    },
    {
      "auxiliary_loss_clip": 0.01220672,
      "auxiliary_loss_mlp": 0.01040077,
      "balance_loss_clip": 1.06165624,
      "balance_loss_mlp": 1.02827537,
      "epoch": 0.12914086454638368,
      "flos": 24170610850560.0,
      "grad_norm": 1.6770865917990314,
      "language_loss": 0.83236742,
      "learning_rate": 3.897880934963007e-06,
      "loss": 0.85497493,
      "num_input_tokens_seen": 22750970,
      "step": 1074,
      "time_per_iteration": 2.6896181106567383
    },
    {
      "auxiliary_loss_clip": 0.01198033,
      "auxiliary_loss_mlp": 0.01042275,
      "balance_loss_clip": 1.05532527,
      "balance_loss_mlp": 1.03106952,
      "epoch": 0.1292611074370228,
      "flos": 20267157081600.0,
      "grad_norm": 2.3132023180962418,
      "language_loss": 0.7876969,
      "learning_rate": 3.89763506035154e-06,
      "loss": 0.81009996,
      "num_input_tokens_seen": 22768820,
      "step": 1075,
      "time_per_iteration": 2.739361047744751
    },
    {
      "auxiliary_loss_clip": 0.01206987,
      "auxiliary_loss_mlp": 0.0103626,
      "balance_loss_clip": 1.05725479,
      "balance_loss_mlp": 1.02519774,
      "epoch": 0.12938135032766188,
      "flos": 27377684668800.0,
      "grad_norm": 1.7018010414803082,
      "language_loss": 0.81401211,
      "learning_rate": 3.897388897870343e-06,
      "loss": 0.83644462,
      "num_input_tokens_seen": 22789460,
      "step": 1076,
      "time_per_iteration": 2.751187562942505
    },
    {
      "auxiliary_loss_clip": 0.01213479,
      "auxiliary_loss_mlp": 0.01040322,
      "balance_loss_clip": 1.05825496,
      "balance_loss_mlp": 1.02873468,
      "epoch": 0.12950159321830096,
      "flos": 29277107861760.0,
      "grad_norm": 3.708839246961591,
      "language_loss": 0.75073683,
      "learning_rate": 3.89714244755676e-06,
      "loss": 0.77327484,
      "num_input_tokens_seen": 22810820,
      "step": 1077,
      "time_per_iteration": 2.834130048751831
    },
    {
      "auxiliary_loss_clip": 0.01155544,
      "auxiliary_loss_mlp": 0.01045835,
      "balance_loss_clip": 1.04715598,
      "balance_loss_mlp": 1.03367567,
      "epoch": 0.12962183610894007,
      "flos": 24534888629760.0,
      "grad_norm": 2.7983480906951224,
      "language_loss": 0.86467409,
      "learning_rate": 3.896895709448175e-06,
      "loss": 0.88668787,
      "num_input_tokens_seen": 22830570,
      "step": 1078,
      "time_per_iteration": 2.7367212772369385
    },
    {
      "auxiliary_loss_clip": 0.0115194,
      "auxiliary_loss_mlp": 0.01042561,
      "balance_loss_clip": 1.04961538,
      "balance_loss_mlp": 1.0310514,
      "epoch": 0.12974207899957915,
      "flos": 11215944552960.0,
      "grad_norm": 3.0087592756580124,
      "language_loss": 0.77225995,
      "learning_rate": 3.896648683582019e-06,
      "loss": 0.79420495,
      "num_input_tokens_seen": 22845905,
      "step": 1079,
      "time_per_iteration": 2.7789998054504395
    },
    {
      "auxiliary_loss_clip": 0.01174121,
      "auxiliary_loss_mlp": 0.01037101,
      "balance_loss_clip": 1.05867374,
      "balance_loss_mlp": 1.0264194,
      "epoch": 0.12986232189021824,
      "flos": 24717889445760.0,
      "grad_norm": 2.182101251924361,
      "language_loss": 0.80719298,
      "learning_rate": 3.896401369995766e-06,
      "loss": 0.82930523,
      "num_input_tokens_seen": 22865710,
      "step": 1080,
      "time_per_iteration": 2.873300552368164
    },
    {
      "auxiliary_loss_clip": 0.01233011,
      "auxiliary_loss_mlp": 0.01037571,
      "balance_loss_clip": 1.06226301,
      "balance_loss_mlp": 1.02643728,
      "epoch": 0.12998256478085732,
      "flos": 23915357827200.0,
      "grad_norm": 1.7951023402405335,
      "language_loss": 0.79566765,
      "learning_rate": 3.896153768726932e-06,
      "loss": 0.81837344,
      "num_input_tokens_seen": 22886020,
      "step": 1081,
      "time_per_iteration": 2.732997417449951
    },
    {
      "auxiliary_loss_clip": 0.01221374,
      "auxiliary_loss_mlp": 0.01039582,
      "balance_loss_clip": 1.06381392,
      "balance_loss_mlp": 1.02795959,
      "epoch": 0.13010280767149643,
      "flos": 18624207974400.0,
      "grad_norm": 2.280801468611434,
      "language_loss": 0.88251865,
      "learning_rate": 3.8959058798130806e-06,
      "loss": 0.90512824,
      "num_input_tokens_seen": 22903995,
      "step": 1082,
      "time_per_iteration": 2.659595251083374
    },
    {
      "auxiliary_loss_clip": 0.01201629,
      "auxiliary_loss_mlp": 0.00765898,
      "balance_loss_clip": 1.05810475,
      "balance_loss_mlp": 1.00033975,
      "epoch": 0.1302230505621355,
      "flos": 22783992174720.0,
      "grad_norm": 1.756937336143744,
      "language_loss": 0.74862212,
      "learning_rate": 3.895657703291814e-06,
      "loss": 0.76829737,
      "num_input_tokens_seen": 22924100,
      "step": 1083,
      "time_per_iteration": 3.6165482997894287
    },
    {
      "auxiliary_loss_clip": 0.01210244,
      "auxiliary_loss_mlp": 0.01036063,
      "balance_loss_clip": 1.05787182,
      "balance_loss_mlp": 1.02393985,
      "epoch": 0.1303432934527746,
      "flos": 21323612920320.0,
      "grad_norm": 3.2739882441738,
      "language_loss": 0.7936691,
      "learning_rate": 3.895409239200781e-06,
      "loss": 0.81613219,
      "num_input_tokens_seen": 22939985,
      "step": 1084,
      "time_per_iteration": 2.6997861862182617
    },
    {
      "auxiliary_loss_clip": 0.01207569,
      "auxiliary_loss_mlp": 0.01038799,
      "balance_loss_clip": 1.05651629,
      "balance_loss_mlp": 1.02730763,
      "epoch": 0.1304635363434137,
      "flos": 20922490765440.0,
      "grad_norm": 2.341261451370678,
      "language_loss": 0.91384739,
      "learning_rate": 3.895160487577673e-06,
      "loss": 0.93631113,
      "num_input_tokens_seen": 22957555,
      "step": 1085,
      "time_per_iteration": 2.704479694366455
    },
    {
      "auxiliary_loss_clip": 0.0111998,
      "auxiliary_loss_mlp": 0.0101245,
      "balance_loss_clip": 1.03094923,
      "balance_loss_mlp": 1.00873089,
      "epoch": 0.1305837792340528,
      "flos": 63245659080960.0,
      "grad_norm": 0.7894819968500649,
      "language_loss": 0.60918617,
      "learning_rate": 3.894911448460226e-06,
      "loss": 0.63051045,
      "num_input_tokens_seen": 23016870,
      "step": 1086,
      "time_per_iteration": 3.1110501289367676
    },
    {
      "auxiliary_loss_clip": 0.01118464,
      "auxiliary_loss_mlp": 0.01039785,
      "balance_loss_clip": 1.04593015,
      "balance_loss_mlp": 1.02813816,
      "epoch": 0.13070402212469187,
      "flos": 26428852955520.0,
      "grad_norm": 2.1994215901221303,
      "language_loss": 0.72558993,
      "learning_rate": 3.8946621218862195e-06,
      "loss": 0.74717236,
      "num_input_tokens_seen": 23037870,
      "step": 1087,
      "time_per_iteration": 4.6585118770599365
    },
    {
      "auxiliary_loss_clip": 0.01184638,
      "auxiliary_loss_mlp": 0.01034235,
      "balance_loss_clip": 1.05637622,
      "balance_loss_mlp": 1.02360749,
      "epoch": 0.13082426501533098,
      "flos": 27673409341440.0,
      "grad_norm": 1.9108873457378215,
      "language_loss": 0.88826632,
      "learning_rate": 3.894412507893475e-06,
      "loss": 0.91045499,
      "num_input_tokens_seen": 23058150,
      "step": 1088,
      "time_per_iteration": 2.8063507080078125
    },
    {
      "auxiliary_loss_clip": 0.01177255,
      "auxiliary_loss_mlp": 0.01038468,
      "balance_loss_clip": 1.05203068,
      "balance_loss_mlp": 1.02625453,
      "epoch": 0.13094450790597006,
      "flos": 24826770547200.0,
      "grad_norm": 4.639505596642658,
      "language_loss": 0.72179878,
      "learning_rate": 3.894162606519859e-06,
      "loss": 0.74395597,
      "num_input_tokens_seen": 23077100,
      "step": 1089,
      "time_per_iteration": 2.7616124153137207
    },
    {
      "auxiliary_loss_clip": 0.01172562,
      "auxiliary_loss_mlp": 0.01041618,
      "balance_loss_clip": 1.05457127,
      "balance_loss_mlp": 1.03031731,
      "epoch": 0.13106475079660915,
      "flos": 19062605468160.0,
      "grad_norm": 2.0353053773280845,
      "language_loss": 0.77093685,
      "learning_rate": 3.893912417803282e-06,
      "loss": 0.79307866,
      "num_input_tokens_seen": 23096815,
      "step": 1090,
      "time_per_iteration": 2.8263602256774902
    },
    {
      "auxiliary_loss_clip": 0.01165562,
      "auxiliary_loss_mlp": 0.01039427,
      "balance_loss_clip": 1.04913771,
      "balance_loss_mlp": 1.02798295,
      "epoch": 0.13118499368724823,
      "flos": 28913189218560.0,
      "grad_norm": 2.2737363490734106,
      "language_loss": 0.76612949,
      "learning_rate": 3.8936619417816975e-06,
      "loss": 0.7881794,
      "num_input_tokens_seen": 23117145,
      "step": 1091,
      "time_per_iteration": 2.8181614875793457
    },
    {
      "auxiliary_loss_clip": 0.01187025,
      "auxiliary_loss_mlp": 0.010471,
      "balance_loss_clip": 1.05856049,
      "balance_loss_mlp": 1.03533411,
      "epoch": 0.13130523657788734,
      "flos": 14283398206080.0,
      "grad_norm": 1.968124414740172,
      "language_loss": 0.71391261,
      "learning_rate": 3.8934111784931015e-06,
      "loss": 0.73625386,
      "num_input_tokens_seen": 23134595,
      "step": 1092,
      "time_per_iteration": 2.6925787925720215
    },
    {
      "auxiliary_loss_clip": 0.01109423,
      "auxiliary_loss_mlp": 0.01006085,
      "balance_loss_clip": 1.02961278,
      "balance_loss_mlp": 1.00234187,
      "epoch": 0.13142547946852642,
      "flos": 70174155519360.0,
      "grad_norm": 0.9159114935045927,
      "language_loss": 0.59043086,
      "learning_rate": 3.893160127975535e-06,
      "loss": 0.61158592,
      "num_input_tokens_seen": 23195285,
      "step": 1093,
      "time_per_iteration": 3.3218305110931396
    },
    {
      "auxiliary_loss_clip": 0.01178296,
      "auxiliary_loss_mlp": 0.01036219,
      "balance_loss_clip": 1.05390394,
      "balance_loss_mlp": 1.02592492,
      "epoch": 0.1315457223591655,
      "flos": 45805998844800.0,
      "grad_norm": 2.700905205803401,
      "language_loss": 0.81337857,
      "learning_rate": 3.8929087902670826e-06,
      "loss": 0.83552372,
      "num_input_tokens_seen": 23216915,
      "step": 1094,
      "time_per_iteration": 3.0302677154541016
    },
    {
      "auxiliary_loss_clip": 0.0112343,
      "auxiliary_loss_mlp": 0.01003488,
      "balance_loss_clip": 1.02861845,
      "balance_loss_mlp": 0.99988824,
      "epoch": 0.13166596524980462,
      "flos": 62881165820160.0,
      "grad_norm": 0.9420288168330911,
      "language_loss": 0.60766435,
      "learning_rate": 3.8926571654058715e-06,
      "loss": 0.62893343,
      "num_input_tokens_seen": 23273560,
      "step": 1095,
      "time_per_iteration": 3.179468870162964
    },
    {
      "auxiliary_loss_clip": 0.01182283,
      "auxiliary_loss_mlp": 0.01044188,
      "balance_loss_clip": 1.05502522,
      "balance_loss_mlp": 1.03344774,
      "epoch": 0.1317862081404437,
      "flos": 23586523793280.0,
      "grad_norm": 2.3637812483979315,
      "language_loss": 0.77033162,
      "learning_rate": 3.892405253430074e-06,
      "loss": 0.79259634,
      "num_input_tokens_seen": 23291080,
      "step": 1096,
      "time_per_iteration": 2.7917513847351074
    },
    {
      "auxiliary_loss_clip": 0.01198824,
      "auxiliary_loss_mlp": 0.00765316,
      "balance_loss_clip": 1.05416965,
      "balance_loss_mlp": 1.00035751,
      "epoch": 0.13190645103108278,
      "flos": 20260764460800.0,
      "grad_norm": 1.924255025083963,
      "language_loss": 0.82513064,
      "learning_rate": 3.892153054377904e-06,
      "loss": 0.84477204,
      "num_input_tokens_seen": 23308485,
      "step": 1097,
      "time_per_iteration": 2.7591538429260254
    },
    {
      "auxiliary_loss_clip": 0.0106837,
      "auxiliary_loss_mlp": 0.0101432,
      "balance_loss_clip": 1.02872324,
      "balance_loss_mlp": 1.01079178,
      "epoch": 0.13202669392172187,
      "flos": 53455440136320.0,
      "grad_norm": 0.9429081832227368,
      "language_loss": 0.59368134,
      "learning_rate": 3.891900568287619e-06,
      "loss": 0.61450827,
      "num_input_tokens_seen": 23360870,
      "step": 1098,
      "time_per_iteration": 3.1594460010528564
    },
    {
      "auxiliary_loss_clip": 0.0118799,
      "auxiliary_loss_mlp": 0.01040992,
      "balance_loss_clip": 1.05458307,
      "balance_loss_mlp": 1.02978635,
      "epoch": 0.13214693681236098,
      "flos": 15851293845120.0,
      "grad_norm": 2.6126593444008064,
      "language_loss": 0.72153783,
      "learning_rate": 3.891647795197523e-06,
      "loss": 0.7438277,
      "num_input_tokens_seen": 23376910,
      "step": 1099,
      "time_per_iteration": 2.7069716453552246
    },
    {
      "auxiliary_loss_clip": 0.01188433,
      "auxiliary_loss_mlp": 0.01042092,
      "balance_loss_clip": 1.05194664,
      "balance_loss_mlp": 1.03023672,
      "epoch": 0.13226717970300006,
      "flos": 19353840940800.0,
      "grad_norm": 8.599552645340156,
      "language_loss": 0.68919945,
      "learning_rate": 3.8913947351459605e-06,
      "loss": 0.7115047,
      "num_input_tokens_seen": 23394450,
      "step": 1100,
      "time_per_iteration": 2.7359695434570312
    },
    {
      "auxiliary_loss_clip": 0.01230747,
      "auxiliary_loss_mlp": 0.01035633,
      "balance_loss_clip": 1.06164312,
      "balance_loss_mlp": 1.02514899,
      "epoch": 0.13238742259363914,
      "flos": 20698084546560.0,
      "grad_norm": 1.8958803167784046,
      "language_loss": 0.6770587,
      "learning_rate": 3.89114138817132e-06,
      "loss": 0.69972253,
      "num_input_tokens_seen": 23411115,
      "step": 1101,
      "time_per_iteration": 2.6670451164245605
    },
    {
      "auxiliary_loss_clip": 0.01212073,
      "auxiliary_loss_mlp": 0.0104838,
      "balance_loss_clip": 1.05928802,
      "balance_loss_mlp": 1.03783584,
      "epoch": 0.13250766548427825,
      "flos": 21032449274880.0,
      "grad_norm": 2.1332691254620326,
      "language_loss": 0.84176129,
      "learning_rate": 3.890887754312035e-06,
      "loss": 0.86436582,
      "num_input_tokens_seen": 23429360,
      "step": 1102,
      "time_per_iteration": 2.7282679080963135
    },
    {
      "auxiliary_loss_clip": 0.0119349,
      "auxiliary_loss_mlp": 0.0103637,
      "balance_loss_clip": 1.05276346,
      "balance_loss_mlp": 1.02558792,
      "epoch": 0.13262790837491734,
      "flos": 22637871648000.0,
      "grad_norm": 2.1614699592731084,
      "language_loss": 0.87940717,
      "learning_rate": 3.890633833606581e-06,
      "loss": 0.90170586,
      "num_input_tokens_seen": 23449050,
      "step": 1103,
      "time_per_iteration": 2.740064859390259
    },
    {
      "auxiliary_loss_clip": 0.01219616,
      "auxiliary_loss_mlp": 0.01038827,
      "balance_loss_clip": 1.06710124,
      "balance_loss_mlp": 1.02788973,
      "epoch": 0.13274815126555642,
      "flos": 19683141851520.0,
      "grad_norm": 2.4230482170485046,
      "language_loss": 0.69789934,
      "learning_rate": 3.890379626093477e-06,
      "loss": 0.72048378,
      "num_input_tokens_seen": 23468800,
      "step": 1104,
      "time_per_iteration": 2.725937604904175
    },
    {
      "auxiliary_loss_clip": 0.01155155,
      "auxiliary_loss_mlp": 0.01039585,
      "balance_loss_clip": 1.05263281,
      "balance_loss_mlp": 1.02857614,
      "epoch": 0.1328683941561955,
      "flos": 21317687176320.0,
      "grad_norm": 5.961143744005907,
      "language_loss": 0.92816705,
      "learning_rate": 3.890125131811287e-06,
      "loss": 0.95011449,
      "num_input_tokens_seen": 23486850,
      "step": 1105,
      "time_per_iteration": 2.753812313079834
    },
    {
      "auxiliary_loss_clip": 0.01187978,
      "auxiliary_loss_mlp": 0.01026744,
      "balance_loss_clip": 1.05181253,
      "balance_loss_mlp": 1.01636147,
      "epoch": 0.1329886370468346,
      "flos": 13699131580800.0,
      "grad_norm": 1.9337355698622487,
      "language_loss": 0.75507557,
      "learning_rate": 3.889870350798618e-06,
      "loss": 0.77722275,
      "num_input_tokens_seen": 23504195,
      "step": 1106,
      "time_per_iteration": 2.675607919692993
    },
    {
      "auxiliary_loss_clip": 0.01229542,
      "auxiliary_loss_mlp": 0.01036862,
      "balance_loss_clip": 1.06031942,
      "balance_loss_mlp": 1.02615142,
      "epoch": 0.1331088799374737,
      "flos": 21032413361280.0,
      "grad_norm": 1.594137305097391,
      "language_loss": 0.78603721,
      "learning_rate": 3.889615283094119e-06,
      "loss": 0.80870122,
      "num_input_tokens_seen": 23523385,
      "step": 1107,
      "time_per_iteration": 2.5718166828155518
    },
    {
      "auxiliary_loss_clip": 0.01233561,
      "auxiliary_loss_mlp": 0.01044299,
      "balance_loss_clip": 1.06054628,
      "balance_loss_mlp": 1.03256917,
      "epoch": 0.13322912282811278,
      "flos": 18260432985600.0,
      "grad_norm": 2.727846927857355,
      "language_loss": 0.84934723,
      "learning_rate": 3.889359928736485e-06,
      "loss": 0.87212574,
      "num_input_tokens_seen": 23541330,
      "step": 1108,
      "time_per_iteration": 2.651667594909668
    },
    {
      "auxiliary_loss_clip": 0.01200675,
      "auxiliary_loss_mlp": 0.0076558,
      "balance_loss_clip": 1.06160474,
      "balance_loss_mlp": 1.0002625,
      "epoch": 0.1333493657187519,
      "flos": 24460876656000.0,
      "grad_norm": 2.412021822735133,
      "language_loss": 0.91338766,
      "learning_rate": 3.889104287764451e-06,
      "loss": 0.93305016,
      "num_input_tokens_seen": 23561705,
      "step": 1109,
      "time_per_iteration": 3.6379072666168213
    },
    {
      "auxiliary_loss_clip": 0.01194874,
      "auxiliary_loss_mlp": 0.01037843,
      "balance_loss_clip": 1.05716181,
      "balance_loss_mlp": 1.02661324,
      "epoch": 0.13346960860939097,
      "flos": 22158930677760.0,
      "grad_norm": 2.0191153906746653,
      "language_loss": 0.90590143,
      "learning_rate": 3.888848360216798e-06,
      "loss": 0.92822862,
      "num_input_tokens_seen": 23579350,
      "step": 1110,
      "time_per_iteration": 2.685408115386963
    },
    {
      "auxiliary_loss_clip": 0.01114857,
      "auxiliary_loss_mlp": 0.01016668,
      "balance_loss_clip": 1.0304563,
      "balance_loss_mlp": 1.01323438,
      "epoch": 0.13358985150003005,
      "flos": 67931212608000.0,
      "grad_norm": 0.8156939426202992,
      "language_loss": 0.56590152,
      "learning_rate": 3.888592146132351e-06,
      "loss": 0.58721685,
      "num_input_tokens_seen": 23640620,
      "step": 1111,
      "time_per_iteration": 3.3684709072113037
    },
    {
      "auxiliary_loss_clip": 0.01215355,
      "auxiliary_loss_mlp": 0.0103084,
      "balance_loss_clip": 1.06140423,
      "balance_loss_mlp": 1.02018881,
      "epoch": 0.13371009439066917,
      "flos": 26834284742400.0,
      "grad_norm": 2.239995209620398,
      "language_loss": 0.7872861,
      "learning_rate": 3.888335645549978e-06,
      "loss": 0.80974805,
      "num_input_tokens_seen": 23661040,
      "step": 1112,
      "time_per_iteration": 2.7217676639556885
    },
    {
      "auxiliary_loss_clip": 0.01234141,
      "auxiliary_loss_mlp": 0.01044273,
      "balance_loss_clip": 1.06363797,
      "balance_loss_mlp": 1.03347874,
      "epoch": 0.13383033728130825,
      "flos": 26322844942080.0,
      "grad_norm": 2.8639483862568844,
      "language_loss": 0.81577265,
      "learning_rate": 3.888078858508588e-06,
      "loss": 0.83855677,
      "num_input_tokens_seen": 23680900,
      "step": 1113,
      "time_per_iteration": 4.483829021453857
    },
    {
      "auxiliary_loss_clip": 0.01198444,
      "auxiliary_loss_mlp": 0.01035513,
      "balance_loss_clip": 1.06125247,
      "balance_loss_mlp": 1.02426529,
      "epoch": 0.13395058017194733,
      "flos": 22563931501440.0,
      "grad_norm": 2.1132454005450003,
      "language_loss": 0.84611714,
      "learning_rate": 3.8878217850471365e-06,
      "loss": 0.86845672,
      "num_input_tokens_seen": 23700815,
      "step": 1114,
      "time_per_iteration": 2.7248823642730713
    },
    {
      "auxiliary_loss_clip": 0.01238173,
      "auxiliary_loss_mlp": 0.01045272,
      "balance_loss_clip": 1.06580782,
      "balance_loss_mlp": 1.03391171,
      "epoch": 0.13407082306258641,
      "flos": 25810938264960.0,
      "grad_norm": 1.7238083315414132,
      "language_loss": 0.73686063,
      "learning_rate": 3.887564425204621e-06,
      "loss": 0.75969517,
      "num_input_tokens_seen": 23722500,
      "step": 1115,
      "time_per_iteration": 2.6231260299682617
    },
    {
      "auxiliary_loss_clip": 0.01093754,
      "auxiliary_loss_mlp": 0.01006334,
      "balance_loss_clip": 1.02726567,
      "balance_loss_mlp": 1.00294876,
      "epoch": 0.13419106595322552,
      "flos": 68338365269760.0,
      "grad_norm": 0.8383876770233804,
      "language_loss": 0.54665816,
      "learning_rate": 3.887306779020083e-06,
      "loss": 0.56765902,
      "num_input_tokens_seen": 23777155,
      "step": 1116,
      "time_per_iteration": 3.2276229858398438
    },
    {
      "auxiliary_loss_clip": 0.01219183,
      "auxiliary_loss_mlp": 0.01041377,
      "balance_loss_clip": 1.06195617,
      "balance_loss_mlp": 1.02999878,
      "epoch": 0.1343113088438646,
      "flos": 20449080489600.0,
      "grad_norm": 2.2046856556632224,
      "language_loss": 0.70686364,
      "learning_rate": 3.887048846532608e-06,
      "loss": 0.72946918,
      "num_input_tokens_seen": 23794130,
      "step": 1117,
      "time_per_iteration": 2.6227176189422607
    },
    {
      "auxiliary_loss_clip": 0.01094477,
      "auxiliary_loss_mlp": 0.01004296,
      "balance_loss_clip": 1.02672219,
      "balance_loss_mlp": 1.00086308,
      "epoch": 0.1344315517345037,
      "flos": 67389784951680.0,
      "grad_norm": 0.7909576807685992,
      "language_loss": 0.5811367,
      "learning_rate": 3.8867906277813224e-06,
      "loss": 0.60212445,
      "num_input_tokens_seen": 23852285,
      "step": 1118,
      "time_per_iteration": 3.170527696609497
    },
    {
      "auxiliary_loss_clip": 0.01220801,
      "auxiliary_loss_mlp": 0.00765383,
      "balance_loss_clip": 1.06221616,
      "balance_loss_mlp": 1.00063288,
      "epoch": 0.1345517946251428,
      "flos": 40734442788480.0,
      "grad_norm": 2.1802861608348363,
      "language_loss": 0.74295092,
      "learning_rate": 3.886532122805399e-06,
      "loss": 0.76281273,
      "num_input_tokens_seen": 23874765,
      "step": 1119,
      "time_per_iteration": 2.8791720867156982
    },
    {
      "auxiliary_loss_clip": 0.01144738,
      "auxiliary_loss_mlp": 0.01042451,
      "balance_loss_clip": 1.05097055,
      "balance_loss_mlp": 1.03069723,
      "epoch": 0.13467203751578188,
      "flos": 22816850140800.0,
      "grad_norm": 2.25042591410831,
      "language_loss": 0.8974508,
      "learning_rate": 3.886273331644053e-06,
      "loss": 0.91932273,
      "num_input_tokens_seen": 23893635,
      "step": 1120,
      "time_per_iteration": 2.7626118659973145
    },
    {
      "auxiliary_loss_clip": 0.01164678,
      "auxiliary_loss_mlp": 0.01033792,
      "balance_loss_clip": 1.05545461,
      "balance_loss_mlp": 1.02331352,
      "epoch": 0.13479228040642097,
      "flos": 17091576512640.0,
      "grad_norm": 1.9161704471879897,
      "language_loss": 0.82289207,
      "learning_rate": 3.886014254336542e-06,
      "loss": 0.84487683,
      "num_input_tokens_seen": 23910110,
      "step": 1121,
      "time_per_iteration": 2.8524169921875
    },
    {
      "auxiliary_loss_clip": 0.01211465,
      "auxiliary_loss_mlp": 0.01040985,
      "balance_loss_clip": 1.05755019,
      "balance_loss_mlp": 1.03028643,
      "epoch": 0.13491252329706005,
      "flos": 23730525417600.0,
      "grad_norm": 1.6925843210919918,
      "language_loss": 0.92409372,
      "learning_rate": 3.885754890922168e-06,
      "loss": 0.94661826,
      "num_input_tokens_seen": 23930440,
      "step": 1122,
      "time_per_iteration": 2.7263529300689697
    },
    {
      "auxiliary_loss_clip": 0.01130127,
      "auxiliary_loss_mlp": 0.01037053,
      "balance_loss_clip": 1.05130744,
      "balance_loss_mlp": 1.02582991,
      "epoch": 0.13503276618769916,
      "flos": 34127058960000.0,
      "grad_norm": 1.8272213429318391,
      "language_loss": 0.78646427,
      "learning_rate": 3.885495241440277e-06,
      "loss": 0.80813605,
      "num_input_tokens_seen": 23954535,
      "step": 1123,
      "time_per_iteration": 2.904585361480713
    },
    {
      "auxiliary_loss_clip": 0.01233631,
      "auxiliary_loss_mlp": 0.01036961,
      "balance_loss_clip": 1.0627377,
      "balance_loss_mlp": 1.02583325,
      "epoch": 0.13515300907833824,
      "flos": 17712328377600.0,
      "grad_norm": 2.3698409967968836,
      "language_loss": 0.74266928,
      "learning_rate": 3.885235305930257e-06,
      "loss": 0.7653752,
      "num_input_tokens_seen": 23972735,
      "step": 1124,
      "time_per_iteration": 2.5818779468536377
    },
    {
      "auxiliary_loss_clip": 0.01178054,
      "auxiliary_loss_mlp": 0.01040764,
      "balance_loss_clip": 1.05522275,
      "balance_loss_mlp": 1.03000534,
      "epoch": 0.13527325196897733,
      "flos": 20260872201600.0,
      "grad_norm": 4.456976906464268,
      "language_loss": 0.85422242,
      "learning_rate": 3.884975084431539e-06,
      "loss": 0.8764106,
      "num_input_tokens_seen": 23987685,
      "step": 1125,
      "time_per_iteration": 2.7775962352752686
    },
    {
      "auxiliary_loss_clip": 0.01205798,
      "auxiliary_loss_mlp": 0.00765478,
      "balance_loss_clip": 1.05913711,
      "balance_loss_mlp": 1.00073028,
      "epoch": 0.13539349485961644,
      "flos": 18186492839040.0,
      "grad_norm": 2.183887364763836,
      "language_loss": 0.92090023,
      "learning_rate": 3.8847145769836e-06,
      "loss": 0.94061297,
      "num_input_tokens_seen": 24004105,
      "step": 1126,
      "time_per_iteration": 2.8285653591156006
    },
    {
      "auxiliary_loss_clip": 0.01232596,
      "auxiliary_loss_mlp": 0.0103485,
      "balance_loss_clip": 1.06419539,
      "balance_loss_mlp": 1.02397203,
      "epoch": 0.13551373775025552,
      "flos": 19317463441920.0,
      "grad_norm": 2.3048519473988374,
      "language_loss": 0.66589153,
      "learning_rate": 3.884453783625959e-06,
      "loss": 0.68856609,
      "num_input_tokens_seen": 24021715,
      "step": 1127,
      "time_per_iteration": 2.595576286315918
    },
    {
      "auxiliary_loss_clip": 0.01198459,
      "auxiliary_loss_mlp": 0.01042349,
      "balance_loss_clip": 1.05889177,
      "balance_loss_mlp": 1.03148329,
      "epoch": 0.1356339806408946,
      "flos": 20850813175680.0,
      "grad_norm": 2.436872275808764,
      "language_loss": 0.84546453,
      "learning_rate": 3.884192704398176e-06,
      "loss": 0.8678726,
      "num_input_tokens_seen": 24038915,
      "step": 1128,
      "time_per_iteration": 2.698815107345581
    },
    {
      "auxiliary_loss_clip": 0.01216812,
      "auxiliary_loss_mlp": 0.01043699,
      "balance_loss_clip": 1.05916715,
      "balance_loss_mlp": 1.03323877,
      "epoch": 0.13575422353153369,
      "flos": 50476037696640.0,
      "grad_norm": 1.700540788388767,
      "language_loss": 0.74835658,
      "learning_rate": 3.883931339339858e-06,
      "loss": 0.7709617,
      "num_input_tokens_seen": 24063300,
      "step": 1129,
      "time_per_iteration": 2.8856260776519775
    },
    {
      "auxiliary_loss_clip": 0.01225187,
      "auxiliary_loss_mlp": 0.01035405,
      "balance_loss_clip": 1.06255329,
      "balance_loss_mlp": 1.02425933,
      "epoch": 0.1358744664221728,
      "flos": 18150797698560.0,
      "grad_norm": 2.2860937054266564,
      "language_loss": 0.78828573,
      "learning_rate": 3.883669688490654e-06,
      "loss": 0.81089163,
      "num_input_tokens_seen": 24081070,
      "step": 1130,
      "time_per_iteration": 2.690363883972168
    },
    {
      "auxiliary_loss_clip": 0.01190336,
      "auxiliary_loss_mlp": 0.00765934,
      "balance_loss_clip": 1.05718076,
      "balance_loss_mlp": 1.00061572,
      "epoch": 0.13599470931281188,
      "flos": 18442966924800.0,
      "grad_norm": 2.686854094856528,
      "language_loss": 0.85423708,
      "learning_rate": 3.883407751890256e-06,
      "loss": 0.8737998,
      "num_input_tokens_seen": 24099675,
      "step": 1131,
      "time_per_iteration": 2.7279837131500244
    },
    {
      "auxiliary_loss_clip": 0.01180831,
      "auxiliary_loss_mlp": 0.01036642,
      "balance_loss_clip": 1.05225122,
      "balance_loss_mlp": 1.02554321,
      "epoch": 0.13611495220345096,
      "flos": 26680766014080.0,
      "grad_norm": 2.087136934452564,
      "language_loss": 0.85873151,
      "learning_rate": 3.8831455295783994e-06,
      "loss": 0.88090622,
      "num_input_tokens_seen": 24118925,
      "step": 1132,
      "time_per_iteration": 2.7800850868225098
    },
    {
      "auxiliary_loss_clip": 0.0119255,
      "auxiliary_loss_mlp": 0.01032127,
      "balance_loss_clip": 1.05634964,
      "balance_loss_mlp": 1.02083218,
      "epoch": 0.13623519509409007,
      "flos": 21686238673920.0,
      "grad_norm": 1.7862779758109626,
      "language_loss": 0.74030447,
      "learning_rate": 3.882883021594864e-06,
      "loss": 0.76255125,
      "num_input_tokens_seen": 24137065,
      "step": 1133,
      "time_per_iteration": 2.650254011154175
    },
    {
      "auxiliary_loss_clip": 0.01178653,
      "auxiliary_loss_mlp": 0.01045186,
      "balance_loss_clip": 1.05707145,
      "balance_loss_mlp": 1.0343914,
      "epoch": 0.13635543798472916,
      "flos": 14830389492480.0,
      "grad_norm": 2.1664577873178548,
      "language_loss": 0.8674885,
      "learning_rate": 3.8826202279794705e-06,
      "loss": 0.88972694,
      "num_input_tokens_seen": 24154125,
      "step": 1134,
      "time_per_iteration": 2.7394473552703857
    },
    {
      "auxiliary_loss_clip": 0.01236867,
      "auxiliary_loss_mlp": 0.01039344,
      "balance_loss_clip": 1.06683183,
      "balance_loss_mlp": 1.02805483,
      "epoch": 0.13647568087536824,
      "flos": 22890323410560.0,
      "grad_norm": 1.9451314843504632,
      "language_loss": 0.70131779,
      "learning_rate": 3.882357148772085e-06,
      "loss": 0.72407985,
      "num_input_tokens_seen": 24171550,
      "step": 1135,
      "time_per_iteration": 3.5486795902252197
    },
    {
      "auxiliary_loss_clip": 0.01175206,
      "auxiliary_loss_mlp": 0.0104224,
      "balance_loss_clip": 1.05700278,
      "balance_loss_mlp": 1.03152943,
      "epoch": 0.13659592376600732,
      "flos": 19937927998080.0,
      "grad_norm": 4.231213380106844,
      "language_loss": 0.84945726,
      "learning_rate": 3.882093784012617e-06,
      "loss": 0.87163174,
      "num_input_tokens_seen": 24190190,
      "step": 1136,
      "time_per_iteration": 2.725208044052124
    },
    {
      "auxiliary_loss_clip": 0.01199978,
      "auxiliary_loss_mlp": 0.01049518,
      "balance_loss_clip": 1.05827248,
      "balance_loss_mlp": 1.03803182,
      "epoch": 0.13671616665664643,
      "flos": 21428579439360.0,
      "grad_norm": 1.8720530265014979,
      "language_loss": 0.8411597,
      "learning_rate": 3.881830133741019e-06,
      "loss": 0.86365467,
      "num_input_tokens_seen": 24209055,
      "step": 1137,
      "time_per_iteration": 2.752316951751709
    },
    {
      "auxiliary_loss_clip": 0.01191245,
      "auxiliary_loss_mlp": 0.01043746,
      "balance_loss_clip": 1.06197071,
      "balance_loss_mlp": 1.03258181,
      "epoch": 0.13683640954728551,
      "flos": 22778138257920.0,
      "grad_norm": 2.0765504247654656,
      "language_loss": 0.76102978,
      "learning_rate": 3.881566197997285e-06,
      "loss": 0.78337967,
      "num_input_tokens_seen": 24225490,
      "step": 1138,
      "time_per_iteration": 2.6883456707000732
    },
    {
      "auxiliary_loss_clip": 0.0119772,
      "auxiliary_loss_mlp": 0.01039232,
      "balance_loss_clip": 1.06064487,
      "balance_loss_mlp": 1.02933741,
      "epoch": 0.1369566524379246,
      "flos": 21725884310400.0,
      "grad_norm": 1.5279681114569097,
      "language_loss": 0.74543828,
      "learning_rate": 3.881301976821456e-06,
      "loss": 0.76780784,
      "num_input_tokens_seen": 24245520,
      "step": 1139,
      "time_per_iteration": 4.50075101852417
    },
    {
      "auxiliary_loss_clip": 0.01214294,
      "auxiliary_loss_mlp": 0.01036008,
      "balance_loss_clip": 1.06149733,
      "balance_loss_mlp": 1.02482069,
      "epoch": 0.1370768953285637,
      "flos": 18624459369600.0,
      "grad_norm": 1.8182741834650884,
      "language_loss": 0.90331072,
      "learning_rate": 3.881037470253612e-06,
      "loss": 0.92581379,
      "num_input_tokens_seen": 24265035,
      "step": 1140,
      "time_per_iteration": 2.67482590675354
    },
    {
      "auxiliary_loss_clip": 0.01168962,
      "auxiliary_loss_mlp": 0.01041079,
      "balance_loss_clip": 1.05458057,
      "balance_loss_mlp": 1.03022528,
      "epoch": 0.1371971382192028,
      "flos": 14939521989120.0,
      "grad_norm": 2.727979033283011,
      "language_loss": 0.79404998,
      "learning_rate": 3.88077267833388e-06,
      "loss": 0.81615037,
      "num_input_tokens_seen": 24281550,
      "step": 1141,
      "time_per_iteration": 2.7099382877349854
    },
    {
      "auxiliary_loss_clip": 0.01164416,
      "auxiliary_loss_mlp": 0.01040793,
      "balance_loss_clip": 1.05650043,
      "balance_loss_mlp": 1.02991509,
      "epoch": 0.13731738110984187,
      "flos": 19023785844480.0,
      "grad_norm": 2.0539160383646857,
      "language_loss": 0.83984089,
      "learning_rate": 3.880507601102427e-06,
      "loss": 0.861893,
      "num_input_tokens_seen": 24299485,
      "step": 1142,
      "time_per_iteration": 2.76448917388916
    },
    {
      "auxiliary_loss_clip": 0.01233585,
      "auxiliary_loss_mlp": 0.01049758,
      "balance_loss_clip": 1.06611609,
      "balance_loss_mlp": 1.03861237,
      "epoch": 0.13743762400048098,
      "flos": 18187462506240.0,
      "grad_norm": 2.0171180441402456,
      "language_loss": 0.82084405,
      "learning_rate": 3.880242238599467e-06,
      "loss": 0.84367746,
      "num_input_tokens_seen": 24316010,
      "step": 1143,
      "time_per_iteration": 2.577920436859131
    },
    {
      "auxiliary_loss_clip": 0.01229473,
      "auxiliary_loss_mlp": 0.01039626,
      "balance_loss_clip": 1.0626955,
      "balance_loss_mlp": 1.0285337,
      "epoch": 0.13755786689112007,
      "flos": 21031982398080.0,
      "grad_norm": 2.4309083029303786,
      "language_loss": 0.8335923,
      "learning_rate": 3.879976590865254e-06,
      "loss": 0.85628331,
      "num_input_tokens_seen": 24335465,
      "step": 1144,
      "time_per_iteration": 2.6204025745391846
    },
    {
      "auxiliary_loss_clip": 0.01200242,
      "auxiliary_loss_mlp": 0.01038145,
      "balance_loss_clip": 1.06092632,
      "balance_loss_mlp": 1.02701688,
      "epoch": 0.13767810978175915,
      "flos": 21360636864000.0,
      "grad_norm": 2.6693306466522433,
      "language_loss": 0.87186241,
      "learning_rate": 3.879710657940087e-06,
      "loss": 0.89424634,
      "num_input_tokens_seen": 24354415,
      "step": 1145,
      "time_per_iteration": 2.848679780960083
    },
    {
      "auxiliary_loss_clip": 0.01218746,
      "auxiliary_loss_mlp": 0.01041509,
      "balance_loss_clip": 1.0635097,
      "balance_loss_mlp": 1.03043461,
      "epoch": 0.13779835267239823,
      "flos": 30592084861440.0,
      "grad_norm": 1.7909629185773799,
      "language_loss": 0.70142758,
      "learning_rate": 3.879444439864308e-06,
      "loss": 0.72403014,
      "num_input_tokens_seen": 24373990,
      "step": 1146,
      "time_per_iteration": 2.7371418476104736
    },
    {
      "auxiliary_loss_clip": 0.01213803,
      "auxiliary_loss_mlp": 0.00765845,
      "balance_loss_clip": 1.05822802,
      "balance_loss_mlp": 1.00067079,
      "epoch": 0.13791859556303734,
      "flos": 22669867687680.0,
      "grad_norm": 2.352702958173118,
      "language_loss": 0.85997128,
      "learning_rate": 3.879177936678301e-06,
      "loss": 0.87976778,
      "num_input_tokens_seen": 24392995,
      "step": 1147,
      "time_per_iteration": 2.6676275730133057
    },
    {
      "auxiliary_loss_clip": 0.01208088,
      "auxiliary_loss_mlp": 0.01047576,
      "balance_loss_clip": 1.05927896,
      "balance_loss_mlp": 1.03579831,
      "epoch": 0.13803883845367643,
      "flos": 35224166016000.0,
      "grad_norm": 1.9074180614889205,
      "language_loss": 0.77291101,
      "learning_rate": 3.878911148422496e-06,
      "loss": 0.79546762,
      "num_input_tokens_seen": 24414470,
      "step": 1148,
      "time_per_iteration": 2.8286795616149902
    },
    {
      "auxiliary_loss_clip": 0.01218961,
      "auxiliary_loss_mlp": 0.01039196,
      "balance_loss_clip": 1.06102681,
      "balance_loss_mlp": 1.02759075,
      "epoch": 0.1381590813443155,
      "flos": 32014542332160.0,
      "grad_norm": 2.3257573166904,
      "language_loss": 0.70484996,
      "learning_rate": 3.878644075137364e-06,
      "loss": 0.72743154,
      "num_input_tokens_seen": 24435120,
      "step": 1149,
      "time_per_iteration": 2.8019440174102783
    },
    {
      "auxiliary_loss_clip": 0.01162127,
      "auxiliary_loss_mlp": 0.01035836,
      "balance_loss_clip": 1.05043757,
      "balance_loss_mlp": 1.02475595,
      "epoch": 0.13827932423495462,
      "flos": 17821855923840.0,
      "grad_norm": 2.3140551464775303,
      "language_loss": 0.79149461,
      "learning_rate": 3.878376716863418e-06,
      "loss": 0.81347418,
      "num_input_tokens_seen": 24451420,
      "step": 1150,
      "time_per_iteration": 2.6529364585876465
    },
    {
      "auxiliary_loss_clip": 0.01196322,
      "auxiliary_loss_mlp": 0.01042837,
      "balance_loss_clip": 1.05552399,
      "balance_loss_mlp": 1.03212023,
      "epoch": 0.1383995671255937,
      "flos": 19427098728960.0,
      "grad_norm": 2.8178864798301815,
      "language_loss": 0.71838844,
      "learning_rate": 3.878109073641219e-06,
      "loss": 0.74078,
      "num_input_tokens_seen": 24470450,
      "step": 1151,
      "time_per_iteration": 2.709325075149536
    },
    {
      "auxiliary_loss_clip": 0.01167708,
      "auxiliary_loss_mlp": 0.01038636,
      "balance_loss_clip": 1.05609381,
      "balance_loss_mlp": 1.02786517,
      "epoch": 0.13851981001623279,
      "flos": 28296603331200.0,
      "grad_norm": 1.6107904379459834,
      "language_loss": 0.81167781,
      "learning_rate": 3.877841145511366e-06,
      "loss": 0.83374119,
      "num_input_tokens_seen": 24493190,
      "step": 1152,
      "time_per_iteration": 2.815868377685547
    },
    {
      "auxiliary_loss_clip": 0.01222692,
      "auxiliary_loss_mlp": 0.01045542,
      "balance_loss_clip": 1.06283748,
      "balance_loss_mlp": 1.03430092,
      "epoch": 0.13864005290687187,
      "flos": 21213079793280.0,
      "grad_norm": 1.7222197307130964,
      "language_loss": 0.82926863,
      "learning_rate": 3.8775729325145035e-06,
      "loss": 0.851951,
      "num_input_tokens_seen": 24512425,
      "step": 1153,
      "time_per_iteration": 2.681976795196533
    },
    {
      "auxiliary_loss_clip": 0.01080075,
      "auxiliary_loss_mlp": 0.01006332,
      "balance_loss_clip": 1.02467167,
      "balance_loss_mlp": 1.00299442,
      "epoch": 0.13876029579751098,
      "flos": 71653389413760.0,
      "grad_norm": 0.7937130332311666,
      "language_loss": 0.64775288,
      "learning_rate": 3.877304434691321e-06,
      "loss": 0.66861701,
      "num_input_tokens_seen": 24579275,
      "step": 1154,
      "time_per_iteration": 3.415860891342163
    },
    {
      "auxiliary_loss_clip": 0.01186598,
      "auxiliary_loss_mlp": 0.01036052,
      "balance_loss_clip": 1.06064188,
      "balance_loss_mlp": 1.02568126,
      "epoch": 0.13888053868815006,
      "flos": 21941348042880.0,
      "grad_norm": 1.8181160544596704,
      "language_loss": 0.79606926,
      "learning_rate": 3.877035652082548e-06,
      "loss": 0.81829584,
      "num_input_tokens_seen": 24598720,
      "step": 1155,
      "time_per_iteration": 2.730027675628662
    },
    {
      "auxiliary_loss_clip": 0.01195761,
      "auxiliary_loss_mlp": 0.01040635,
      "balance_loss_clip": 1.06201911,
      "balance_loss_mlp": 1.02876735,
      "epoch": 0.13900078157878915,
      "flos": 19608627087360.0,
      "grad_norm": 1.7243531977411166,
      "language_loss": 0.85330909,
      "learning_rate": 3.87676658472896e-06,
      "loss": 0.87567306,
      "num_input_tokens_seen": 24617530,
      "step": 1156,
      "time_per_iteration": 2.717961311340332
    },
    {
      "auxiliary_loss_clip": 0.01211747,
      "auxiliary_loss_mlp": 0.01040129,
      "balance_loss_clip": 1.05571079,
      "balance_loss_mlp": 1.02926302,
      "epoch": 0.13912102446942826,
      "flos": 22638051216000.0,
      "grad_norm": 2.3832685797276123,
      "language_loss": 0.85097796,
      "learning_rate": 3.876497232671372e-06,
      "loss": 0.87349671,
      "num_input_tokens_seen": 24637485,
      "step": 1157,
      "time_per_iteration": 2.6454429626464844
    },
    {
      "auxiliary_loss_clip": 0.01170328,
      "auxiliary_loss_mlp": 0.0103933,
      "balance_loss_clip": 1.05421853,
      "balance_loss_mlp": 1.02819645,
      "epoch": 0.13924126736006734,
      "flos": 29643324975360.0,
      "grad_norm": 2.1507306919810114,
      "language_loss": 0.83803481,
      "learning_rate": 3.876227595950647e-06,
      "loss": 0.86013138,
      "num_input_tokens_seen": 24656915,
      "step": 1158,
      "time_per_iteration": 2.7870540618896484
    },
    {
      "auxiliary_loss_clip": 0.01230742,
      "auxiliary_loss_mlp": 0.01035152,
      "balance_loss_clip": 1.06337821,
      "balance_loss_mlp": 1.02357054,
      "epoch": 0.13936151025070642,
      "flos": 27417653527680.0,
      "grad_norm": 1.5555805836656236,
      "language_loss": 0.79169214,
      "learning_rate": 3.875957674607686e-06,
      "loss": 0.81435108,
      "num_input_tokens_seen": 24679190,
      "step": 1159,
      "time_per_iteration": 2.68554949760437
    },
    {
      "auxiliary_loss_clip": 0.01204482,
      "auxiliary_loss_mlp": 0.0076674,
      "balance_loss_clip": 1.05572021,
      "balance_loss_mlp": 1.00058579,
      "epoch": 0.1394817531413455,
      "flos": 16399326625920.0,
      "grad_norm": 1.7379403025435036,
      "language_loss": 0.87995005,
      "learning_rate": 3.8756874686834386e-06,
      "loss": 0.89966232,
      "num_input_tokens_seen": 24697405,
      "step": 1160,
      "time_per_iteration": 3.6171417236328125
    },
    {
      "auxiliary_loss_clip": 0.01220245,
      "auxiliary_loss_mlp": 0.00766102,
      "balance_loss_clip": 1.06036556,
      "balance_loss_mlp": 1.00062156,
      "epoch": 0.13960199603198462,
      "flos": 30922319525760.0,
      "grad_norm": 1.6647464933110165,
      "language_loss": 0.8043012,
      "learning_rate": 3.875416978218893e-06,
      "loss": 0.82416463,
      "num_input_tokens_seen": 24720600,
      "step": 1161,
      "time_per_iteration": 2.7196340560913086
    },
    {
      "auxiliary_loss_clip": 0.01193903,
      "auxiliary_loss_mlp": 0.01034022,
      "balance_loss_clip": 1.0545342,
      "balance_loss_mlp": 1.02288806,
      "epoch": 0.1397222389226237,
      "flos": 18113773754880.0,
      "grad_norm": 2.280341845333764,
      "language_loss": 0.82957339,
      "learning_rate": 3.8751462032550835e-06,
      "loss": 0.85185266,
      "num_input_tokens_seen": 24737605,
      "step": 1162,
      "time_per_iteration": 2.7208597660064697
    },
    {
      "auxiliary_loss_clip": 0.01199853,
      "auxiliary_loss_mlp": 0.01036763,
      "balance_loss_clip": 1.06250226,
      "balance_loss_mlp": 1.02624309,
      "epoch": 0.13984248181326278,
      "flos": 16872772815360.0,
      "grad_norm": 3.4816704094840296,
      "language_loss": 0.82980108,
      "learning_rate": 3.874875143833085e-06,
      "loss": 0.85216725,
      "num_input_tokens_seen": 24755845,
      "step": 1163,
      "time_per_iteration": 2.719616174697876
    },
    {
      "auxiliary_loss_clip": 0.01216468,
      "auxiliary_loss_mlp": 0.01037641,
      "balance_loss_clip": 1.06163049,
      "balance_loss_mlp": 1.02632189,
      "epoch": 0.1399627247039019,
      "flos": 54121401267840.0,
      "grad_norm": 3.286100487865083,
      "language_loss": 0.68820703,
      "learning_rate": 3.874603799994019e-06,
      "loss": 0.71074814,
      "num_input_tokens_seen": 24779380,
      "step": 1164,
      "time_per_iteration": 2.927874803543091
    },
    {
      "auxiliary_loss_clip": 0.01176223,
      "auxiliary_loss_mlp": 0.01045843,
      "balance_loss_clip": 1.05394125,
      "balance_loss_mlp": 1.03487015,
      "epoch": 0.14008296759454097,
      "flos": 11765521618560.0,
      "grad_norm": 2.9795500256942535,
      "language_loss": 0.87145269,
      "learning_rate": 3.874332171779046e-06,
      "loss": 0.8936733,
      "num_input_tokens_seen": 24794260,
      "step": 1165,
      "time_per_iteration": 4.516380071640015
    },
    {
      "auxiliary_loss_clip": 0.01183177,
      "auxiliary_loss_mlp": 0.01038037,
      "balance_loss_clip": 1.05745125,
      "balance_loss_mlp": 1.02730799,
      "epoch": 0.14020321048518006,
      "flos": 22017514832640.0,
      "grad_norm": 1.6906577585042804,
      "language_loss": 0.75717759,
      "learning_rate": 3.874060259229373e-06,
      "loss": 0.77938974,
      "num_input_tokens_seen": 24815835,
      "step": 1166,
      "time_per_iteration": 2.8095974922180176
    },
    {
      "auxiliary_loss_clip": 0.01218601,
      "auxiliary_loss_mlp": 0.01047861,
      "balance_loss_clip": 1.06300867,
      "balance_loss_mlp": 1.03625,
      "epoch": 0.14032345337581917,
      "flos": 23404313076480.0,
      "grad_norm": 2.197214655886737,
      "language_loss": 0.93849444,
      "learning_rate": 3.873788062386249e-06,
      "loss": 0.96115911,
      "num_input_tokens_seen": 24834095,
      "step": 1167,
      "time_per_iteration": 2.707846164703369
    },
    {
      "auxiliary_loss_clip": 0.01189982,
      "auxiliary_loss_mlp": 0.01041922,
      "balance_loss_clip": 1.0601542,
      "balance_loss_mlp": 1.03032839,
      "epoch": 0.14044369626645825,
      "flos": 29645767100160.0,
      "grad_norm": 2.115652955981573,
      "language_loss": 0.82456028,
      "learning_rate": 3.873515581290965e-06,
      "loss": 0.84687924,
      "num_input_tokens_seen": 24858900,
      "step": 1168,
      "time_per_iteration": 2.7815768718719482
    },
    {
      "auxiliary_loss_clip": 0.01183767,
      "auxiliary_loss_mlp": 0.01033851,
      "balance_loss_clip": 1.05891788,
      "balance_loss_mlp": 1.02299142,
      "epoch": 0.14056393915709733,
      "flos": 18332972501760.0,
      "grad_norm": 2.4220738244679985,
      "language_loss": 0.75516158,
      "learning_rate": 3.8732428159848575e-06,
      "loss": 0.77733773,
      "num_input_tokens_seen": 24877875,
      "step": 1169,
      "time_per_iteration": 2.7488629817962646
    },
    {
      "auxiliary_loss_clip": 0.01217163,
      "auxiliary_loss_mlp": 0.01042,
      "balance_loss_clip": 1.06533909,
      "balance_loss_mlp": 1.03046656,
      "epoch": 0.14068418204773642,
      "flos": 26687517770880.0,
      "grad_norm": 1.9219501981153029,
      "language_loss": 0.78350127,
      "learning_rate": 3.872969766509304e-06,
      "loss": 0.80609292,
      "num_input_tokens_seen": 24898430,
      "step": 1170,
      "time_per_iteration": 2.770106077194214
    },
    {
      "auxiliary_loss_clip": 0.01082778,
      "auxiliary_loss_mlp": 0.01010405,
      "balance_loss_clip": 1.02597535,
      "balance_loss_mlp": 1.00704372,
      "epoch": 0.14080442493837553,
      "flos": 65259314501760.0,
      "grad_norm": 0.7687104911707842,
      "language_loss": 0.55604547,
      "learning_rate": 3.872696432905726e-06,
      "loss": 0.57697731,
      "num_input_tokens_seen": 24959250,
      "step": 1171,
      "time_per_iteration": 3.2625792026519775
    },
    {
      "auxiliary_loss_clip": 0.01217254,
      "auxiliary_loss_mlp": 0.01037148,
      "balance_loss_clip": 1.05836725,
      "balance_loss_mlp": 1.02498317,
      "epoch": 0.1409246678290146,
      "flos": 25776715582080.0,
      "grad_norm": 2.784654170309308,
      "language_loss": 0.71696085,
      "learning_rate": 3.872422815215589e-06,
      "loss": 0.73950487,
      "num_input_tokens_seen": 24978330,
      "step": 1172,
      "time_per_iteration": 2.719391345977783
    },
    {
      "auxiliary_loss_clip": 0.012069,
      "auxiliary_loss_mlp": 0.01041764,
      "balance_loss_clip": 1.05529308,
      "balance_loss_mlp": 1.03029013,
      "epoch": 0.1410449107196537,
      "flos": 21868521217920.0,
      "grad_norm": 1.7473169302375895,
      "language_loss": 0.74482375,
      "learning_rate": 3.8721489134803994e-06,
      "loss": 0.76731038,
      "num_input_tokens_seen": 24997120,
      "step": 1173,
      "time_per_iteration": 2.660723924636841
    },
    {
      "auxiliary_loss_clip": 0.01215377,
      "auxiliary_loss_mlp": 0.01030005,
      "balance_loss_clip": 1.06171513,
      "balance_loss_mlp": 1.01867449,
      "epoch": 0.1411651536102928,
      "flos": 16684133564160.0,
      "grad_norm": 2.301418495531914,
      "language_loss": 0.72769761,
      "learning_rate": 3.871874727741707e-06,
      "loss": 0.75015146,
      "num_input_tokens_seen": 25014350,
      "step": 1174,
      "time_per_iteration": 2.620027780532837
    },
    {
      "auxiliary_loss_clip": 0.01209984,
      "auxiliary_loss_mlp": 0.01040879,
      "balance_loss_clip": 1.06016052,
      "balance_loss_mlp": 1.03018546,
      "epoch": 0.1412853965009319,
      "flos": 20992264934400.0,
      "grad_norm": 1.965947201254203,
      "language_loss": 0.96875179,
      "learning_rate": 3.871600258041108e-06,
      "loss": 0.99126047,
      "num_input_tokens_seen": 25033875,
      "step": 1175,
      "time_per_iteration": 2.658416986465454
    },
    {
      "auxiliary_loss_clip": 0.01190663,
      "auxiliary_loss_mlp": 0.01032429,
      "balance_loss_clip": 1.05542505,
      "balance_loss_mlp": 1.02121711,
      "epoch": 0.14140563939157097,
      "flos": 20335279224960.0,
      "grad_norm": 2.6292616194805283,
      "language_loss": 0.8574124,
      "learning_rate": 3.871325504420238e-06,
      "loss": 0.87964332,
      "num_input_tokens_seen": 25052865,
      "step": 1176,
      "time_per_iteration": 2.683713436126709
    },
    {
      "auxiliary_loss_clip": 0.01228045,
      "auxiliary_loss_mlp": 0.010313,
      "balance_loss_clip": 1.06136012,
      "balance_loss_mlp": 1.02043962,
      "epoch": 0.14152588228221005,
      "flos": 21068826773760.0,
      "grad_norm": 2.195623364277583,
      "language_loss": 0.81630832,
      "learning_rate": 3.871050466920776e-06,
      "loss": 0.83890182,
      "num_input_tokens_seen": 25072770,
      "step": 1177,
      "time_per_iteration": 2.615522623062134
    },
    {
      "auxiliary_loss_clip": 0.011715,
      "auxiliary_loss_mlp": 0.01039215,
      "balance_loss_clip": 1.05102813,
      "balance_loss_mlp": 1.02843833,
      "epoch": 0.14164612517284916,
      "flos": 18223157646720.0,
      "grad_norm": 2.139111381767639,
      "language_loss": 0.79918897,
      "learning_rate": 3.870775145584447e-06,
      "loss": 0.82129616,
      "num_input_tokens_seen": 25090550,
      "step": 1178,
      "time_per_iteration": 2.6630585193634033
    },
    {
      "auxiliary_loss_clip": 0.01203337,
      "auxiliary_loss_mlp": 0.01041069,
      "balance_loss_clip": 1.0582974,
      "balance_loss_mlp": 1.02753234,
      "epoch": 0.14176636806348825,
      "flos": 22744454279040.0,
      "grad_norm": 2.6819696655023972,
      "language_loss": 0.64154553,
      "learning_rate": 3.8704995404530145e-06,
      "loss": 0.66398954,
      "num_input_tokens_seen": 25106175,
      "step": 1179,
      "time_per_iteration": 2.688589334487915
    },
    {
      "auxiliary_loss_clip": 0.01226157,
      "auxiliary_loss_mlp": 0.01051918,
      "balance_loss_clip": 1.06128585,
      "balance_loss_mlp": 1.04065275,
      "epoch": 0.14188661095412733,
      "flos": 22091095843200.0,
      "grad_norm": 2.1022701180732386,
      "language_loss": 0.85243762,
      "learning_rate": 3.87022365156829e-06,
      "loss": 0.87521839,
      "num_input_tokens_seen": 25126890,
      "step": 1180,
      "time_per_iteration": 2.6162021160125732
    },
    {
      "auxiliary_loss_clip": 0.01137869,
      "auxiliary_loss_mlp": 0.01034291,
      "balance_loss_clip": 1.050282,
      "balance_loss_mlp": 1.02328217,
      "epoch": 0.14200685384476644,
      "flos": 24352390604160.0,
      "grad_norm": 2.1447473297802455,
      "language_loss": 0.81489068,
      "learning_rate": 3.869947478972123e-06,
      "loss": 0.83661234,
      "num_input_tokens_seen": 25147915,
      "step": 1181,
      "time_per_iteration": 2.9183812141418457
    },
    {
      "auxiliary_loss_clip": 0.01203629,
      "auxiliary_loss_mlp": 0.01038404,
      "balance_loss_clip": 1.05627489,
      "balance_loss_mlp": 1.02728188,
      "epoch": 0.14212709673540552,
      "flos": 24022048199040.0,
      "grad_norm": 2.3093325161314473,
      "language_loss": 0.82918704,
      "learning_rate": 3.869671022706412e-06,
      "loss": 0.85160738,
      "num_input_tokens_seen": 25166645,
      "step": 1182,
      "time_per_iteration": 2.7755186557769775
    },
    {
      "auxiliary_loss_clip": 0.01150239,
      "auxiliary_loss_mlp": 0.01040033,
      "balance_loss_clip": 1.05009019,
      "balance_loss_mlp": 1.02799892,
      "epoch": 0.1422473396260446,
      "flos": 26431797870720.0,
      "grad_norm": 2.2288387068433466,
      "language_loss": 0.64955342,
      "learning_rate": 3.869394282813092e-06,
      "loss": 0.6714561,
      "num_input_tokens_seen": 25185845,
      "step": 1183,
      "time_per_iteration": 2.79331111907959
    },
    {
      "auxiliary_loss_clip": 0.0119128,
      "auxiliary_loss_mlp": 0.01035796,
      "balance_loss_clip": 1.05605412,
      "balance_loss_mlp": 1.02455449,
      "epoch": 0.1423675825166837,
      "flos": 17055306754560.0,
      "grad_norm": 2.4592609011729376,
      "language_loss": 0.89634633,
      "learning_rate": 3.869117259334147e-06,
      "loss": 0.91861713,
      "num_input_tokens_seen": 25203770,
      "step": 1184,
      "time_per_iteration": 2.6829237937927246
    },
    {
      "auxiliary_loss_clip": 0.01206939,
      "auxiliary_loss_mlp": 0.01036988,
      "balance_loss_clip": 1.05743837,
      "balance_loss_mlp": 1.0265336,
      "epoch": 0.1424878254073228,
      "flos": 17929480049280.0,
      "grad_norm": 1.7359719081151015,
      "language_loss": 0.82280344,
      "learning_rate": 3.868839952311599e-06,
      "loss": 0.84524274,
      "num_input_tokens_seen": 25221725,
      "step": 1185,
      "time_per_iteration": 2.662015438079834
    },
    {
      "auxiliary_loss_clip": 0.01196969,
      "auxiliary_loss_mlp": 0.01039444,
      "balance_loss_clip": 1.06051517,
      "balance_loss_mlp": 1.02850044,
      "epoch": 0.14260806829796188,
      "flos": 20303606407680.0,
      "grad_norm": 2.797085976501659,
      "language_loss": 0.80312598,
      "learning_rate": 3.868562361787516e-06,
      "loss": 0.82549012,
      "num_input_tokens_seen": 25240855,
      "step": 1186,
      "time_per_iteration": 2.7269883155822754
    },
    {
      "auxiliary_loss_clip": 0.01131864,
      "auxiliary_loss_mlp": 0.01037084,
      "balance_loss_clip": 1.04704022,
      "balance_loss_mlp": 1.02575374,
      "epoch": 0.14272831118860096,
      "flos": 23185724860800.0,
      "grad_norm": 2.105933509796222,
      "language_loss": 0.68887997,
      "learning_rate": 3.868284487804009e-06,
      "loss": 0.7105695,
      "num_input_tokens_seen": 25260085,
      "step": 1187,
      "time_per_iteration": 3.7207744121551514
    },
    {
      "auxiliary_loss_clip": 0.01198911,
      "auxiliary_loss_mlp": 0.01040894,
      "balance_loss_clip": 1.05511713,
      "balance_loss_mlp": 1.02993321,
      "epoch": 0.14284855407924008,
      "flos": 27232210586880.0,
      "grad_norm": 1.6211412083852588,
      "language_loss": 0.78175557,
      "learning_rate": 3.86800633040323e-06,
      "loss": 0.80415356,
      "num_input_tokens_seen": 25280675,
      "step": 1188,
      "time_per_iteration": 2.7550668716430664
    },
    {
      "auxiliary_loss_clip": 0.01199021,
      "auxiliary_loss_mlp": 0.00765738,
      "balance_loss_clip": 1.06147051,
      "balance_loss_mlp": 1.00074446,
      "epoch": 0.14296879696987916,
      "flos": 28184202696960.0,
      "grad_norm": 2.2045211015118773,
      "language_loss": 0.78446126,
      "learning_rate": 3.867727889627376e-06,
      "loss": 0.80410886,
      "num_input_tokens_seen": 25300290,
      "step": 1189,
      "time_per_iteration": 2.740311861038208
    },
    {
      "auxiliary_loss_clip": 0.01171647,
      "auxiliary_loss_mlp": 0.01038412,
      "balance_loss_clip": 1.05613017,
      "balance_loss_mlp": 1.02647328,
      "epoch": 0.14308903986051824,
      "flos": 19390290266880.0,
      "grad_norm": 2.693677260777442,
      "language_loss": 0.78339458,
      "learning_rate": 3.867449165518687e-06,
      "loss": 0.80549526,
      "num_input_tokens_seen": 25316760,
      "step": 1190,
      "time_per_iteration": 2.7284646034240723
    },
    {
      "auxiliary_loss_clip": 0.01232593,
      "auxiliary_loss_mlp": 0.00765744,
      "balance_loss_clip": 1.06187487,
      "balance_loss_mlp": 1.00061345,
      "epoch": 0.14320928275115732,
      "flos": 17457506317440.0,
      "grad_norm": 1.7217594128849356,
      "language_loss": 0.71078998,
      "learning_rate": 3.867170158119444e-06,
      "loss": 0.73077333,
      "num_input_tokens_seen": 25335760,
      "step": 1191,
      "time_per_iteration": 4.493774890899658
    },
    {
      "auxiliary_loss_clip": 0.01231627,
      "auxiliary_loss_mlp": 0.01043424,
      "balance_loss_clip": 1.06233311,
      "balance_loss_mlp": 1.03211069,
      "epoch": 0.14332952564179643,
      "flos": 21466070259840.0,
      "grad_norm": 4.853829715682026,
      "language_loss": 0.75158399,
      "learning_rate": 3.866890867471972e-06,
      "loss": 0.77433449,
      "num_input_tokens_seen": 25354230,
      "step": 1192,
      "time_per_iteration": 2.6253933906555176
    },
    {
      "auxiliary_loss_clip": 0.01195578,
      "auxiliary_loss_mlp": 0.01033362,
      "balance_loss_clip": 1.05474019,
      "balance_loss_mlp": 1.02222824,
      "epoch": 0.14344976853243552,
      "flos": 16396992241920.0,
      "grad_norm": 2.5418725894431273,
      "language_loss": 0.89557326,
      "learning_rate": 3.86661129361864e-06,
      "loss": 0.91786265,
      "num_input_tokens_seen": 25368720,
      "step": 1193,
      "time_per_iteration": 2.7305965423583984
    },
    {
      "auxiliary_loss_clip": 0.01197874,
      "auxiliary_loss_mlp": 0.01035207,
      "balance_loss_clip": 1.06050396,
      "balance_loss_mlp": 1.0240016,
      "epoch": 0.1435700114230746,
      "flos": 18916736336640.0,
      "grad_norm": 1.8605033795656913,
      "language_loss": 0.86038327,
      "learning_rate": 3.866331436601859e-06,
      "loss": 0.88271403,
      "num_input_tokens_seen": 25386715,
      "step": 1194,
      "time_per_iteration": 2.6115710735321045
    },
    {
      "auxiliary_loss_clip": 0.01231446,
      "auxiliary_loss_mlp": 0.01037489,
      "balance_loss_clip": 1.06520653,
      "balance_loss_mlp": 1.02621222,
      "epoch": 0.1436902543137137,
      "flos": 19755394058880.0,
      "grad_norm": 3.7635740208777135,
      "language_loss": 0.73784709,
      "learning_rate": 3.866051296464083e-06,
      "loss": 0.76053643,
      "num_input_tokens_seen": 25405550,
      "step": 1195,
      "time_per_iteration": 2.6558682918548584
    },
    {
      "auxiliary_loss_clip": 0.012311,
      "auxiliary_loss_mlp": 0.00766063,
      "balance_loss_clip": 1.06124532,
      "balance_loss_mlp": 1.00073349,
      "epoch": 0.1438104972043528,
      "flos": 14684807669760.0,
      "grad_norm": 2.7261894411531684,
      "language_loss": 0.85220695,
      "learning_rate": 3.86577087324781e-06,
      "loss": 0.87217861,
      "num_input_tokens_seen": 25422040,
      "step": 1196,
      "time_per_iteration": 2.589552164077759
    },
    {
      "auxiliary_loss_clip": 0.01214195,
      "auxiliary_loss_mlp": 0.01033367,
      "balance_loss_clip": 1.06420279,
      "balance_loss_mlp": 1.0221374,
      "epoch": 0.14393074009499188,
      "flos": 17092330698240.0,
      "grad_norm": 4.845764736605291,
      "language_loss": 0.77121109,
      "learning_rate": 3.865490166995578e-06,
      "loss": 0.79368675,
      "num_input_tokens_seen": 25440270,
      "step": 1197,
      "time_per_iteration": 2.657001495361328
    },
    {
      "auxiliary_loss_clip": 0.01216498,
      "auxiliary_loss_mlp": 0.01035292,
      "balance_loss_clip": 1.06313467,
      "balance_loss_mlp": 1.02373409,
      "epoch": 0.144050982985631,
      "flos": 30476200608000.0,
      "grad_norm": 2.1936411710860506,
      "language_loss": 0.84291196,
      "learning_rate": 3.86520917774997e-06,
      "loss": 0.86542982,
      "num_input_tokens_seen": 25459705,
      "step": 1198,
      "time_per_iteration": 2.7346129417419434
    },
    {
      "auxiliary_loss_clip": 0.0121134,
      "auxiliary_loss_mlp": 0.01035997,
      "balance_loss_clip": 1.06165838,
      "balance_loss_mlp": 1.02542877,
      "epoch": 0.14417122587627007,
      "flos": 17858484817920.0,
      "grad_norm": 2.334827196203235,
      "language_loss": 0.74881321,
      "learning_rate": 3.864927905553614e-06,
      "loss": 0.77128661,
      "num_input_tokens_seen": 25477615,
      "step": 1199,
      "time_per_iteration": 2.621856451034546
    },
    {
      "auxiliary_loss_clip": 0.01181762,
      "auxiliary_loss_mlp": 0.01035753,
      "balance_loss_clip": 1.05647326,
      "balance_loss_mlp": 1.02470207,
      "epoch": 0.14429146876690915,
      "flos": 21613914639360.0,
      "grad_norm": 1.5863821606740192,
      "language_loss": 0.88727188,
      "learning_rate": 3.8646463504491765e-06,
      "loss": 0.90944707,
      "num_input_tokens_seen": 25497750,
      "step": 1200,
      "time_per_iteration": 2.7526328563690186
    },
    {
      "auxiliary_loss_clip": 0.01220059,
      "auxiliary_loss_mlp": 0.01045156,
      "balance_loss_clip": 1.06212604,
      "balance_loss_mlp": 1.03417075,
      "epoch": 0.14441171165754824,
      "flos": 23258120722560.0,
      "grad_norm": 2.1066813009988463,
      "language_loss": 0.83281273,
      "learning_rate": 3.8643645124793705e-06,
      "loss": 0.85546494,
      "num_input_tokens_seen": 25516650,
      "step": 1201,
      "time_per_iteration": 2.6592607498168945
    },
    {
      "auxiliary_loss_clip": 0.0120934,
      "auxiliary_loss_mlp": 0.01033881,
      "balance_loss_clip": 1.05877054,
      "balance_loss_mlp": 1.02354527,
      "epoch": 0.14453195454818735,
      "flos": 42854213963520.0,
      "grad_norm": 1.6566024758929419,
      "language_loss": 0.74934459,
      "learning_rate": 3.8640823916869515e-06,
      "loss": 0.7717768,
      "num_input_tokens_seen": 25540960,
      "step": 1202,
      "time_per_iteration": 2.8913087844848633
    },
    {
      "auxiliary_loss_clip": 0.0122843,
      "auxiliary_loss_mlp": 0.01046486,
      "balance_loss_clip": 1.06119752,
      "balance_loss_mlp": 1.0356375,
      "epoch": 0.14465219743882643,
      "flos": 27235873774080.0,
      "grad_norm": 1.5238064837978664,
      "language_loss": 0.78454107,
      "learning_rate": 3.863799988114714e-06,
      "loss": 0.8072902,
      "num_input_tokens_seen": 25562990,
      "step": 1203,
      "time_per_iteration": 2.723848819732666
    },
    {
      "auxiliary_loss_clip": 0.01231348,
      "auxiliary_loss_mlp": 0.01037456,
      "balance_loss_clip": 1.06131637,
      "balance_loss_mlp": 1.02622056,
      "epoch": 0.1447724403294655,
      "flos": 16690705752960.0,
      "grad_norm": 2.343943914485854,
      "language_loss": 0.70939064,
      "learning_rate": 3.863517301805502e-06,
      "loss": 0.73207867,
      "num_input_tokens_seen": 25581380,
      "step": 1204,
      "time_per_iteration": 2.6028659343719482
    },
    {
      "auxiliary_loss_clip": 0.01189008,
      "auxiliary_loss_mlp": 0.01042675,
      "balance_loss_clip": 1.06226504,
      "balance_loss_mlp": 1.03102195,
      "epoch": 0.14489268322010462,
      "flos": 20073741321600.0,
      "grad_norm": 2.828405867536094,
      "language_loss": 0.9720484,
      "learning_rate": 3.863234332802196e-06,
      "loss": 0.99436522,
      "num_input_tokens_seen": 25593585,
      "step": 1205,
      "time_per_iteration": 2.671290636062622
    },
    {
      "auxiliary_loss_clip": 0.01193546,
      "auxiliary_loss_mlp": 0.0104118,
      "balance_loss_clip": 1.05712402,
      "balance_loss_mlp": 1.03001642,
      "epoch": 0.1450129261107437,
      "flos": 27125627955840.0,
      "grad_norm": 3.209296282681346,
      "language_loss": 0.74069047,
      "learning_rate": 3.862951081147723e-06,
      "loss": 0.76303774,
      "num_input_tokens_seen": 25613750,
      "step": 1206,
      "time_per_iteration": 2.7038304805755615
    },
    {
      "auxiliary_loss_clip": 0.01219416,
      "auxiliary_loss_mlp": 0.01043425,
      "balance_loss_clip": 1.0648191,
      "balance_loss_mlp": 1.03272605,
      "epoch": 0.1451331690013828,
      "flos": 25702344472320.0,
      "grad_norm": 3.069748800779454,
      "language_loss": 0.78790188,
      "learning_rate": 3.862667546885053e-06,
      "loss": 0.81053019,
      "num_input_tokens_seen": 25632300,
      "step": 1207,
      "time_per_iteration": 2.6833574771881104
    },
    {
      "auxiliary_loss_clip": 0.01202918,
      "auxiliary_loss_mlp": 0.01040394,
      "balance_loss_clip": 1.05875838,
      "balance_loss_mlp": 1.0294503,
      "epoch": 0.14525341189202187,
      "flos": 25737393168000.0,
      "grad_norm": 1.912362446360852,
      "language_loss": 0.7377848,
      "learning_rate": 3.8623837300571965e-06,
      "loss": 0.76021796,
      "num_input_tokens_seen": 25651285,
      "step": 1208,
      "time_per_iteration": 2.729201078414917
    },
    {
      "auxiliary_loss_clip": 0.01230605,
      "auxiliary_loss_mlp": 0.01039819,
      "balance_loss_clip": 1.0626955,
      "balance_loss_mlp": 1.02857161,
      "epoch": 0.14537365478266098,
      "flos": 23073898844160.0,
      "grad_norm": 2.9063878517480988,
      "language_loss": 0.84056526,
      "learning_rate": 3.8620996307072085e-06,
      "loss": 0.86326945,
      "num_input_tokens_seen": 25671990,
      "step": 1209,
      "time_per_iteration": 2.649345636367798
    },
    {
      "auxiliary_loss_clip": 0.01187367,
      "auxiliary_loss_mlp": 0.01042372,
      "balance_loss_clip": 1.0552119,
      "balance_loss_mlp": 1.0307076,
      "epoch": 0.14549389767330007,
      "flos": 20595021448320.0,
      "grad_norm": 2.448998006961012,
      "language_loss": 0.64815927,
      "learning_rate": 3.861815248878188e-06,
      "loss": 0.67045665,
      "num_input_tokens_seen": 25689475,
      "step": 1210,
      "time_per_iteration": 2.6924753189086914
    },
    {
      "auxiliary_loss_clip": 0.01191914,
      "auxiliary_loss_mlp": 0.0103581,
      "balance_loss_clip": 1.05885947,
      "balance_loss_mlp": 1.02431774,
      "epoch": 0.14561414056393915,
      "flos": 15121804533120.0,
      "grad_norm": 7.0438819024340225,
      "language_loss": 0.80189013,
      "learning_rate": 3.861530584613274e-06,
      "loss": 0.82416731,
      "num_input_tokens_seen": 25707475,
      "step": 1211,
      "time_per_iteration": 2.6943118572235107
    },
    {
      "auxiliary_loss_clip": 0.01217193,
      "auxiliary_loss_mlp": 0.00765645,
      "balance_loss_clip": 1.06248999,
      "balance_loss_mlp": 1.00071084,
      "epoch": 0.14573438345457826,
      "flos": 19427493778560.0,
      "grad_norm": 2.217084007034587,
      "language_loss": 0.82685435,
      "learning_rate": 3.86124563795565e-06,
      "loss": 0.84668279,
      "num_input_tokens_seen": 25726290,
      "step": 1212,
      "time_per_iteration": 3.449018716812134
    },
    {
      "auxiliary_loss_clip": 0.01229928,
      "auxiliary_loss_mlp": 0.01042511,
      "balance_loss_clip": 1.06475008,
      "balance_loss_mlp": 1.03225303,
      "epoch": 0.14585462634521734,
      "flos": 24828422572800.0,
      "grad_norm": 1.8308858316356038,
      "language_loss": 0.70092964,
      "learning_rate": 3.860960408948543e-06,
      "loss": 0.72365397,
      "num_input_tokens_seen": 25748040,
      "step": 1213,
      "time_per_iteration": 2.797030448913574
    },
    {
      "auxiliary_loss_clip": 0.01210057,
      "auxiliary_loss_mlp": 0.01038355,
      "balance_loss_clip": 1.06341398,
      "balance_loss_mlp": 1.02776289,
      "epoch": 0.14597486923585642,
      "flos": 15448627405440.0,
      "grad_norm": 2.4749967712640317,
      "language_loss": 0.89233148,
      "learning_rate": 3.860674897635222e-06,
      "loss": 0.91481566,
      "num_input_tokens_seen": 25764525,
      "step": 1214,
      "time_per_iteration": 2.7346909046173096
    },
    {
      "auxiliary_loss_clip": 0.0121561,
      "auxiliary_loss_mlp": 0.0103526,
      "balance_loss_clip": 1.06315351,
      "balance_loss_mlp": 1.02474582,
      "epoch": 0.1460951121264955,
      "flos": 16655154266880.0,
      "grad_norm": 2.6773754531319502,
      "language_loss": 0.83647525,
      "learning_rate": 3.860389104058998e-06,
      "loss": 0.85898393,
      "num_input_tokens_seen": 25782755,
      "step": 1215,
      "time_per_iteration": 2.7007076740264893
    },
    {
      "auxiliary_loss_clip": 0.011968,
      "auxiliary_loss_mlp": 0.01034712,
      "balance_loss_clip": 1.05718493,
      "balance_loss_mlp": 1.02351797,
      "epoch": 0.14621535501713462,
      "flos": 24863291700480.0,
      "grad_norm": 1.8507080037177677,
      "language_loss": 0.72697031,
      "learning_rate": 3.860103028263227e-06,
      "loss": 0.74928546,
      "num_input_tokens_seen": 25805860,
      "step": 1216,
      "time_per_iteration": 2.741389751434326
    },
    {
      "auxiliary_loss_clip": 0.01157454,
      "auxiliary_loss_mlp": 0.01030721,
      "balance_loss_clip": 1.04766464,
      "balance_loss_mlp": 1.01968241,
      "epoch": 0.1463355979077737,
      "flos": 25228000442880.0,
      "grad_norm": 2.113300153065524,
      "language_loss": 0.70305192,
      "learning_rate": 3.859816670291304e-06,
      "loss": 0.72493374,
      "num_input_tokens_seen": 25824955,
      "step": 1217,
      "time_per_iteration": 4.647987365722656
    },
    {
      "auxiliary_loss_clip": 0.01146763,
      "auxiliary_loss_mlp": 0.01038952,
      "balance_loss_clip": 1.05488086,
      "balance_loss_mlp": 1.02829432,
      "epoch": 0.14645584079841278,
      "flos": 22054143726720.0,
      "grad_norm": 1.9008766322596444,
      "language_loss": 0.9021554,
      "learning_rate": 3.859530030186672e-06,
      "loss": 0.92401266,
      "num_input_tokens_seen": 25841965,
      "step": 1218,
      "time_per_iteration": 2.8006250858306885
    },
    {
      "auxiliary_loss_clip": 0.01201595,
      "auxiliary_loss_mlp": 0.01035074,
      "balance_loss_clip": 1.06003046,
      "balance_loss_mlp": 1.02304006,
      "epoch": 0.1465760836890519,
      "flos": 23623870959360.0,
      "grad_norm": 2.4363108320231266,
      "language_loss": 0.82830942,
      "learning_rate": 3.859243107992813e-06,
      "loss": 0.85067612,
      "num_input_tokens_seen": 25860770,
      "step": 1219,
      "time_per_iteration": 2.7219016551971436
    },
    {
      "auxiliary_loss_clip": 0.01183486,
      "auxiliary_loss_mlp": 0.01039444,
      "balance_loss_clip": 1.05102491,
      "balance_loss_mlp": 1.02775562,
      "epoch": 0.14669632657969098,
      "flos": 37407893356800.0,
      "grad_norm": 4.637679592290566,
      "language_loss": 0.78136623,
      "learning_rate": 3.858955903753252e-06,
      "loss": 0.80359548,
      "num_input_tokens_seen": 25879410,
      "step": 1220,
      "time_per_iteration": 2.8418514728546143
    },
    {
      "auxiliary_loss_clip": 0.01215223,
      "auxiliary_loss_mlp": 0.01048955,
      "balance_loss_clip": 1.0612855,
      "balance_loss_mlp": 1.03872085,
      "epoch": 0.14681656947033006,
      "flos": 28365910623360.0,
      "grad_norm": 1.5695511245293694,
      "language_loss": 0.83525586,
      "learning_rate": 3.858668417511559e-06,
      "loss": 0.85789764,
      "num_input_tokens_seen": 25902160,
      "step": 1221,
      "time_per_iteration": 2.74118709564209
    },
    {
      "auxiliary_loss_clip": 0.01202199,
      "auxiliary_loss_mlp": 0.01041443,
      "balance_loss_clip": 1.06095386,
      "balance_loss_mlp": 1.03038001,
      "epoch": 0.14693681236096917,
      "flos": 18479488078080.0,
      "grad_norm": 2.0825384376447484,
      "language_loss": 0.76783532,
      "learning_rate": 3.8583806493113445e-06,
      "loss": 0.7902717,
      "num_input_tokens_seen": 25920505,
      "step": 1222,
      "time_per_iteration": 2.758145809173584
    },
    {
      "auxiliary_loss_clip": 0.0121494,
      "auxiliary_loss_mlp": 0.01046133,
      "balance_loss_clip": 1.06453562,
      "balance_loss_mlp": 1.0351007,
      "epoch": 0.14705705525160825,
      "flos": 20777806782720.0,
      "grad_norm": 2.0887704716655238,
      "language_loss": 0.82391608,
      "learning_rate": 3.858092599196263e-06,
      "loss": 0.84652674,
      "num_input_tokens_seen": 25938460,
      "step": 1223,
      "time_per_iteration": 2.689225673675537
    },
    {
      "auxiliary_loss_clip": 0.01214397,
      "auxiliary_loss_mlp": 0.01041592,
      "balance_loss_clip": 1.06148612,
      "balance_loss_mlp": 1.03051138,
      "epoch": 0.14717729814224734,
      "flos": 29932944336000.0,
      "grad_norm": 3.23279300194928,
      "language_loss": 0.82237494,
      "learning_rate": 3.857804267210012e-06,
      "loss": 0.84493482,
      "num_input_tokens_seen": 25957760,
      "step": 1224,
      "time_per_iteration": 2.6660141944885254
    },
    {
      "auxiliary_loss_clip": 0.01172052,
      "auxiliary_loss_mlp": 0.01036203,
      "balance_loss_clip": 1.05426693,
      "balance_loss_mlp": 1.02502692,
      "epoch": 0.14729754103288642,
      "flos": 20047491457920.0,
      "grad_norm": 2.16563211812386,
      "language_loss": 0.88379008,
      "learning_rate": 3.857515653396331e-06,
      "loss": 0.90587264,
      "num_input_tokens_seen": 25974970,
      "step": 1225,
      "time_per_iteration": 2.6902647018432617
    },
    {
      "auxiliary_loss_clip": 0.01171989,
      "auxiliary_loss_mlp": 0.01037484,
      "balance_loss_clip": 1.0557642,
      "balance_loss_mlp": 1.02654099,
      "epoch": 0.14741778392352553,
      "flos": 19281516906240.0,
      "grad_norm": 2.1523590221682247,
      "language_loss": 0.87250215,
      "learning_rate": 3.857226757799002e-06,
      "loss": 0.89459693,
      "num_input_tokens_seen": 25992525,
      "step": 1226,
      "time_per_iteration": 2.755070924758911
    },
    {
      "auxiliary_loss_clip": 0.01198168,
      "auxiliary_loss_mlp": 0.01035144,
      "balance_loss_clip": 1.05614901,
      "balance_loss_mlp": 1.02449214,
      "epoch": 0.1475380268141646,
      "flos": 25411108999680.0,
      "grad_norm": 3.049291465948468,
      "language_loss": 0.74183846,
      "learning_rate": 3.85693758046185e-06,
      "loss": 0.7641716,
      "num_input_tokens_seen": 26010815,
      "step": 1227,
      "time_per_iteration": 2.7398159503936768
    },
    {
      "auxiliary_loss_clip": 0.01233688,
      "auxiliary_loss_mlp": 0.01039627,
      "balance_loss_clip": 1.06636882,
      "balance_loss_mlp": 1.0278852,
      "epoch": 0.1476582697048037,
      "flos": 20847652778880.0,
      "grad_norm": 1.719989944343744,
      "language_loss": 0.83042467,
      "learning_rate": 3.8566481214287435e-06,
      "loss": 0.85315788,
      "num_input_tokens_seen": 26028935,
      "step": 1228,
      "time_per_iteration": 2.6091153621673584
    },
    {
      "auxiliary_loss_clip": 0.01175198,
      "auxiliary_loss_mlp": 0.01046738,
      "balance_loss_clip": 1.05245531,
      "balance_loss_mlp": 1.03518116,
      "epoch": 0.1477785125954428,
      "flos": 14028109269120.0,
      "grad_norm": 2.024697828877587,
      "language_loss": 0.90692252,
      "learning_rate": 3.8563583807435935e-06,
      "loss": 0.92914188,
      "num_input_tokens_seen": 26045080,
      "step": 1229,
      "time_per_iteration": 2.8040363788604736
    },
    {
      "auxiliary_loss_clip": 0.01213082,
      "auxiliary_loss_mlp": 0.00765619,
      "balance_loss_clip": 1.05977607,
      "balance_loss_mlp": 1.00045073,
      "epoch": 0.1478987554860819,
      "flos": 20516699842560.0,
      "grad_norm": 1.7971076226353624,
      "language_loss": 0.77343976,
      "learning_rate": 3.856068358450353e-06,
      "loss": 0.79322672,
      "num_input_tokens_seen": 26065030,
      "step": 1230,
      "time_per_iteration": 2.712881088256836
    },
    {
      "auxiliary_loss_clip": 0.01196417,
      "auxiliary_loss_mlp": 0.01036849,
      "balance_loss_clip": 1.06255507,
      "balance_loss_mlp": 1.02533972,
      "epoch": 0.14801899837672097,
      "flos": 17857012360320.0,
      "grad_norm": 1.704728709384041,
      "language_loss": 0.86052501,
      "learning_rate": 3.8557780545930186e-06,
      "loss": 0.88285768,
      "num_input_tokens_seen": 26083445,
      "step": 1231,
      "time_per_iteration": 2.6880502700805664
    },
    {
      "auxiliary_loss_clip": 0.01195278,
      "auxiliary_loss_mlp": 0.01036512,
      "balance_loss_clip": 1.05822134,
      "balance_loss_mlp": 1.025527,
      "epoch": 0.14813924126736006,
      "flos": 20881408584960.0,
      "grad_norm": 1.8138645288613187,
      "language_loss": 0.79300463,
      "learning_rate": 3.855487469215628e-06,
      "loss": 0.81532246,
      "num_input_tokens_seen": 26102375,
      "step": 1232,
      "time_per_iteration": 2.676414966583252
    },
    {
      "auxiliary_loss_clip": 0.01179423,
      "auxiliary_loss_mlp": 0.01032504,
      "balance_loss_clip": 1.05518603,
      "balance_loss_mlp": 1.02151346,
      "epoch": 0.14825948415799917,
      "flos": 37414070496000.0,
      "grad_norm": 2.1431688119116306,
      "language_loss": 0.72569293,
      "learning_rate": 3.855196602362264e-06,
      "loss": 0.74781221,
      "num_input_tokens_seen": 26125295,
      "step": 1233,
      "time_per_iteration": 2.8021509647369385
    },
    {
      "auxiliary_loss_clip": 0.01213111,
      "auxiliary_loss_mlp": 0.01036529,
      "balance_loss_clip": 1.05954039,
      "balance_loss_mlp": 1.02527595,
      "epoch": 0.14837972704863825,
      "flos": 22014641744640.0,
      "grad_norm": 2.160107862357133,
      "language_loss": 0.94309986,
      "learning_rate": 3.854905454077051e-06,
      "loss": 0.96559632,
      "num_input_tokens_seen": 26142905,
      "step": 1234,
      "time_per_iteration": 2.6102452278137207
    },
    {
      "auxiliary_loss_clip": 0.01132882,
      "auxiliary_loss_mlp": 0.0103181,
      "balance_loss_clip": 1.0498848,
      "balance_loss_mlp": 1.02058685,
      "epoch": 0.14849996993927733,
      "flos": 20996323171200.0,
      "grad_norm": 2.0762463665928665,
      "language_loss": 0.8834399,
      "learning_rate": 3.854614024404155e-06,
      "loss": 0.90508682,
      "num_input_tokens_seen": 26161215,
      "step": 1235,
      "time_per_iteration": 2.753676652908325
    },
    {
      "auxiliary_loss_clip": 0.01183526,
      "auxiliary_loss_mlp": 0.0104003,
      "balance_loss_clip": 1.05339622,
      "balance_loss_mlp": 1.02891397,
      "epoch": 0.14862021282991644,
      "flos": 20047994248320.0,
      "grad_norm": 2.1900345219156496,
      "language_loss": 0.89382362,
      "learning_rate": 3.8543223133877865e-06,
      "loss": 0.91605914,
      "num_input_tokens_seen": 26179810,
      "step": 1236,
      "time_per_iteration": 2.686641216278076
    },
    {
      "auxiliary_loss_clip": 0.01178187,
      "auxiliary_loss_mlp": 0.01033472,
      "balance_loss_clip": 1.05352545,
      "balance_loss_mlp": 1.02150917,
      "epoch": 0.14874045572055553,
      "flos": 22712027276160.0,
      "grad_norm": 1.7254784548751605,
      "language_loss": 0.88607669,
      "learning_rate": 3.854030321072198e-06,
      "loss": 0.90819323,
      "num_input_tokens_seen": 26199715,
      "step": 1237,
      "time_per_iteration": 2.7118351459503174
    },
    {
      "auxiliary_loss_clip": 0.01187648,
      "auxiliary_loss_mlp": 0.01036741,
      "balance_loss_clip": 1.05731344,
      "balance_loss_mlp": 1.02608418,
      "epoch": 0.1488606986111946,
      "flos": 25411288567680.0,
      "grad_norm": 1.947605796348242,
      "language_loss": 0.73387992,
      "learning_rate": 3.853738047501682e-06,
      "loss": 0.75612384,
      "num_input_tokens_seen": 26220275,
      "step": 1238,
      "time_per_iteration": 3.721106767654419
    },
    {
      "auxiliary_loss_clip": 0.01214274,
      "auxiliary_loss_mlp": 0.01038241,
      "balance_loss_clip": 1.06100142,
      "balance_loss_mlp": 1.02708316,
      "epoch": 0.1489809415018337,
      "flos": 17018749687680.0,
      "grad_norm": 1.8854486018911207,
      "language_loss": 0.77631086,
      "learning_rate": 3.85344549272058e-06,
      "loss": 0.79883605,
      "num_input_tokens_seen": 26238255,
      "step": 1239,
      "time_per_iteration": 2.6427366733551025
    },
    {
      "auxiliary_loss_clip": 0.01206202,
      "auxiliary_loss_mlp": 0.01045409,
      "balance_loss_clip": 1.05728757,
      "balance_loss_mlp": 1.03465641,
      "epoch": 0.1491011843924728,
      "flos": 33659394860160.0,
      "grad_norm": 1.8034476970958224,
      "language_loss": 0.82624644,
      "learning_rate": 3.853152656773269e-06,
      "loss": 0.84876251,
      "num_input_tokens_seen": 26259690,
      "step": 1240,
      "time_per_iteration": 2.7432587146759033
    },
    {
      "auxiliary_loss_clip": 0.01195803,
      "auxiliary_loss_mlp": 0.01037,
      "balance_loss_clip": 1.05880284,
      "balance_loss_mlp": 1.02664661,
      "epoch": 0.14922142728311188,
      "flos": 21179000764800.0,
      "grad_norm": 1.8038287663705896,
      "language_loss": 0.85037917,
      "learning_rate": 3.852859539704174e-06,
      "loss": 0.87270713,
      "num_input_tokens_seen": 26278990,
      "step": 1241,
      "time_per_iteration": 2.663396120071411
    },
    {
      "auxiliary_loss_clip": 0.01168018,
      "auxiliary_loss_mlp": 0.01034375,
      "balance_loss_clip": 1.05496144,
      "balance_loss_mlp": 1.02315199,
      "epoch": 0.14934167017375097,
      "flos": 29860548474240.0,
      "grad_norm": 2.1693647055375997,
      "language_loss": 0.76379752,
      "learning_rate": 3.85256614155776e-06,
      "loss": 0.7858215,
      "num_input_tokens_seen": 26299120,
      "step": 1242,
      "time_per_iteration": 2.764119863510132
    },
    {
      "auxiliary_loss_clip": 0.01211666,
      "auxiliary_loss_mlp": 0.01037132,
      "balance_loss_clip": 1.05849314,
      "balance_loss_mlp": 1.02582479,
      "epoch": 0.14946191306439008,
      "flos": 17019216564480.0,
      "grad_norm": 3.867528129114803,
      "language_loss": 0.74389017,
      "learning_rate": 3.852272462378535e-06,
      "loss": 0.7663781,
      "num_input_tokens_seen": 26316995,
      "step": 1243,
      "time_per_iteration": 4.421874046325684
    },
    {
      "auxiliary_loss_clip": 0.01197605,
      "auxiliary_loss_mlp": 0.0104493,
      "balance_loss_clip": 1.0596931,
      "balance_loss_mlp": 1.03321171,
      "epoch": 0.14958215595502916,
      "flos": 15669047214720.0,
      "grad_norm": 2.2717501066451358,
      "language_loss": 0.77568674,
      "learning_rate": 3.85197850221105e-06,
      "loss": 0.79811215,
      "num_input_tokens_seen": 26333295,
      "step": 1244,
      "time_per_iteration": 2.577883243560791
    },
    {
      "auxiliary_loss_clip": 0.01213947,
      "auxiliary_loss_mlp": 0.01036381,
      "balance_loss_clip": 1.06350434,
      "balance_loss_mlp": 1.02564073,
      "epoch": 0.14970239884566824,
      "flos": 33108560818560.0,
      "grad_norm": 1.9109854236198984,
      "language_loss": 0.76320648,
      "learning_rate": 3.851684261099899e-06,
      "loss": 0.78570974,
      "num_input_tokens_seen": 26355035,
      "step": 1245,
      "time_per_iteration": 2.642407178878784
    },
    {
      "auxiliary_loss_clip": 0.01199269,
      "auxiliary_loss_mlp": 0.01039544,
      "balance_loss_clip": 1.06269789,
      "balance_loss_mlp": 1.02831483,
      "epoch": 0.14982264173630733,
      "flos": 17821245392640.0,
      "grad_norm": 1.9225540144171578,
      "language_loss": 0.86979043,
      "learning_rate": 3.851389739089718e-06,
      "loss": 0.89217854,
      "num_input_tokens_seen": 26371655,
      "step": 1246,
      "time_per_iteration": 2.55082368850708
    },
    {
      "auxiliary_loss_clip": 0.01217051,
      "auxiliary_loss_mlp": 0.01048271,
      "balance_loss_clip": 1.06520402,
      "balance_loss_mlp": 1.03636813,
      "epoch": 0.14994288462694644,
      "flos": 32409559175040.0,
      "grad_norm": 1.9765406680640472,
      "language_loss": 0.80425286,
      "learning_rate": 3.851094936225186e-06,
      "loss": 0.82690603,
      "num_input_tokens_seen": 26392540,
      "step": 1247,
      "time_per_iteration": 2.6267290115356445
    },
    {
      "auxiliary_loss_clip": 0.01193472,
      "auxiliary_loss_mlp": 0.01039562,
      "balance_loss_clip": 1.0598979,
      "balance_loss_mlp": 1.02893472,
      "epoch": 0.15006312751758552,
      "flos": 31794661226880.0,
      "grad_norm": 1.4403886486846316,
      "language_loss": 0.76860803,
      "learning_rate": 3.850799852551024e-06,
      "loss": 0.79093838,
      "num_input_tokens_seen": 26414960,
      "step": 1248,
      "time_per_iteration": 2.815251588821411
    },
    {
      "auxiliary_loss_clip": 0.01205783,
      "auxiliary_loss_mlp": 0.01036844,
      "balance_loss_clip": 1.05807889,
      "balance_loss_mlp": 1.0249052,
      "epoch": 0.1501833704082246,
      "flos": 16618022582400.0,
      "grad_norm": 2.098768869862525,
      "language_loss": 0.86086851,
      "learning_rate": 3.850504488111995e-06,
      "loss": 0.88329476,
      "num_input_tokens_seen": 26431635,
      "step": 1249,
      "time_per_iteration": 2.586219310760498
    },
    {
      "auxiliary_loss_clip": 0.01190596,
      "auxiliary_loss_mlp": 0.01037037,
      "balance_loss_clip": 1.05498421,
      "balance_loss_mlp": 1.02667797,
      "epoch": 0.15030361329886371,
      "flos": 23471178243840.0,
      "grad_norm": 1.7697391737190888,
      "language_loss": 0.82803303,
      "learning_rate": 3.850208842952907e-06,
      "loss": 0.85030931,
      "num_input_tokens_seen": 26450440,
      "step": 1250,
      "time_per_iteration": 2.7449941635131836
    },
    {
      "auxiliary_loss_clip": 0.01178829,
      "auxiliary_loss_mlp": 0.01038221,
      "balance_loss_clip": 1.05690706,
      "balance_loss_mlp": 1.02768242,
      "epoch": 0.1504238561895028,
      "flos": 25629409906560.0,
      "grad_norm": 2.6101226267658206,
      "language_loss": 0.79159009,
      "learning_rate": 3.849912917118608e-06,
      "loss": 0.81376064,
      "num_input_tokens_seen": 26471480,
      "step": 1251,
      "time_per_iteration": 2.779860258102417
    },
    {
      "auxiliary_loss_clip": 0.01122472,
      "auxiliary_loss_mlp": 0.0101448,
      "balance_loss_clip": 1.03202558,
      "balance_loss_mlp": 1.01111794,
      "epoch": 0.15054409908014188,
      "flos": 52095146129280.0,
      "grad_norm": 0.8775674771255021,
      "language_loss": 0.59289014,
      "learning_rate": 3.849616710653992e-06,
      "loss": 0.61425966,
      "num_input_tokens_seen": 26532950,
      "step": 1252,
      "time_per_iteration": 3.168269634246826
    },
    {
      "auxiliary_loss_clip": 0.01210076,
      "auxiliary_loss_mlp": 0.01040261,
      "balance_loss_clip": 1.05928135,
      "balance_loss_mlp": 1.02890658,
      "epoch": 0.150664341970781,
      "flos": 18880251096960.0,
      "grad_norm": 1.8689901611023974,
      "language_loss": 0.75104451,
      "learning_rate": 3.84932022360399e-06,
      "loss": 0.77354789,
      "num_input_tokens_seen": 26551615,
      "step": 1253,
      "time_per_iteration": 2.6280484199523926
    },
    {
      "auxiliary_loss_clip": 0.01195692,
      "auxiliary_loss_mlp": 0.01038377,
      "balance_loss_clip": 1.06087136,
      "balance_loss_mlp": 1.0280292,
      "epoch": 0.15078458486142007,
      "flos": 22163240309760.0,
      "grad_norm": 2.6089142786859885,
      "language_loss": 0.84899712,
      "learning_rate": 3.849023456013581e-06,
      "loss": 0.87133777,
      "num_input_tokens_seen": 26569175,
      "step": 1254,
      "time_per_iteration": 2.639106035232544
    },
    {
      "auxiliary_loss_clip": 0.01218913,
      "auxiliary_loss_mlp": 0.01039083,
      "balance_loss_clip": 1.06209481,
      "balance_loss_mlp": 1.02866983,
      "epoch": 0.15090482775205916,
      "flos": 26651894457600.0,
      "grad_norm": 2.120969529222285,
      "language_loss": 0.62871814,
      "learning_rate": 3.848726407927784e-06,
      "loss": 0.65129805,
      "num_input_tokens_seen": 26589560,
      "step": 1255,
      "time_per_iteration": 2.6973910331726074
    },
    {
      "auxiliary_loss_clip": 0.01202101,
      "auxiliary_loss_mlp": 0.01038571,
      "balance_loss_clip": 1.06353927,
      "balance_loss_mlp": 1.0270493,
      "epoch": 0.15102507064269824,
      "flos": 21798998444160.0,
      "grad_norm": 2.8240874565370095,
      "language_loss": 0.86614645,
      "learning_rate": 3.84842907939166e-06,
      "loss": 0.88855314,
      "num_input_tokens_seen": 26608785,
      "step": 1256,
      "time_per_iteration": 2.6589887142181396
    },
    {
      "auxiliary_loss_clip": 0.01173358,
      "auxiliary_loss_mlp": 0.01038454,
      "balance_loss_clip": 1.05466747,
      "balance_loss_mlp": 1.02765989,
      "epoch": 0.15114531353333735,
      "flos": 22820908377600.0,
      "grad_norm": 2.866018843632024,
      "language_loss": 0.70662761,
      "learning_rate": 3.8481314704503146e-06,
      "loss": 0.72874576,
      "num_input_tokens_seen": 26628615,
      "step": 1257,
      "time_per_iteration": 2.727477550506592
    },
    {
      "auxiliary_loss_clip": 0.01209913,
      "auxiliary_loss_mlp": 0.01044317,
      "balance_loss_clip": 1.06303227,
      "balance_loss_mlp": 1.03382051,
      "epoch": 0.15126555642397643,
      "flos": 19682674974720.0,
      "grad_norm": 2.314754622612884,
      "language_loss": 0.88067603,
      "learning_rate": 3.847833581148895e-06,
      "loss": 0.90321833,
      "num_input_tokens_seen": 26647525,
      "step": 1258,
      "time_per_iteration": 2.6216983795166016
    },
    {
      "auxiliary_loss_clip": 0.01224628,
      "auxiliary_loss_mlp": 0.01036248,
      "balance_loss_clip": 1.05974698,
      "balance_loss_mlp": 1.02529299,
      "epoch": 0.15138579931461552,
      "flos": 28726022424960.0,
      "grad_norm": 1.9793562406692984,
      "language_loss": 0.81299448,
      "learning_rate": 3.84753541153259e-06,
      "loss": 0.83560324,
      "num_input_tokens_seen": 26667095,
      "step": 1259,
      "time_per_iteration": 2.7240817546844482
    },
    {
      "auxiliary_loss_clip": 0.01214749,
      "auxiliary_loss_mlp": 0.01042208,
      "balance_loss_clip": 1.06173253,
      "balance_loss_mlp": 1.03119278,
      "epoch": 0.15150604220525463,
      "flos": 22127006465280.0,
      "grad_norm": 1.6380330321145067,
      "language_loss": 0.83293808,
      "learning_rate": 3.847236961646633e-06,
      "loss": 0.85550773,
      "num_input_tokens_seen": 26686075,
      "step": 1260,
      "time_per_iteration": 2.708862781524658
    },
    {
      "auxiliary_loss_clip": 0.01191439,
      "auxiliary_loss_mlp": 0.01043375,
      "balance_loss_clip": 1.05748093,
      "balance_loss_mlp": 1.03234172,
      "epoch": 0.1516262850958937,
      "flos": 12968708515200.0,
      "grad_norm": 2.715761959035085,
      "language_loss": 0.78000396,
      "learning_rate": 3.846938231536296e-06,
      "loss": 0.80235207,
      "num_input_tokens_seen": 26701695,
      "step": 1261,
      "time_per_iteration": 2.661027431488037
    },
    {
      "auxiliary_loss_clip": 0.01217566,
      "auxiliary_loss_mlp": 0.01040465,
      "balance_loss_clip": 1.06398702,
      "balance_loss_mlp": 1.02859211,
      "epoch": 0.1517465279865328,
      "flos": 21797130936960.0,
      "grad_norm": 1.8020792689945944,
      "language_loss": 0.81134558,
      "learning_rate": 3.8466392212468995e-06,
      "loss": 0.83392584,
      "num_input_tokens_seen": 26721885,
      "step": 1262,
      "time_per_iteration": 2.6364200115203857
    },
    {
      "auxiliary_loss_clip": 0.01098162,
      "auxiliary_loss_mlp": 0.01009657,
      "balance_loss_clip": 1.02627468,
      "balance_loss_mlp": 1.00629544,
      "epoch": 0.15186677087717187,
      "flos": 58174569901440.0,
      "grad_norm": 0.8279155592851141,
      "language_loss": 0.61927038,
      "learning_rate": 3.8463399308238e-06,
      "loss": 0.64034855,
      "num_input_tokens_seen": 26780990,
      "step": 1263,
      "time_per_iteration": 3.2125496864318848
    },
    {
      "auxiliary_loss_clip": 0.01213478,
      "auxiliary_loss_mlp": 0.01044309,
      "balance_loss_clip": 1.06178522,
      "balance_loss_mlp": 1.03349018,
      "epoch": 0.15198701376781099,
      "flos": 32669696448000.0,
      "grad_norm": 2.8151698714914954,
      "language_loss": 0.6421411,
      "learning_rate": 3.846040360312402e-06,
      "loss": 0.66471899,
      "num_input_tokens_seen": 26804250,
      "step": 1264,
      "time_per_iteration": 3.662544012069702
    },
    {
      "auxiliary_loss_clip": 0.01225093,
      "auxiliary_loss_mlp": 0.01042173,
      "balance_loss_clip": 1.05884719,
      "balance_loss_mlp": 1.0310626,
      "epoch": 0.15210725665845007,
      "flos": 28402575431040.0,
      "grad_norm": 2.203914782878996,
      "language_loss": 0.81170654,
      "learning_rate": 3.8457405097581485e-06,
      "loss": 0.83437932,
      "num_input_tokens_seen": 26823240,
      "step": 1265,
      "time_per_iteration": 2.6349384784698486
    },
    {
      "auxiliary_loss_clip": 0.01168788,
      "auxiliary_loss_mlp": 0.01036556,
      "balance_loss_clip": 1.0509758,
      "balance_loss_mlp": 1.02489734,
      "epoch": 0.15222749954908915,
      "flos": 19938179393280.0,
      "grad_norm": 1.7320737088909786,
      "language_loss": 0.78480041,
      "learning_rate": 3.8454403792065275e-06,
      "loss": 0.80685389,
      "num_input_tokens_seen": 26842060,
      "step": 1266,
      "time_per_iteration": 2.740194082260132
    },
    {
      "auxiliary_loss_clip": 0.01172993,
      "auxiliary_loss_mlp": 0.01048142,
      "balance_loss_clip": 1.05526578,
      "balance_loss_mlp": 1.0371685,
      "epoch": 0.15234774243972826,
      "flos": 21324223451520.0,
      "grad_norm": 2.0101612906626887,
      "language_loss": 0.85991836,
      "learning_rate": 3.845139968703068e-06,
      "loss": 0.88212967,
      "num_input_tokens_seen": 26859580,
      "step": 1267,
      "time_per_iteration": 2.6715123653411865
    },
    {
      "auxiliary_loss_clip": 0.01162759,
      "auxiliary_loss_mlp": 0.01038033,
      "balance_loss_clip": 1.05241513,
      "balance_loss_mlp": 1.02732229,
      "epoch": 0.15246798533036734,
      "flos": 25957812977280.0,
      "grad_norm": 1.9489894574471673,
      "language_loss": 0.8275584,
      "learning_rate": 3.844839278293342e-06,
      "loss": 0.84956634,
      "num_input_tokens_seen": 26880430,
      "step": 1268,
      "time_per_iteration": 2.7834231853485107
    },
    {
      "auxiliary_loss_clip": 0.01230933,
      "auxiliary_loss_mlp": 0.01038224,
      "balance_loss_clip": 1.0663023,
      "balance_loss_mlp": 1.02758491,
      "epoch": 0.15258822822100643,
      "flos": 25811907932160.0,
      "grad_norm": 2.2605617477309905,
      "language_loss": 0.76936567,
      "learning_rate": 3.8445383080229654e-06,
      "loss": 0.79205728,
      "num_input_tokens_seen": 26896445,
      "step": 1269,
      "time_per_iteration": 3.5650291442871094
    },
    {
      "auxiliary_loss_clip": 0.01187059,
      "auxiliary_loss_mlp": 0.01036245,
      "balance_loss_clip": 1.05597997,
      "balance_loss_mlp": 1.02438378,
      "epoch": 0.1527084711116455,
      "flos": 25265455349760.0,
      "grad_norm": 2.3146121832539244,
      "language_loss": 0.73455602,
      "learning_rate": 3.844237057937593e-06,
      "loss": 0.75678909,
      "num_input_tokens_seen": 26915450,
      "step": 1270,
      "time_per_iteration": 3.644118309020996
    },
    {
      "auxiliary_loss_clip": 0.01215851,
      "auxiliary_loss_mlp": 0.01035536,
      "balance_loss_clip": 1.06042027,
      "balance_loss_mlp": 1.02463377,
      "epoch": 0.15282871400228462,
      "flos": 29240227572480.0,
      "grad_norm": 2.379709745001183,
      "language_loss": 0.78080845,
      "learning_rate": 3.843935528082926e-06,
      "loss": 0.80332232,
      "num_input_tokens_seen": 26936475,
      "step": 1271,
      "time_per_iteration": 2.6751418113708496
    },
    {
      "auxiliary_loss_clip": 0.01210739,
      "auxiliary_loss_mlp": 0.0103945,
      "balance_loss_clip": 1.05877256,
      "balance_loss_mlp": 1.02797627,
      "epoch": 0.1529489568929237,
      "flos": 20882952869760.0,
      "grad_norm": 1.7119555867298408,
      "language_loss": 0.84927511,
      "learning_rate": 3.843633718504704e-06,
      "loss": 0.87177694,
      "num_input_tokens_seen": 26954920,
      "step": 1272,
      "time_per_iteration": 2.6493210792541504
    },
    {
      "auxiliary_loss_clip": 0.01182148,
      "auxiliary_loss_mlp": 0.01036479,
      "balance_loss_clip": 1.05598104,
      "balance_loss_mlp": 1.02492166,
      "epoch": 0.1530691997835628,
      "flos": 20083833043200.0,
      "grad_norm": 2.41101394043956,
      "language_loss": 0.90229356,
      "learning_rate": 3.843331629248715e-06,
      "loss": 0.92447984,
      "num_input_tokens_seen": 26972520,
      "step": 1273,
      "time_per_iteration": 2.736314296722412
    },
    {
      "auxiliary_loss_clip": 0.01227278,
      "auxiliary_loss_mlp": 0.01035449,
      "balance_loss_clip": 1.06334352,
      "balance_loss_mlp": 1.02500033,
      "epoch": 0.1531894426742019,
      "flos": 28759814144640.0,
      "grad_norm": 2.11349647091796,
      "language_loss": 0.76929057,
      "learning_rate": 3.843029260360782e-06,
      "loss": 0.79191786,
      "num_input_tokens_seen": 26990890,
      "step": 1274,
      "time_per_iteration": 2.681389331817627
    },
    {
      "auxiliary_loss_clip": 0.01209068,
      "auxiliary_loss_mlp": 0.01039604,
      "balance_loss_clip": 1.06064916,
      "balance_loss_mlp": 1.02941132,
      "epoch": 0.15330968556484098,
      "flos": 22236282616320.0,
      "grad_norm": 2.0508964171238455,
      "language_loss": 0.78829205,
      "learning_rate": 3.8427266118867755e-06,
      "loss": 0.8107788,
      "num_input_tokens_seen": 27010640,
      "step": 1275,
      "time_per_iteration": 2.7134149074554443
    },
    {
      "auxiliary_loss_clip": 0.01196377,
      "auxiliary_loss_mlp": 0.0103788,
      "balance_loss_clip": 1.05854177,
      "balance_loss_mlp": 1.02660239,
      "epoch": 0.15342992845548006,
      "flos": 27527504296320.0,
      "grad_norm": 3.9984130307014487,
      "language_loss": 0.83514953,
      "learning_rate": 3.842423683872608e-06,
      "loss": 0.85749209,
      "num_input_tokens_seen": 27031215,
      "step": 1276,
      "time_per_iteration": 2.768399953842163
    },
    {
      "auxiliary_loss_clip": 0.01212443,
      "auxiliary_loss_mlp": 0.01037727,
      "balance_loss_clip": 1.06028962,
      "balance_loss_mlp": 1.02625966,
      "epoch": 0.15355017134611917,
      "flos": 19609596754560.0,
      "grad_norm": 5.278392600736629,
      "language_loss": 0.77593124,
      "learning_rate": 3.842120476364232e-06,
      "loss": 0.79843295,
      "num_input_tokens_seen": 27049665,
      "step": 1277,
      "time_per_iteration": 2.5766642093658447
    },
    {
      "auxiliary_loss_clip": 0.01214933,
      "auxiliary_loss_mlp": 0.01048325,
      "balance_loss_clip": 1.05987394,
      "balance_loss_mlp": 1.03670812,
      "epoch": 0.15367041423675826,
      "flos": 18478590238080.0,
      "grad_norm": 2.2094644234524177,
      "language_loss": 0.8361268,
      "learning_rate": 3.841816989407644e-06,
      "loss": 0.8587594,
      "num_input_tokens_seen": 27065155,
      "step": 1278,
      "time_per_iteration": 2.6689441204071045
    },
    {
      "auxiliary_loss_clip": 0.01180367,
      "auxiliary_loss_mlp": 0.0103468,
      "balance_loss_clip": 1.05645967,
      "balance_loss_mlp": 1.0239929,
      "epoch": 0.15379065712739734,
      "flos": 41427662342400.0,
      "grad_norm": 2.026232934352127,
      "language_loss": 0.7673614,
      "learning_rate": 3.841513223048884e-06,
      "loss": 0.78951192,
      "num_input_tokens_seen": 27085840,
      "step": 1279,
      "time_per_iteration": 2.832150459289551
    },
    {
      "auxiliary_loss_clip": 0.01180082,
      "auxiliary_loss_mlp": 0.01041048,
      "balance_loss_clip": 1.05566669,
      "balance_loss_mlp": 1.030123,
      "epoch": 0.15391090001803642,
      "flos": 22054215553920.0,
      "grad_norm": 2.593592270047382,
      "language_loss": 0.78879118,
      "learning_rate": 3.841209177334031e-06,
      "loss": 0.81100249,
      "num_input_tokens_seen": 27104200,
      "step": 1280,
      "time_per_iteration": 2.7665064334869385
    },
    {
      "auxiliary_loss_clip": 0.01209514,
      "auxiliary_loss_mlp": 0.01041037,
      "balance_loss_clip": 1.0622201,
      "balance_loss_mlp": 1.03032613,
      "epoch": 0.15403114290867553,
      "flos": 15450351258240.0,
      "grad_norm": 1.8842519683169598,
      "language_loss": 0.7498098,
      "learning_rate": 3.84090485230921e-06,
      "loss": 0.77231526,
      "num_input_tokens_seen": 27122440,
      "step": 1281,
      "time_per_iteration": 2.598557949066162
    },
    {
      "auxiliary_loss_clip": 0.01226399,
      "auxiliary_loss_mlp": 0.01032292,
      "balance_loss_clip": 1.06296921,
      "balance_loss_mlp": 1.02124786,
      "epoch": 0.15415138579931462,
      "flos": 17929156826880.0,
      "grad_norm": 7.22433559557855,
      "language_loss": 0.76386023,
      "learning_rate": 3.840600248020588e-06,
      "loss": 0.78644717,
      "num_input_tokens_seen": 27139380,
      "step": 1282,
      "time_per_iteration": 2.8246712684631348
    },
    {
      "auxiliary_loss_clip": 0.01201647,
      "auxiliary_loss_mlp": 0.0103965,
      "balance_loss_clip": 1.05748272,
      "balance_loss_mlp": 1.02825916,
      "epoch": 0.1542716286899537,
      "flos": 11429325296640.0,
      "grad_norm": 3.2395284955879258,
      "language_loss": 0.80170673,
      "learning_rate": 3.840295364514371e-06,
      "loss": 0.82411969,
      "num_input_tokens_seen": 27156760,
      "step": 1283,
      "time_per_iteration": 2.645066261291504
    },
    {
      "auxiliary_loss_clip": 0.01195441,
      "auxiliary_loss_mlp": 0.01037822,
      "balance_loss_clip": 1.05800343,
      "balance_loss_mlp": 1.02767777,
      "epoch": 0.1543918715805928,
      "flos": 17420338719360.0,
      "grad_norm": 2.6566751164067064,
      "language_loss": 0.78854489,
      "learning_rate": 3.83999020183681e-06,
      "loss": 0.81087756,
      "num_input_tokens_seen": 27175455,
      "step": 1284,
      "time_per_iteration": 2.6616122722625732
    },
    {
      "auxiliary_loss_clip": 0.01147475,
      "auxiliary_loss_mlp": 0.0103352,
      "balance_loss_clip": 1.05147338,
      "balance_loss_mlp": 1.02289808,
      "epoch": 0.1545121144712319,
      "flos": 17786376264960.0,
      "grad_norm": 1.8245517485877292,
      "language_loss": 0.78946495,
      "learning_rate": 3.839684760034199e-06,
      "loss": 0.81127489,
      "num_input_tokens_seen": 27193660,
      "step": 1285,
      "time_per_iteration": 2.7298710346221924
    },
    {
      "auxiliary_loss_clip": 0.01174063,
      "auxiliary_loss_mlp": 0.01034765,
      "balance_loss_clip": 1.05754209,
      "balance_loss_mlp": 1.02365494,
      "epoch": 0.15463235736187098,
      "flos": 28220185146240.0,
      "grad_norm": 2.63103091412387,
      "language_loss": 0.65142268,
      "learning_rate": 3.8393790391528716e-06,
      "loss": 0.67351097,
      "num_input_tokens_seen": 27214355,
      "step": 1286,
      "time_per_iteration": 2.7687134742736816
    },
    {
      "auxiliary_loss_clip": 0.01195758,
      "auxiliary_loss_mlp": 0.01033107,
      "balance_loss_clip": 1.05759835,
      "balance_loss_mlp": 1.02236056,
      "epoch": 0.15475260025251006,
      "flos": 22856890826880.0,
      "grad_norm": 1.9558125119649594,
      "language_loss": 0.89503682,
      "learning_rate": 3.8390730392392075e-06,
      "loss": 0.9173255,
      "num_input_tokens_seen": 27234335,
      "step": 1287,
      "time_per_iteration": 2.7378504276275635
    },
    {
      "auxiliary_loss_clip": 0.01230329,
      "auxiliary_loss_mlp": 0.01030061,
      "balance_loss_clip": 1.06604218,
      "balance_loss_mlp": 1.01977909,
      "epoch": 0.15487284314314917,
      "flos": 17602872658560.0,
      "grad_norm": 2.393964559601533,
      "language_loss": 0.79352933,
      "learning_rate": 3.838766760339626e-06,
      "loss": 0.81613326,
      "num_input_tokens_seen": 27252860,
      "step": 1288,
      "time_per_iteration": 2.583246946334839
    },
    {
      "auxiliary_loss_clip": 0.01159915,
      "auxiliary_loss_mlp": 0.01035622,
      "balance_loss_clip": 1.05227792,
      "balance_loss_mlp": 1.0242312,
      "epoch": 0.15499308603378825,
      "flos": 20082037363200.0,
      "grad_norm": 2.5679836129045537,
      "language_loss": 0.792768,
      "learning_rate": 3.838460202500587e-06,
      "loss": 0.81472337,
      "num_input_tokens_seen": 27268650,
      "step": 1289,
      "time_per_iteration": 2.7179596424102783
    },
    {
      "auxiliary_loss_clip": 0.01181499,
      "auxiliary_loss_mlp": 0.01036261,
      "balance_loss_clip": 1.06191945,
      "balance_loss_mlp": 1.02535355,
      "epoch": 0.15511332892442733,
      "flos": 15918051271680.0,
      "grad_norm": 2.8762997256528533,
      "language_loss": 0.7399981,
      "learning_rate": 3.838153365768599e-06,
      "loss": 0.76217568,
      "num_input_tokens_seen": 27285160,
      "step": 1290,
      "time_per_iteration": 3.6000165939331055
    },
    {
      "auxiliary_loss_clip": 0.01181552,
      "auxiliary_loss_mlp": 0.01036968,
      "balance_loss_clip": 1.06176651,
      "balance_loss_mlp": 1.02560091,
      "epoch": 0.15523357181506645,
      "flos": 41282475569280.0,
      "grad_norm": 2.338496168978605,
      "language_loss": 0.7477892,
      "learning_rate": 3.837846250190206e-06,
      "loss": 0.76997441,
      "num_input_tokens_seen": 27308025,
      "step": 1291,
      "time_per_iteration": 2.849884510040283
    },
    {
      "auxiliary_loss_clip": 0.01164162,
      "auxiliary_loss_mlp": 0.00764859,
      "balance_loss_clip": 1.05550861,
      "balance_loss_mlp": 1.00049973,
      "epoch": 0.15535381470570553,
      "flos": 18478769806080.0,
      "grad_norm": 2.0433099659576057,
      "language_loss": 0.76871532,
      "learning_rate": 3.837538855811998e-06,
      "loss": 0.78800553,
      "num_input_tokens_seen": 27326200,
      "step": 1292,
      "time_per_iteration": 2.691004514694214
    },
    {
      "auxiliary_loss_clip": 0.01200183,
      "auxiliary_loss_mlp": 0.01047123,
      "balance_loss_clip": 1.05987108,
      "balance_loss_mlp": 1.03559542,
      "epoch": 0.1554740575963446,
      "flos": 13918150759680.0,
      "grad_norm": 2.3016446650573936,
      "language_loss": 0.70889395,
      "learning_rate": 3.837231182680606e-06,
      "loss": 0.73136699,
      "num_input_tokens_seen": 27344165,
      "step": 1293,
      "time_per_iteration": 2.7653911113739014
    },
    {
      "auxiliary_loss_clip": 0.01212321,
      "auxiliary_loss_mlp": 0.0103523,
      "balance_loss_clip": 1.05959964,
      "balance_loss_mlp": 1.02510309,
      "epoch": 0.1555943004869837,
      "flos": 20847078161280.0,
      "grad_norm": 1.856622557441132,
      "language_loss": 0.76360977,
      "learning_rate": 3.836923230842706e-06,
      "loss": 0.78608525,
      "num_input_tokens_seen": 27363280,
      "step": 1294,
      "time_per_iteration": 2.653813123703003
    },
    {
      "auxiliary_loss_clip": 0.01166589,
      "auxiliary_loss_mlp": 0.01034666,
      "balance_loss_clip": 1.05222058,
      "balance_loss_mlp": 1.02415776,
      "epoch": 0.1557145433776228,
      "flos": 22085888371200.0,
      "grad_norm": 2.4241691458439214,
      "language_loss": 0.8078742,
      "learning_rate": 3.836615000345011e-06,
      "loss": 0.82988679,
      "num_input_tokens_seen": 27381460,
      "step": 1295,
      "time_per_iteration": 5.447427272796631
    },
    {
      "auxiliary_loss_clip": 0.01226344,
      "auxiliary_loss_mlp": 0.01034741,
      "balance_loss_clip": 1.06285632,
      "balance_loss_mlp": 1.02392864,
      "epoch": 0.1558347862682619,
      "flos": 19791987039360.0,
      "grad_norm": 2.239363065816323,
      "language_loss": 0.77996665,
      "learning_rate": 3.836306491234282e-06,
      "loss": 0.8025775,
      "num_input_tokens_seen": 27399310,
      "step": 1296,
      "time_per_iteration": 2.57647705078125
    },
    {
      "auxiliary_loss_clip": 0.01193919,
      "auxiliary_loss_mlp": 0.01040276,
      "balance_loss_clip": 1.06381035,
      "balance_loss_mlp": 1.02985716,
      "epoch": 0.15595502915890097,
      "flos": 17237086508160.0,
      "grad_norm": 2.4698675423644625,
      "language_loss": 0.76080388,
      "learning_rate": 3.835997703557317e-06,
      "loss": 0.78314584,
      "num_input_tokens_seen": 27416050,
      "step": 1297,
      "time_per_iteration": 2.6633152961730957
    },
    {
      "auxiliary_loss_clip": 0.01164481,
      "auxiliary_loss_mlp": 0.01035777,
      "balance_loss_clip": 1.05081856,
      "balance_loss_mlp": 1.0254004,
      "epoch": 0.15607527204954008,
      "flos": 19719519350400.0,
      "grad_norm": 1.881595972596585,
      "language_loss": 0.80037606,
      "learning_rate": 3.83568863736096e-06,
      "loss": 0.82237864,
      "num_input_tokens_seen": 27434920,
      "step": 1298,
      "time_per_iteration": 2.730390787124634
    },
    {
      "auxiliary_loss_clip": 0.0117888,
      "auxiliary_loss_mlp": 0.01039281,
      "balance_loss_clip": 1.05119789,
      "balance_loss_mlp": 1.02861762,
      "epoch": 0.15619551494017916,
      "flos": 18515650095360.0,
      "grad_norm": 3.5045548507245923,
      "language_loss": 0.89135909,
      "learning_rate": 3.8353792926920975e-06,
      "loss": 0.91354072,
      "num_input_tokens_seen": 27453570,
      "step": 1299,
      "time_per_iteration": 2.6947710514068604
    },
    {
      "auxiliary_loss_clip": 0.01220166,
      "auxiliary_loss_mlp": 0.01036328,
      "balance_loss_clip": 1.06393528,
      "balance_loss_mlp": 1.02431214,
      "epoch": 0.15631575783081825,
      "flos": 19902125116800.0,
      "grad_norm": 2.4166952092060554,
      "language_loss": 0.81919348,
      "learning_rate": 3.835069669597655e-06,
      "loss": 0.84175843,
      "num_input_tokens_seen": 27471960,
      "step": 1300,
      "time_per_iteration": 2.6733474731445312
    },
    {
      "auxiliary_loss_clip": 0.01215533,
      "auxiliary_loss_mlp": 0.00765728,
      "balance_loss_clip": 1.06076205,
      "balance_loss_mlp": 1.0005548,
      "epoch": 0.15643600072145733,
      "flos": 20777663128320.0,
      "grad_norm": 2.7407490348045576,
      "language_loss": 0.79506087,
      "learning_rate": 3.834759768124603e-06,
      "loss": 0.81487346,
      "num_input_tokens_seen": 27490835,
      "step": 1301,
      "time_per_iteration": 2.6863908767700195
    },
    {
      "auxiliary_loss_clip": 0.01186797,
      "auxiliary_loss_mlp": 0.0103821,
      "balance_loss_clip": 1.06176066,
      "balance_loss_mlp": 1.02717161,
      "epoch": 0.15655624361209644,
      "flos": 18546389159040.0,
      "grad_norm": 2.166714178251868,
      "language_loss": 0.76257992,
      "learning_rate": 3.834449588319953e-06,
      "loss": 0.78482997,
      "num_input_tokens_seen": 27508870,
      "step": 1302,
      "time_per_iteration": 2.7003772258758545
    },
    {
      "auxiliary_loss_clip": 0.0121121,
      "auxiliary_loss_mlp": 0.0103945,
      "balance_loss_clip": 1.06599391,
      "balance_loss_mlp": 1.02938914,
      "epoch": 0.15667648650273552,
      "flos": 25229544727680.0,
      "grad_norm": 1.9437962396540251,
      "language_loss": 0.85080099,
      "learning_rate": 3.834139130230758e-06,
      "loss": 0.87330759,
      "num_input_tokens_seen": 27528175,
      "step": 1303,
      "time_per_iteration": 2.7817845344543457
    },
    {
      "auxiliary_loss_clip": 0.01197612,
      "auxiliary_loss_mlp": 0.01040529,
      "balance_loss_clip": 1.0567503,
      "balance_loss_mlp": 1.02953172,
      "epoch": 0.1567967293933746,
      "flos": 24827093769600.0,
      "grad_norm": 1.6468386960634478,
      "language_loss": 0.81439406,
      "learning_rate": 3.833828393904117e-06,
      "loss": 0.83677548,
      "num_input_tokens_seen": 27548455,
      "step": 1304,
      "time_per_iteration": 2.730656623840332
    },
    {
      "auxiliary_loss_clip": 0.01162226,
      "auxiliary_loss_mlp": 0.01047403,
      "balance_loss_clip": 1.05428195,
      "balance_loss_mlp": 1.03696668,
      "epoch": 0.15691697228401372,
      "flos": 19164555244800.0,
      "grad_norm": 2.263429126983323,
      "language_loss": 0.77465695,
      "learning_rate": 3.833517379387165e-06,
      "loss": 0.79675323,
      "num_input_tokens_seen": 27564910,
      "step": 1305,
      "time_per_iteration": 2.7459049224853516
    },
    {
      "auxiliary_loss_clip": 0.01212993,
      "auxiliary_loss_mlp": 0.01040287,
      "balance_loss_clip": 1.06077123,
      "balance_loss_mlp": 1.02987432,
      "epoch": 0.1570372151746528,
      "flos": 24790931752320.0,
      "grad_norm": 1.898011724123316,
      "language_loss": 0.88723552,
      "learning_rate": 3.833206086727085e-06,
      "loss": 0.9097684,
      "num_input_tokens_seen": 27584260,
      "step": 1306,
      "time_per_iteration": 2.86450457572937
    },
    {
      "auxiliary_loss_clip": 0.01181492,
      "auxiliary_loss_mlp": 0.01036395,
      "balance_loss_clip": 1.05282438,
      "balance_loss_mlp": 1.02578568,
      "epoch": 0.15715745806529188,
      "flos": 24863650836480.0,
      "grad_norm": 2.2892520885911556,
      "language_loss": 0.70732301,
      "learning_rate": 3.8328945159710994e-06,
      "loss": 0.7295019,
      "num_input_tokens_seen": 27604440,
      "step": 1307,
      "time_per_iteration": 2.7197964191436768
    },
    {
      "auxiliary_loss_clip": 0.01223028,
      "auxiliary_loss_mlp": 0.00765431,
      "balance_loss_clip": 1.06799102,
      "balance_loss_mlp": 1.00047064,
      "epoch": 0.157277700955931,
      "flos": 21872148491520.0,
      "grad_norm": 2.2442982339529753,
      "language_loss": 0.88882279,
      "learning_rate": 3.832582667166473e-06,
      "loss": 0.90870732,
      "num_input_tokens_seen": 27624250,
      "step": 1308,
      "time_per_iteration": 2.7076189517974854
    },
    {
      "auxiliary_loss_clip": 0.01195606,
      "auxiliary_loss_mlp": 0.01044342,
      "balance_loss_clip": 1.05981553,
      "balance_loss_mlp": 1.03311241,
      "epoch": 0.15739794384657008,
      "flos": 24533344344960.0,
      "grad_norm": 1.8370767738617375,
      "language_loss": 0.81799906,
      "learning_rate": 3.8322705403605125e-06,
      "loss": 0.84039855,
      "num_input_tokens_seen": 27644595,
      "step": 1309,
      "time_per_iteration": 2.751643657684326
    },
    {
      "auxiliary_loss_clip": 0.01188594,
      "auxiliary_loss_mlp": 0.01033042,
      "balance_loss_clip": 1.05698895,
      "balance_loss_mlp": 1.02235484,
      "epoch": 0.15751818673720916,
      "flos": 17745329998080.0,
      "grad_norm": 2.025354039780842,
      "language_loss": 0.81564081,
      "learning_rate": 3.831958135600568e-06,
      "loss": 0.83785725,
      "num_input_tokens_seen": 27662145,
      "step": 1310,
      "time_per_iteration": 2.747087240219116
    },
    {
      "auxiliary_loss_clip": 0.01214308,
      "auxiliary_loss_mlp": 0.01041787,
      "balance_loss_clip": 1.06346655,
      "balance_loss_mlp": 1.03123116,
      "epoch": 0.15763842962784824,
      "flos": 17858520731520.0,
      "grad_norm": 2.0888562355766638,
      "language_loss": 0.7936219,
      "learning_rate": 3.831645452934032e-06,
      "loss": 0.81618285,
      "num_input_tokens_seen": 27680575,
      "step": 1311,
      "time_per_iteration": 2.6973438262939453
    },
    {
      "auxiliary_loss_clip": 0.01230768,
      "auxiliary_loss_mlp": 0.01040346,
      "balance_loss_clip": 1.0676899,
      "balance_loss_mlp": 1.02847838,
      "epoch": 0.15775867251848735,
      "flos": 26980908059520.0,
      "grad_norm": 1.918264281530322,
      "language_loss": 0.80265933,
      "learning_rate": 3.831332492408336e-06,
      "loss": 0.82537049,
      "num_input_tokens_seen": 27701985,
      "step": 1312,
      "time_per_iteration": 2.6607227325439453
    },
    {
      "auxiliary_loss_clip": 0.01192595,
      "auxiliary_loss_mlp": 0.01031664,
      "balance_loss_clip": 1.05737209,
      "balance_loss_mlp": 1.02103686,
      "epoch": 0.15787891540912644,
      "flos": 19240398812160.0,
      "grad_norm": 2.7755379098816166,
      "language_loss": 0.69333172,
      "learning_rate": 3.831019254070957e-06,
      "loss": 0.71557438,
      "num_input_tokens_seen": 27719770,
      "step": 1313,
      "time_per_iteration": 2.6310677528381348
    },
    {
      "auxiliary_loss_clip": 0.01167983,
      "auxiliary_loss_mlp": 0.01039068,
      "balance_loss_clip": 1.05372524,
      "balance_loss_mlp": 1.028548,
      "epoch": 0.15799915829976552,
      "flos": 27271102037760.0,
      "grad_norm": 3.2528012643946744,
      "language_loss": 0.94788027,
      "learning_rate": 3.8307057379694135e-06,
      "loss": 0.96995068,
      "num_input_tokens_seen": 27739105,
      "step": 1314,
      "time_per_iteration": 2.79300594329834
    },
    {
      "auxiliary_loss_clip": 0.01228975,
      "auxiliary_loss_mlp": 0.01038354,
      "balance_loss_clip": 1.06251121,
      "balance_loss_mlp": 1.02802491,
      "epoch": 0.15811940119040463,
      "flos": 20405520270720.0,
      "grad_norm": 2.2933259017625085,
      "language_loss": 0.82580692,
      "learning_rate": 3.830391944151264e-06,
      "loss": 0.84848017,
      "num_input_tokens_seen": 27754985,
      "step": 1315,
      "time_per_iteration": 2.5895400047302246
    },
    {
      "auxiliary_loss_clip": 0.01198072,
      "auxiliary_loss_mlp": 0.01035952,
      "balance_loss_clip": 1.05833983,
      "balance_loss_mlp": 1.02533031,
      "epoch": 0.1582396440810437,
      "flos": 32599347661440.0,
      "grad_norm": 1.7421330026681197,
      "language_loss": 0.67478943,
      "learning_rate": 3.830077872664114e-06,
      "loss": 0.69712973,
      "num_input_tokens_seen": 27776110,
      "step": 1316,
      "time_per_iteration": 3.8042240142822266
    },
    {
      "auxiliary_loss_clip": 0.01154932,
      "auxiliary_loss_mlp": 0.01045039,
      "balance_loss_clip": 1.05350304,
      "balance_loss_mlp": 1.03472114,
      "epoch": 0.1583598869716828,
      "flos": 33800559310080.0,
      "grad_norm": 1.7706233568705718,
      "language_loss": 0.73570228,
      "learning_rate": 3.829763523555604e-06,
      "loss": 0.75770199,
      "num_input_tokens_seen": 27796510,
      "step": 1317,
      "time_per_iteration": 2.906458616256714
    },
    {
      "auxiliary_loss_clip": 0.01211216,
      "auxiliary_loss_mlp": 0.01038609,
      "balance_loss_clip": 1.06701887,
      "balance_loss_mlp": 1.02843451,
      "epoch": 0.15848012986232188,
      "flos": 24681332378880.0,
      "grad_norm": 2.29847518511585,
      "language_loss": 0.7858097,
      "learning_rate": 3.829448896873423e-06,
      "loss": 0.80830801,
      "num_input_tokens_seen": 27815610,
      "step": 1318,
      "time_per_iteration": 2.6642143726348877
    },
    {
      "auxiliary_loss_clip": 0.01155234,
      "auxiliary_loss_mlp": 0.00765891,
      "balance_loss_clip": 1.05721354,
      "balance_loss_mlp": 1.00052655,
      "epoch": 0.158600372752961,
      "flos": 22602068766720.0,
      "grad_norm": 2.067408346680746,
      "language_loss": 0.79493648,
      "learning_rate": 3.829133992665299e-06,
      "loss": 0.81414771,
      "num_input_tokens_seen": 27834735,
      "step": 1319,
      "time_per_iteration": 2.7497541904449463
    },
    {
      "auxiliary_loss_clip": 0.01202104,
      "auxiliary_loss_mlp": 0.01034725,
      "balance_loss_clip": 1.06061745,
      "balance_loss_mlp": 1.02459788,
      "epoch": 0.15872061564360007,
      "flos": 27927944092800.0,
      "grad_norm": 2.3398467983765707,
      "language_loss": 0.8901279,
      "learning_rate": 3.828818810979002e-06,
      "loss": 0.91249615,
      "num_input_tokens_seen": 27853065,
      "step": 1320,
      "time_per_iteration": 4.507674217224121
    },
    {
      "auxiliary_loss_clip": 0.01228672,
      "auxiliary_loss_mlp": 0.010434,
      "balance_loss_clip": 1.06649208,
      "balance_loss_mlp": 1.03298068,
      "epoch": 0.15884085853423915,
      "flos": 23696805525120.0,
      "grad_norm": 1.7204090021091543,
      "language_loss": 0.80384701,
      "learning_rate": 3.8285033518623454e-06,
      "loss": 0.82656777,
      "num_input_tokens_seen": 27873315,
      "step": 1321,
      "time_per_iteration": 3.5840232372283936
    },
    {
      "auxiliary_loss_clip": 0.01218653,
      "auxiliary_loss_mlp": 0.01044678,
      "balance_loss_clip": 1.06445539,
      "balance_loss_mlp": 1.03343678,
      "epoch": 0.15896110142487826,
      "flos": 23112359331840.0,
      "grad_norm": 2.7430607004755263,
      "language_loss": 0.81399989,
      "learning_rate": 3.8281876153631845e-06,
      "loss": 0.83663321,
      "num_input_tokens_seen": 27890070,
      "step": 1322,
      "time_per_iteration": 2.6522269248962402
    },
    {
      "auxiliary_loss_clip": 0.01163285,
      "auxiliary_loss_mlp": 0.01036927,
      "balance_loss_clip": 1.05560255,
      "balance_loss_mlp": 1.02592444,
      "epoch": 0.15908134431551735,
      "flos": 14685238632960.0,
      "grad_norm": 1.956180091575244,
      "language_loss": 0.65208936,
      "learning_rate": 3.827871601529416e-06,
      "loss": 0.67409152,
      "num_input_tokens_seen": 27908590,
      "step": 1323,
      "time_per_iteration": 2.762910842895508
    },
    {
      "auxiliary_loss_clip": 0.01180463,
      "auxiliary_loss_mlp": 0.01032917,
      "balance_loss_clip": 1.05753696,
      "balance_loss_mlp": 1.02264094,
      "epoch": 0.15920158720615643,
      "flos": 20193611984640.0,
      "grad_norm": 1.8267453161054032,
      "language_loss": 0.80638754,
      "learning_rate": 3.827555310408979e-06,
      "loss": 0.82852137,
      "num_input_tokens_seen": 27927985,
      "step": 1324,
      "time_per_iteration": 2.7244620323181152
    },
    {
      "auxiliary_loss_clip": 0.01183984,
      "auxiliary_loss_mlp": 0.01039571,
      "balance_loss_clip": 1.0630002,
      "balance_loss_mlp": 1.02829361,
      "epoch": 0.1593218300967955,
      "flos": 24826626892800.0,
      "grad_norm": 1.6246529764713744,
      "language_loss": 0.82331175,
      "learning_rate": 3.827238742049854e-06,
      "loss": 0.84554732,
      "num_input_tokens_seen": 27948280,
      "step": 1325,
      "time_per_iteration": 2.780841112136841
    },
    {
      "auxiliary_loss_clip": 0.01225537,
      "auxiliary_loss_mlp": 0.01038213,
      "balance_loss_clip": 1.06184673,
      "balance_loss_mlp": 1.02729344,
      "epoch": 0.15944207298743462,
      "flos": 28328707111680.0,
      "grad_norm": 1.8793773974062478,
      "language_loss": 0.52288342,
      "learning_rate": 3.826921896500066e-06,
      "loss": 0.5455209,
      "num_input_tokens_seen": 27969565,
      "step": 1326,
      "time_per_iteration": 2.6798951625823975
    },
    {
      "auxiliary_loss_clip": 0.01188077,
      "auxiliary_loss_mlp": 0.01039111,
      "balance_loss_clip": 1.06079531,
      "balance_loss_mlp": 1.02833509,
      "epoch": 0.1595623158780737,
      "flos": 22964838174720.0,
      "grad_norm": 1.8781466863466236,
      "language_loss": 0.78703606,
      "learning_rate": 3.826604773807678e-06,
      "loss": 0.80930793,
      "num_input_tokens_seen": 27987540,
      "step": 1327,
      "time_per_iteration": 2.744772434234619
    },
    {
      "auxiliary_loss_clip": 0.01199564,
      "auxiliary_loss_mlp": 0.01044591,
      "balance_loss_clip": 1.05766582,
      "balance_loss_mlp": 1.03252089,
      "epoch": 0.1596825587687128,
      "flos": 19710540950400.0,
      "grad_norm": 2.809090677346955,
      "language_loss": 0.73693991,
      "learning_rate": 3.826287374020798e-06,
      "loss": 0.75938141,
      "num_input_tokens_seen": 28002345,
      "step": 1328,
      "time_per_iteration": 2.6480305194854736
    },
    {
      "auxiliary_loss_clip": 0.01230119,
      "auxiliary_loss_mlp": 0.01039869,
      "balance_loss_clip": 1.06746447,
      "balance_loss_mlp": 1.02952111,
      "epoch": 0.1598028016593519,
      "flos": 22637727993600.0,
      "grad_norm": 2.0742878498039206,
      "language_loss": 0.82239622,
      "learning_rate": 3.825969697187575e-06,
      "loss": 0.84509611,
      "num_input_tokens_seen": 28021675,
      "step": 1329,
      "time_per_iteration": 2.676424264907837
    },
    {
      "auxiliary_loss_clip": 0.0118076,
      "auxiliary_loss_mlp": 0.0103474,
      "balance_loss_clip": 1.05774117,
      "balance_loss_mlp": 1.02410066,
      "epoch": 0.15992304454999098,
      "flos": 20482908122880.0,
      "grad_norm": 1.9812020783085575,
      "language_loss": 0.69682431,
      "learning_rate": 3.8256517433562015e-06,
      "loss": 0.71897936,
      "num_input_tokens_seen": 28039615,
      "step": 1330,
      "time_per_iteration": 2.699894666671753
    },
    {
      "auxiliary_loss_clip": 0.01230662,
      "auxiliary_loss_mlp": 0.01028853,
      "balance_loss_clip": 1.06663311,
      "balance_loss_mlp": 1.0188452,
      "epoch": 0.16004328744063007,
      "flos": 17676094533120.0,
      "grad_norm": 2.5596096766486403,
      "language_loss": 0.91969979,
      "learning_rate": 3.82533351257491e-06,
      "loss": 0.94229496,
      "num_input_tokens_seen": 28057565,
      "step": 1331,
      "time_per_iteration": 2.6839871406555176
    },
    {
      "auxiliary_loss_clip": 0.01207509,
      "auxiliary_loss_mlp": 0.01044141,
      "balance_loss_clip": 1.06224251,
      "balance_loss_mlp": 1.03287578,
      "epoch": 0.16016353033126918,
      "flos": 24098717779200.0,
      "grad_norm": 1.8189157520705472,
      "language_loss": 0.88406497,
      "learning_rate": 3.825015004891975e-06,
      "loss": 0.90658146,
      "num_input_tokens_seen": 28076305,
      "step": 1332,
      "time_per_iteration": 2.6576993465423584
    },
    {
      "auxiliary_loss_clip": 0.01204506,
      "auxiliary_loss_mlp": 0.01038365,
      "balance_loss_clip": 1.05759454,
      "balance_loss_mlp": 1.02812505,
      "epoch": 0.16028377322190826,
      "flos": 27634841112960.0,
      "grad_norm": 1.9436023243649394,
      "language_loss": 0.75770265,
      "learning_rate": 3.824696220355716e-06,
      "loss": 0.78013134,
      "num_input_tokens_seen": 28097895,
      "step": 1333,
      "time_per_iteration": 2.79388427734375
    },
    {
      "auxiliary_loss_clip": 0.01190389,
      "auxiliary_loss_mlp": 0.01036521,
      "balance_loss_clip": 1.05783224,
      "balance_loss_mlp": 1.0257442,
      "epoch": 0.16040401611254734,
      "flos": 20961202648320.0,
      "grad_norm": 31.501914570555847,
      "language_loss": 0.78889006,
      "learning_rate": 3.824377159014491e-06,
      "loss": 0.81115913,
      "num_input_tokens_seen": 28118790,
      "step": 1334,
      "time_per_iteration": 2.783365249633789
    },
    {
      "auxiliary_loss_clip": 0.01211275,
      "auxiliary_loss_mlp": 0.01037805,
      "balance_loss_clip": 1.06316888,
      "balance_loss_mlp": 1.0275588,
      "epoch": 0.16052425900318643,
      "flos": 21247051080960.0,
      "grad_norm": 2.067934597032914,
      "language_loss": 0.8481943,
      "learning_rate": 3.824057820916702e-06,
      "loss": 0.8706851,
      "num_input_tokens_seen": 28135995,
      "step": 1335,
      "time_per_iteration": 2.6405575275421143
    },
    {
      "auxiliary_loss_clip": 0.0119952,
      "auxiliary_loss_mlp": 0.01036175,
      "balance_loss_clip": 1.0619843,
      "balance_loss_mlp": 1.02478504,
      "epoch": 0.16064450189382554,
      "flos": 15524004096000.0,
      "grad_norm": 2.683351743257417,
      "language_loss": 0.71455777,
      "learning_rate": 3.8237382061107904e-06,
      "loss": 0.73691463,
      "num_input_tokens_seen": 28152715,
      "step": 1336,
      "time_per_iteration": 2.6827914714813232
    },
    {
      "auxiliary_loss_clip": 0.01125225,
      "auxiliary_loss_mlp": 0.01029131,
      "balance_loss_clip": 1.04608703,
      "balance_loss_mlp": 1.01877201,
      "epoch": 0.16076474478446462,
      "flos": 21178497974400.0,
      "grad_norm": 2.068463885737031,
      "language_loss": 0.78623801,
      "learning_rate": 3.823418314645243e-06,
      "loss": 0.80778158,
      "num_input_tokens_seen": 28171590,
      "step": 1337,
      "time_per_iteration": 2.7938859462738037
    },
    {
      "auxiliary_loss_clip": 0.0115092,
      "auxiliary_loss_mlp": 0.01033187,
      "balance_loss_clip": 1.05494809,
      "balance_loss_mlp": 1.0230726,
      "epoch": 0.1608849876751037,
      "flos": 18366476912640.0,
      "grad_norm": 2.014169017637328,
      "language_loss": 0.75178897,
      "learning_rate": 3.823098146568588e-06,
      "loss": 0.77363002,
      "num_input_tokens_seen": 28191295,
      "step": 1338,
      "time_per_iteration": 2.820744276046753
    },
    {
      "auxiliary_loss_clip": 0.01213494,
      "auxiliary_loss_mlp": 0.0103258,
      "balance_loss_clip": 1.06439304,
      "balance_loss_mlp": 1.0229001,
      "epoch": 0.1610052305657428,
      "flos": 29497024880640.0,
      "grad_norm": 1.7866710493257718,
      "language_loss": 0.71435857,
      "learning_rate": 3.822777701929394e-06,
      "loss": 0.73681933,
      "num_input_tokens_seen": 28213120,
      "step": 1339,
      "time_per_iteration": 2.7570526599884033
    },
    {
      "auxiliary_loss_clip": 0.01197902,
      "auxiliary_loss_mlp": 0.01039036,
      "balance_loss_clip": 1.05649436,
      "balance_loss_mlp": 1.02873075,
      "epoch": 0.1611254734563819,
      "flos": 26797871329920.0,
      "grad_norm": 1.981556774085316,
      "language_loss": 0.73508179,
      "learning_rate": 3.8224569807762714e-06,
      "loss": 0.75745118,
      "num_input_tokens_seen": 28232440,
      "step": 1340,
      "time_per_iteration": 2.7074010372161865
    },
    {
      "auxiliary_loss_clip": 0.01143583,
      "auxiliary_loss_mlp": 0.01039269,
      "balance_loss_clip": 1.04812956,
      "balance_loss_mlp": 1.02887416,
      "epoch": 0.16124571634702098,
      "flos": 22419570741120.0,
      "grad_norm": 1.9075958486341633,
      "language_loss": 0.76627779,
      "learning_rate": 3.822135983157873e-06,
      "loss": 0.78810632,
      "num_input_tokens_seen": 28251715,
      "step": 1341,
      "time_per_iteration": 2.7287495136260986
    },
    {
      "auxiliary_loss_clip": 0.01220929,
      "auxiliary_loss_mlp": 0.00764125,
      "balance_loss_clip": 1.06079459,
      "balance_loss_mlp": 1.00042868,
      "epoch": 0.16136595923766006,
      "flos": 10999116103680.0,
      "grad_norm": 2.154973411354358,
      "language_loss": 0.84887254,
      "learning_rate": 3.821814709122896e-06,
      "loss": 0.86872303,
      "num_input_tokens_seen": 28269765,
      "step": 1342,
      "time_per_iteration": 3.506011486053467
    },
    {
      "auxiliary_loss_clip": 0.01194504,
      "auxiliary_loss_mlp": 0.01033668,
      "balance_loss_clip": 1.0599649,
      "balance_loss_mlp": 1.02363086,
      "epoch": 0.16148620212829917,
      "flos": 21214983214080.0,
      "grad_norm": 2.349953692256193,
      "language_loss": 0.84976244,
      "learning_rate": 3.821493158720076e-06,
      "loss": 0.87204409,
      "num_input_tokens_seen": 28288870,
      "step": 1343,
      "time_per_iteration": 2.7098159790039062
    },
    {
      "auxiliary_loss_clip": 0.01179729,
      "auxiliary_loss_mlp": 0.01041791,
      "balance_loss_clip": 1.05304956,
      "balance_loss_mlp": 1.03038824,
      "epoch": 0.16160644501893826,
      "flos": 16758468760320.0,
      "grad_norm": 15.718647329379587,
      "language_loss": 0.73935485,
      "learning_rate": 3.821171331998191e-06,
      "loss": 0.7615701,
      "num_input_tokens_seen": 28305400,
      "step": 1344,
      "time_per_iteration": 2.6615591049194336
    },
    {
      "auxiliary_loss_clip": 0.01101678,
      "auxiliary_loss_mlp": 0.01007389,
      "balance_loss_clip": 1.03582716,
      "balance_loss_mlp": 1.00433719,
      "epoch": 0.16172668790957734,
      "flos": 64444967308800.0,
      "grad_norm": 0.706792882770157,
      "language_loss": 0.54470193,
      "learning_rate": 3.820849229006064e-06,
      "loss": 0.56579262,
      "num_input_tokens_seen": 28373150,
      "step": 1345,
      "time_per_iteration": 3.4247233867645264
    },
    {
      "auxiliary_loss_clip": 0.01226448,
      "auxiliary_loss_mlp": 0.01043442,
      "balance_loss_clip": 1.0619278,
      "balance_loss_mlp": 1.03297579,
      "epoch": 0.16184693080021645,
      "flos": 23257689759360.0,
      "grad_norm": 2.0318291914116298,
      "language_loss": 0.71435952,
      "learning_rate": 3.8205268497925564e-06,
      "loss": 0.7370584,
      "num_input_tokens_seen": 28393620,
      "step": 1346,
      "time_per_iteration": 3.6226305961608887
    },
    {
      "auxiliary_loss_clip": 0.01223749,
      "auxiliary_loss_mlp": 0.01034974,
      "balance_loss_clip": 1.06129885,
      "balance_loss_mlp": 1.0244782,
      "epoch": 0.16196717369085553,
      "flos": 17451113696640.0,
      "grad_norm": 2.146532717180153,
      "language_loss": 0.78589976,
      "learning_rate": 3.8202041944065725e-06,
      "loss": 0.80848694,
      "num_input_tokens_seen": 28409440,
      "step": 1347,
      "time_per_iteration": 4.44841742515564
    },
    {
      "auxiliary_loss_clip": 0.01226323,
      "auxiliary_loss_mlp": 0.01034404,
      "balance_loss_clip": 1.06486773,
      "balance_loss_mlp": 1.02392006,
      "epoch": 0.16208741658149461,
      "flos": 23873377806720.0,
      "grad_norm": 2.059042334785779,
      "language_loss": 0.74059761,
      "learning_rate": 3.819881262897061e-06,
      "loss": 0.76320493,
      "num_input_tokens_seen": 28427575,
      "step": 1348,
      "time_per_iteration": 2.6366240978240967
    },
    {
      "auxiliary_loss_clip": 0.01184674,
      "auxiliary_loss_mlp": 0.01036192,
      "balance_loss_clip": 1.06253815,
      "balance_loss_mlp": 1.02457488,
      "epoch": 0.1622076594721337,
      "flos": 25884806584320.0,
      "grad_norm": 2.082306287602304,
      "language_loss": 0.73841178,
      "learning_rate": 3.819558055313008e-06,
      "loss": 0.76062042,
      "num_input_tokens_seen": 28448260,
      "step": 1349,
      "time_per_iteration": 2.7356936931610107
    },
    {
      "auxiliary_loss_clip": 0.01217748,
      "auxiliary_loss_mlp": 0.01044163,
      "balance_loss_clip": 1.06385183,
      "balance_loss_mlp": 1.03375554,
      "epoch": 0.1623279023627728,
      "flos": 21539759011200.0,
      "grad_norm": 2.370254777971957,
      "language_loss": 0.77185345,
      "learning_rate": 3.819234571703444e-06,
      "loss": 0.79447252,
      "num_input_tokens_seen": 28467085,
      "step": 1350,
      "time_per_iteration": 2.6230547428131104
    },
    {
      "auxiliary_loss_clip": 0.01204067,
      "auxiliary_loss_mlp": 0.01041708,
      "balance_loss_clip": 1.05969977,
      "balance_loss_mlp": 1.03024626,
      "epoch": 0.1624481452534119,
      "flos": 22085421494400.0,
      "grad_norm": 1.90585672177158,
      "language_loss": 0.85788441,
      "learning_rate": 3.8189108121174435e-06,
      "loss": 0.88034213,
      "num_input_tokens_seen": 28486850,
      "step": 1351,
      "time_per_iteration": 2.773103713989258
    },
    {
      "auxiliary_loss_clip": 0.01180693,
      "auxiliary_loss_mlp": 0.01042905,
      "balance_loss_clip": 1.06347191,
      "balance_loss_mlp": 1.03198504,
      "epoch": 0.16256838814405097,
      "flos": 27087490690560.0,
      "grad_norm": 1.6630190077934404,
      "language_loss": 0.83297193,
      "learning_rate": 3.818586776604118e-06,
      "loss": 0.85520798,
      "num_input_tokens_seen": 28507490,
      "step": 1352,
      "time_per_iteration": 2.72802996635437
    },
    {
      "auxiliary_loss_clip": 0.01189281,
      "auxiliary_loss_mlp": 0.01034192,
      "balance_loss_clip": 1.05770707,
      "balance_loss_mlp": 1.02335596,
      "epoch": 0.16268863103469008,
      "flos": 20120354196480.0,
      "grad_norm": 2.3245025433671307,
      "language_loss": 0.61517918,
      "learning_rate": 3.818262465212625e-06,
      "loss": 0.63741386,
      "num_input_tokens_seen": 28527615,
      "step": 1353,
      "time_per_iteration": 2.6996500492095947
    },
    {
      "auxiliary_loss_clip": 0.01203111,
      "auxiliary_loss_mlp": 0.01037914,
      "balance_loss_clip": 1.06055665,
      "balance_loss_mlp": 1.02632642,
      "epoch": 0.16280887392532917,
      "flos": 18332792933760.0,
      "grad_norm": 2.2323350688685633,
      "language_loss": 0.77407175,
      "learning_rate": 3.817937877992161e-06,
      "loss": 0.79648197,
      "num_input_tokens_seen": 28544910,
      "step": 1354,
      "time_per_iteration": 2.609583854675293
    },
    {
      "auxiliary_loss_clip": 0.01181257,
      "auxiliary_loss_mlp": 0.00765538,
      "balance_loss_clip": 1.05624402,
      "balance_loss_mlp": 1.00053239,
      "epoch": 0.16292911681596825,
      "flos": 11874330892800.0,
      "grad_norm": 2.4836465973822905,
      "language_loss": 0.85998744,
      "learning_rate": 3.817613014991967e-06,
      "loss": 0.87945533,
      "num_input_tokens_seen": 28561050,
      "step": 1355,
      "time_per_iteration": 2.723642110824585
    },
    {
      "auxiliary_loss_clip": 0.01169495,
      "auxiliary_loss_mlp": 0.01038797,
      "balance_loss_clip": 1.05239177,
      "balance_loss_mlp": 1.02794862,
      "epoch": 0.16304935970660733,
      "flos": 26103466627200.0,
      "grad_norm": 2.0678664328136875,
      "language_loss": 0.76728565,
      "learning_rate": 3.817287876261323e-06,
      "loss": 0.78936857,
      "num_input_tokens_seen": 28581385,
      "step": 1356,
      "time_per_iteration": 2.7060506343841553
    },
    {
      "auxiliary_loss_clip": 0.01196749,
      "auxiliary_loss_mlp": 0.0103927,
      "balance_loss_clip": 1.06272793,
      "balance_loss_mlp": 1.02831519,
      "epoch": 0.16316960259724644,
      "flos": 29351945848320.0,
      "grad_norm": 1.8330933380018286,
      "language_loss": 0.80030483,
      "learning_rate": 3.816962461849553e-06,
      "loss": 0.82266498,
      "num_input_tokens_seen": 28603255,
      "step": 1357,
      "time_per_iteration": 2.7294507026672363
    },
    {
      "auxiliary_loss_clip": 0.01189794,
      "auxiliary_loss_mlp": 0.01040556,
      "balance_loss_clip": 1.06186652,
      "balance_loss_mlp": 1.02989244,
      "epoch": 0.16328984548788553,
      "flos": 20886759711360.0,
      "grad_norm": 1.914936601577789,
      "language_loss": 0.84594107,
      "learning_rate": 3.8166367718060235e-06,
      "loss": 0.86824459,
      "num_input_tokens_seen": 28623145,
      "step": 1358,
      "time_per_iteration": 2.6864376068115234
    },
    {
      "auxiliary_loss_clip": 0.01208589,
      "auxiliary_loss_mlp": 0.01033362,
      "balance_loss_clip": 1.06032491,
      "balance_loss_mlp": 1.02280593,
      "epoch": 0.1634100883785246,
      "flos": 18041090584320.0,
      "grad_norm": 3.044225928149676,
      "language_loss": 0.76611805,
      "learning_rate": 3.816310806180139e-06,
      "loss": 0.7885375,
      "num_input_tokens_seen": 28641555,
      "step": 1359,
      "time_per_iteration": 2.7074708938598633
    },
    {
      "auxiliary_loss_clip": 0.01197176,
      "auxiliary_loss_mlp": 0.01040138,
      "balance_loss_clip": 1.06260788,
      "balance_loss_mlp": 1.02882504,
      "epoch": 0.16353033126916372,
      "flos": 24572128055040.0,
      "grad_norm": 1.7475665467686825,
      "language_loss": 0.80799806,
      "learning_rate": 3.81598456502135e-06,
      "loss": 0.83037126,
      "num_input_tokens_seen": 28661575,
      "step": 1360,
      "time_per_iteration": 2.709463119506836
    },
    {
      "auxiliary_loss_clip": 0.01194337,
      "auxiliary_loss_mlp": 0.0103987,
      "balance_loss_clip": 1.06141019,
      "balance_loss_mlp": 1.02937388,
      "epoch": 0.1636505741598028,
      "flos": 19892895321600.0,
      "grad_norm": 1.990796875681064,
      "language_loss": 0.87152076,
      "learning_rate": 3.8156580483791455e-06,
      "loss": 0.89386284,
      "num_input_tokens_seen": 28676765,
      "step": 1361,
      "time_per_iteration": 2.6877455711364746
    },
    {
      "auxiliary_loss_clip": 0.01226143,
      "auxiliary_loss_mlp": 0.01038239,
      "balance_loss_clip": 1.06250465,
      "balance_loss_mlp": 1.02760518,
      "epoch": 0.16377081705044189,
      "flos": 28402611344640.0,
      "grad_norm": 2.368438712233047,
      "language_loss": 0.77093369,
      "learning_rate": 3.815331256303059e-06,
      "loss": 0.79357755,
      "num_input_tokens_seen": 28696795,
      "step": 1362,
      "time_per_iteration": 2.694855213165283
    },
    {
      "auxiliary_loss_clip": 0.01183394,
      "auxiliary_loss_mlp": 0.01037893,
      "balance_loss_clip": 1.06333888,
      "balance_loss_mlp": 1.0266571,
      "epoch": 0.163891059941081,
      "flos": 21908059113600.0,
      "grad_norm": 2.3845480664722722,
      "language_loss": 0.77255613,
      "learning_rate": 3.815004188842665e-06,
      "loss": 0.79476899,
      "num_input_tokens_seen": 28714835,
      "step": 1363,
      "time_per_iteration": 2.7188937664031982
    },
    {
      "auxiliary_loss_clip": 0.01188665,
      "auxiliary_loss_mlp": 0.01039981,
      "balance_loss_clip": 1.05492926,
      "balance_loss_mlp": 1.03021169,
      "epoch": 0.16401130283172008,
      "flos": 26797619934720.0,
      "grad_norm": 1.6379327618332375,
      "language_loss": 0.79727268,
      "learning_rate": 3.814676846047578e-06,
      "loss": 0.81955922,
      "num_input_tokens_seen": 28735710,
      "step": 1364,
      "time_per_iteration": 2.7413017749786377
    },
    {
      "auxiliary_loss_clip": 0.0120874,
      "auxiliary_loss_mlp": 0.01040139,
      "balance_loss_clip": 1.05991864,
      "balance_loss_mlp": 1.03023314,
      "epoch": 0.16413154572235916,
      "flos": 32997417160320.0,
      "grad_norm": 1.9972154235236679,
      "language_loss": 0.69828516,
      "learning_rate": 3.8143492279674565e-06,
      "loss": 0.72077399,
      "num_input_tokens_seen": 28758405,
      "step": 1365,
      "time_per_iteration": 2.6900131702423096
    },
    {
      "auxiliary_loss_clip": 0.01097543,
      "auxiliary_loss_mlp": 0.01009562,
      "balance_loss_clip": 1.03295505,
      "balance_loss_mlp": 1.00664163,
      "epoch": 0.16425178861299825,
      "flos": 40113622074240.0,
      "grad_norm": 0.8422414866576833,
      "language_loss": 0.58454561,
      "learning_rate": 3.8140213346519997e-06,
      "loss": 0.60561657,
      "num_input_tokens_seen": 28809000,
      "step": 1366,
      "time_per_iteration": 3.0356743335723877
    },
    {
      "auxiliary_loss_clip": 0.01173608,
      "auxiliary_loss_mlp": 0.01046082,
      "balance_loss_clip": 1.05760479,
      "balance_loss_mlp": 1.03444719,
      "epoch": 0.16437203150363736,
      "flos": 25447486498560.0,
      "grad_norm": 1.890849893652848,
      "language_loss": 0.76837242,
      "learning_rate": 3.813693166150948e-06,
      "loss": 0.79056931,
      "num_input_tokens_seen": 28829210,
      "step": 1367,
      "time_per_iteration": 2.765251636505127
    },
    {
      "auxiliary_loss_clip": 0.01174565,
      "auxiliary_loss_mlp": 0.01036497,
      "balance_loss_clip": 1.05951202,
      "balance_loss_mlp": 1.02507663,
      "epoch": 0.16449227439427644,
      "flos": 23476888506240.0,
      "grad_norm": 4.715004523600085,
      "language_loss": 0.8515414,
      "learning_rate": 3.813364722514086e-06,
      "loss": 0.87365198,
      "num_input_tokens_seen": 28847545,
      "step": 1368,
      "time_per_iteration": 3.6013832092285156
    },
    {
      "auxiliary_loss_clip": 0.01206977,
      "auxiliary_loss_mlp": 0.010446,
      "balance_loss_clip": 1.05973005,
      "balance_loss_mlp": 1.0339129,
      "epoch": 0.16461251728491552,
      "flos": 13545217802880.0,
      "grad_norm": 3.1992866244109663,
      "language_loss": 0.80753827,
      "learning_rate": 3.8130360037912368e-06,
      "loss": 0.83005404,
      "num_input_tokens_seen": 28863990,
      "step": 1369,
      "time_per_iteration": 2.659313678741455
    },
    {
      "auxiliary_loss_clip": 0.01210093,
      "auxiliary_loss_mlp": 0.01037859,
      "balance_loss_clip": 1.05998945,
      "balance_loss_mlp": 1.02702844,
      "epoch": 0.16473276017555463,
      "flos": 23003298662400.0,
      "grad_norm": 2.129268998164014,
      "language_loss": 0.81847328,
      "learning_rate": 3.812707010032268e-06,
      "loss": 0.84095275,
      "num_input_tokens_seen": 28883045,
      "step": 1370,
      "time_per_iteration": 2.676392078399658
    },
    {
      "auxiliary_loss_clip": 0.01218283,
      "auxiliary_loss_mlp": 0.0104467,
      "balance_loss_clip": 1.06622577,
      "balance_loss_mlp": 1.03403068,
      "epoch": 0.16485300306619372,
      "flos": 24790680357120.0,
      "grad_norm": 1.8149263814636039,
      "language_loss": 0.79139954,
      "learning_rate": 3.8123777412870863e-06,
      "loss": 0.8140291,
      "num_input_tokens_seen": 28902545,
      "step": 1371,
      "time_per_iteration": 2.6979081630706787
    },
    {
      "auxiliary_loss_clip": 0.01198602,
      "auxiliary_loss_mlp": 0.01034567,
      "balance_loss_clip": 1.0597682,
      "balance_loss_mlp": 1.02418923,
      "epoch": 0.1649732459568328,
      "flos": 21106497162240.0,
      "grad_norm": 2.141230323830424,
      "language_loss": 0.78432405,
      "learning_rate": 3.812048197605643e-06,
      "loss": 0.80665576,
      "num_input_tokens_seen": 28921440,
      "step": 1372,
      "time_per_iteration": 3.5658137798309326
    },
    {
      "auxiliary_loss_clip": 0.0121085,
      "auxiliary_loss_mlp": 0.01034353,
      "balance_loss_clip": 1.06176257,
      "balance_loss_mlp": 1.02276599,
      "epoch": 0.16509348884747188,
      "flos": 20266726118400.0,
      "grad_norm": 2.0091620776325665,
      "language_loss": 0.81723845,
      "learning_rate": 3.8117183790379277e-06,
      "loss": 0.83969051,
      "num_input_tokens_seen": 28939890,
      "step": 1373,
      "time_per_iteration": 4.375774621963501
    },
    {
      "auxiliary_loss_clip": 0.01227585,
      "auxiliary_loss_mlp": 0.01033683,
      "balance_loss_clip": 1.06346107,
      "balance_loss_mlp": 1.02310896,
      "epoch": 0.165213731738111,
      "flos": 11035493602560.0,
      "grad_norm": 3.3396092069705907,
      "language_loss": 0.93681508,
      "learning_rate": 3.811388285633976e-06,
      "loss": 0.95942777,
      "num_input_tokens_seen": 28955875,
      "step": 1374,
      "time_per_iteration": 2.5478992462158203
    },
    {
      "auxiliary_loss_clip": 0.01172749,
      "auxiliary_loss_mlp": 0.01035772,
      "balance_loss_clip": 1.05834711,
      "balance_loss_mlp": 1.02483439,
      "epoch": 0.16533397462875007,
      "flos": 29972051268480.0,
      "grad_norm": 2.28068910450057,
      "language_loss": 0.62023109,
      "learning_rate": 3.811057917443861e-06,
      "loss": 0.64231622,
      "num_input_tokens_seen": 28975140,
      "step": 1375,
      "time_per_iteration": 2.7869110107421875
    },
    {
      "auxiliary_loss_clip": 0.01109085,
      "auxiliary_loss_mlp": 0.01005692,
      "balance_loss_clip": 1.03020215,
      "balance_loss_mlp": 1.00261676,
      "epoch": 0.16545421751938916,
      "flos": 65556763027200.0,
      "grad_norm": 0.8569516029847498,
      "language_loss": 0.6825732,
      "learning_rate": 3.8107272745177e-06,
      "loss": 0.70372093,
      "num_input_tokens_seen": 29047470,
      "step": 1376,
      "time_per_iteration": 3.3500595092773438
    },
    {
      "auxiliary_loss_clip": 0.01185596,
      "auxiliary_loss_mlp": 0.010356,
      "balance_loss_clip": 1.06089938,
      "balance_loss_mlp": 1.02520537,
      "epoch": 0.16557446041002827,
      "flos": 22492361652480.0,
      "grad_norm": 1.8329689627213155,
      "language_loss": 0.78643268,
      "learning_rate": 3.8103963569056513e-06,
      "loss": 0.80864465,
      "num_input_tokens_seen": 29066605,
      "step": 1377,
      "time_per_iteration": 2.7369496822357178
    },
    {
      "auxiliary_loss_clip": 0.01187069,
      "auxiliary_loss_mlp": 0.01035902,
      "balance_loss_clip": 1.05401182,
      "balance_loss_mlp": 1.02522111,
      "epoch": 0.16569470330066735,
      "flos": 24602723464320.0,
      "grad_norm": 1.644939081810944,
      "language_loss": 0.88104236,
      "learning_rate": 3.8100651646579146e-06,
      "loss": 0.90327215,
      "num_input_tokens_seen": 29085815,
      "step": 1378,
      "time_per_iteration": 2.6735615730285645
    },
    {
      "auxiliary_loss_clip": 0.01187949,
      "auxiliary_loss_mlp": 0.01039061,
      "balance_loss_clip": 1.05586076,
      "balance_loss_mlp": 1.02861261,
      "epoch": 0.16581494619130643,
      "flos": 15006207588480.0,
      "grad_norm": 2.1568701111289914,
      "language_loss": 0.92811346,
      "learning_rate": 3.8097336978247317e-06,
      "loss": 0.95038354,
      "num_input_tokens_seen": 29102520,
      "step": 1379,
      "time_per_iteration": 2.6430978775024414
    },
    {
      "auxiliary_loss_clip": 0.01182745,
      "auxiliary_loss_mlp": 0.01041074,
      "balance_loss_clip": 1.05559313,
      "balance_loss_mlp": 1.0301249,
      "epoch": 0.16593518908194552,
      "flos": 17420338719360.0,
      "grad_norm": 2.2985836695641986,
      "language_loss": 0.88887179,
      "learning_rate": 3.8094019564563854e-06,
      "loss": 0.91110992,
      "num_input_tokens_seen": 29119450,
      "step": 1380,
      "time_per_iteration": 2.612462282180786
    },
    {
      "auxiliary_loss_clip": 0.01225119,
      "auxiliary_loss_mlp": 0.00764697,
      "balance_loss_clip": 1.06286478,
      "balance_loss_mlp": 1.00052512,
      "epoch": 0.16605543197258463,
      "flos": 20412631163520.0,
      "grad_norm": 2.99107971737057,
      "language_loss": 0.75116527,
      "learning_rate": 3.809069940603201e-06,
      "loss": 0.77106339,
      "num_input_tokens_seen": 29137405,
      "step": 1381,
      "time_per_iteration": 2.7121667861938477
    },
    {
      "auxiliary_loss_clip": 0.01184576,
      "auxiliary_loss_mlp": 0.0103806,
      "balance_loss_clip": 1.05897403,
      "balance_loss_mlp": 1.02675259,
      "epoch": 0.1661756748632237,
      "flos": 14209745368320.0,
      "grad_norm": 2.0515317504695503,
      "language_loss": 0.77973926,
      "learning_rate": 3.8087376503155452e-06,
      "loss": 0.80196559,
      "num_input_tokens_seen": 29154890,
      "step": 1382,
      "time_per_iteration": 2.606696128845215
    },
    {
      "auxiliary_loss_clip": 0.01101277,
      "auxiliary_loss_mlp": 0.01003077,
      "balance_loss_clip": 1.02749765,
      "balance_loss_mlp": 1.00032294,
      "epoch": 0.1662959177538628,
      "flos": 66080877350400.0,
      "grad_norm": 0.8929302535826479,
      "language_loss": 0.56240308,
      "learning_rate": 3.808405085643826e-06,
      "loss": 0.58344662,
      "num_input_tokens_seen": 29219770,
      "step": 1383,
      "time_per_iteration": 3.281708002090454
    },
    {
      "auxiliary_loss_clip": 0.01227783,
      "auxiliary_loss_mlp": 0.00765341,
      "balance_loss_clip": 1.06477785,
      "balance_loss_mlp": 1.00056577,
      "epoch": 0.1664161606445019,
      "flos": 20740567357440.0,
      "grad_norm": 2.1881124610304603,
      "language_loss": 0.88889068,
      "learning_rate": 3.8080722466384925e-06,
      "loss": 0.90882194,
      "num_input_tokens_seen": 29237620,
      "step": 1384,
      "time_per_iteration": 2.6021363735198975
    },
    {
      "auxiliary_loss_clip": 0.01229455,
      "auxiliary_loss_mlp": 0.01035733,
      "balance_loss_clip": 1.06344676,
      "balance_loss_mlp": 1.02518904,
      "epoch": 0.166536403535141,
      "flos": 25260930236160.0,
      "grad_norm": 2.4441607083293886,
      "language_loss": 0.71020973,
      "learning_rate": 3.8077391333500376e-06,
      "loss": 0.73286164,
      "num_input_tokens_seen": 29256760,
      "step": 1385,
      "time_per_iteration": 2.7025511264801025
    },
    {
      "auxiliary_loss_clip": 0.01195492,
      "auxiliary_loss_mlp": 0.01037769,
      "balance_loss_clip": 1.0614599,
      "balance_loss_mlp": 1.02754676,
      "epoch": 0.16665664642578007,
      "flos": 25447450584960.0,
      "grad_norm": 1.765808766379165,
      "language_loss": 0.76668763,
      "learning_rate": 3.8074057458289934e-06,
      "loss": 0.78902024,
      "num_input_tokens_seen": 29277450,
      "step": 1386,
      "time_per_iteration": 2.6995954513549805
    },
    {
      "auxiliary_loss_clip": 0.01193856,
      "auxiliary_loss_mlp": 0.01037292,
      "balance_loss_clip": 1.05693769,
      "balance_loss_mlp": 1.02701044,
      "epoch": 0.16677688931641918,
      "flos": 22200767043840.0,
      "grad_norm": 1.9862111102295252,
      "language_loss": 0.82654405,
      "learning_rate": 3.807072084125934e-06,
      "loss": 0.8488555,
      "num_input_tokens_seen": 29299300,
      "step": 1387,
      "time_per_iteration": 2.691457748413086
    },
    {
      "auxiliary_loss_clip": 0.0119372,
      "auxiliary_loss_mlp": 0.01033055,
      "balance_loss_clip": 1.06194925,
      "balance_loss_mlp": 1.02223063,
      "epoch": 0.16689713220705826,
      "flos": 16945958776320.0,
      "grad_norm": 2.135181117493788,
      "language_loss": 0.80272245,
      "learning_rate": 3.806738148291477e-06,
      "loss": 0.82499015,
      "num_input_tokens_seen": 29316125,
      "step": 1388,
      "time_per_iteration": 2.6920456886291504
    },
    {
      "auxiliary_loss_clip": 0.01151089,
      "auxiliary_loss_mlp": 0.01040489,
      "balance_loss_clip": 1.05243015,
      "balance_loss_mlp": 1.02812672,
      "epoch": 0.16701737509769735,
      "flos": 36244423923840.0,
      "grad_norm": 2.0353639825343373,
      "language_loss": 0.71211946,
      "learning_rate": 3.8064039383762793e-06,
      "loss": 0.73403525,
      "num_input_tokens_seen": 29338490,
      "step": 1389,
      "time_per_iteration": 2.875511407852173
    },
    {
      "auxiliary_loss_clip": 0.01213918,
      "auxiliary_loss_mlp": 0.01038134,
      "balance_loss_clip": 1.06601572,
      "balance_loss_mlp": 1.02751803,
      "epoch": 0.16713761798833643,
      "flos": 23258659426560.0,
      "grad_norm": 3.6764840929757336,
      "language_loss": 0.76328599,
      "learning_rate": 3.8060694544310396e-06,
      "loss": 0.78580654,
      "num_input_tokens_seen": 29357000,
      "step": 1390,
      "time_per_iteration": 2.682415008544922
    },
    {
      "auxiliary_loss_clip": 0.01227578,
      "auxiliary_loss_mlp": 0.01041056,
      "balance_loss_clip": 1.06478631,
      "balance_loss_mlp": 1.0302732,
      "epoch": 0.16725786087897554,
      "flos": 25302515207040.0,
      "grad_norm": 1.8325435688495901,
      "language_loss": 0.7877388,
      "learning_rate": 3.8057346965065006e-06,
      "loss": 0.81042516,
      "num_input_tokens_seen": 29378230,
      "step": 1391,
      "time_per_iteration": 2.6407477855682373
    },
    {
      "auxiliary_loss_clip": 0.01197433,
      "auxiliary_loss_mlp": 0.01040505,
      "balance_loss_clip": 1.06410694,
      "balance_loss_mlp": 1.02963269,
      "epoch": 0.16737810376961462,
      "flos": 31831541516160.0,
      "grad_norm": 1.635873384155522,
      "language_loss": 0.84891963,
      "learning_rate": 3.805399664653443e-06,
      "loss": 0.87129903,
      "num_input_tokens_seen": 29400370,
      "step": 1392,
      "time_per_iteration": 2.836308717727661
    },
    {
      "auxiliary_loss_clip": 0.01227529,
      "auxiliary_loss_mlp": 0.01039183,
      "balance_loss_clip": 1.06360567,
      "balance_loss_mlp": 1.02760196,
      "epoch": 0.1674983466602537,
      "flos": 27961843553280.0,
      "grad_norm": 2.3096463813857113,
      "language_loss": 0.74055016,
      "learning_rate": 3.805064358922692e-06,
      "loss": 0.76321733,
      "num_input_tokens_seen": 29418660,
      "step": 1393,
      "time_per_iteration": 2.6994788646698
    },
    {
      "auxiliary_loss_clip": 0.01212539,
      "auxiliary_loss_mlp": 0.01034818,
      "balance_loss_clip": 1.06056798,
      "balance_loss_mlp": 1.02351737,
      "epoch": 0.16761858955089282,
      "flos": 21762656858880.0,
      "grad_norm": 3.2287345784707058,
      "language_loss": 0.81118667,
      "learning_rate": 3.8047287793651136e-06,
      "loss": 0.83366024,
      "num_input_tokens_seen": 29440105,
      "step": 1394,
      "time_per_iteration": 3.605530023574829
    },
    {
      "auxiliary_loss_clip": 0.01183658,
      "auxiliary_loss_mlp": 0.01036202,
      "balance_loss_clip": 1.06014323,
      "balance_loss_mlp": 1.02571142,
      "epoch": 0.1677388324415319,
      "flos": 23805507058560.0,
      "grad_norm": 1.7407172212455306,
      "language_loss": 0.88580894,
      "learning_rate": 3.8043929260316137e-06,
      "loss": 0.9080075,
      "num_input_tokens_seen": 29458260,
      "step": 1395,
      "time_per_iteration": 2.751116991043091
    },
    {
      "auxiliary_loss_clip": 0.01199278,
      "auxiliary_loss_mlp": 0.01032647,
      "balance_loss_clip": 1.06507802,
      "balance_loss_mlp": 1.02223444,
      "epoch": 0.16785907533217098,
      "flos": 20558859431040.0,
      "grad_norm": 1.9246889218890606,
      "language_loss": 0.83547258,
      "learning_rate": 3.8040567989731417e-06,
      "loss": 0.85779184,
      "num_input_tokens_seen": 29476205,
      "step": 1396,
      "time_per_iteration": 2.747361660003662
    },
    {
      "auxiliary_loss_clip": 0.01202938,
      "auxiliary_loss_mlp": 0.0103582,
      "balance_loss_clip": 1.06064296,
      "balance_loss_mlp": 1.02606869,
      "epoch": 0.16797931822281006,
      "flos": 15669657745920.0,
      "grad_norm": 3.2985040043908715,
      "language_loss": 0.79449439,
      "learning_rate": 3.8037203982406876e-06,
      "loss": 0.81688195,
      "num_input_tokens_seen": 29494370,
      "step": 1397,
      "time_per_iteration": 2.649172067642212
    },
    {
      "auxiliary_loss_clip": 0.01225304,
      "auxiliary_loss_mlp": 0.01038749,
      "balance_loss_clip": 1.06486607,
      "balance_loss_mlp": 1.02807415,
      "epoch": 0.16809956111344918,
      "flos": 16541101607040.0,
      "grad_norm": 1.7310485300913174,
      "language_loss": 0.73128581,
      "learning_rate": 3.8033837238852835e-06,
      "loss": 0.7539264,
      "num_input_tokens_seen": 29511070,
      "step": 1398,
      "time_per_iteration": 3.508976936340332
    },
    {
      "auxiliary_loss_clip": 0.01181665,
      "auxiliary_loss_mlp": 0.01039665,
      "balance_loss_clip": 1.05503464,
      "balance_loss_mlp": 1.02956247,
      "epoch": 0.16821980400408826,
      "flos": 23258084808960.0,
      "grad_norm": 1.9845298170789722,
      "language_loss": 0.69455779,
      "learning_rate": 3.8030467759580017e-06,
      "loss": 0.71677113,
      "num_input_tokens_seen": 29531990,
      "step": 1399,
      "time_per_iteration": 4.420074939727783
    },
    {
      "auxiliary_loss_clip": 0.01210636,
      "auxiliary_loss_mlp": 0.01043083,
      "balance_loss_clip": 1.05882168,
      "balance_loss_mlp": 1.03283656,
      "epoch": 0.16834004689472734,
      "flos": 20774754126720.0,
      "grad_norm": 2.074630287643699,
      "language_loss": 0.87269884,
      "learning_rate": 3.802709554509958e-06,
      "loss": 0.89523602,
      "num_input_tokens_seen": 29549790,
      "step": 1400,
      "time_per_iteration": 2.6797807216644287
    },
    {
      "auxiliary_loss_clip": 0.01190603,
      "auxiliary_loss_mlp": 0.01036893,
      "balance_loss_clip": 1.05478728,
      "balance_loss_mlp": 1.02605104,
      "epoch": 0.16846028978536645,
      "flos": 26687302289280.0,
      "grad_norm": 2.160290164775485,
      "language_loss": 0.79165792,
      "learning_rate": 3.8023720595923083e-06,
      "loss": 0.8139329,
      "num_input_tokens_seen": 29569045,
      "step": 1401,
      "time_per_iteration": 2.6895766258239746
    },
    {
      "auxiliary_loss_clip": 0.011608,
      "auxiliary_loss_mlp": 0.01038184,
      "balance_loss_clip": 1.05231881,
      "balance_loss_mlp": 1.02722836,
      "epoch": 0.16858053267600553,
      "flos": 18843298980480.0,
      "grad_norm": 2.6766640808265048,
      "language_loss": 0.87636095,
      "learning_rate": 3.80203429125625e-06,
      "loss": 0.89835083,
      "num_input_tokens_seen": 29587220,
      "step": 1402,
      "time_per_iteration": 2.731353759765625
    },
    {
      "auxiliary_loss_clip": 0.01142428,
      "auxiliary_loss_mlp": 0.0103628,
      "balance_loss_clip": 1.05385792,
      "balance_loss_mlp": 1.0267669,
      "epoch": 0.16870077556664462,
      "flos": 27744548227200.0,
      "grad_norm": 2.3306238265850325,
      "language_loss": 0.70078373,
      "learning_rate": 3.8016962495530225e-06,
      "loss": 0.72257084,
      "num_input_tokens_seen": 29606410,
      "step": 1403,
      "time_per_iteration": 2.8432488441467285
    },
    {
      "auxiliary_loss_clip": 0.0122589,
      "auxiliary_loss_mlp": 0.01042637,
      "balance_loss_clip": 1.06340051,
      "balance_loss_mlp": 1.03131199,
      "epoch": 0.1688210184572837,
      "flos": 13730768484480.0,
      "grad_norm": 2.9182645956867934,
      "language_loss": 0.76328236,
      "learning_rate": 3.8013579345339063e-06,
      "loss": 0.78596759,
      "num_input_tokens_seen": 29621275,
      "step": 1404,
      "time_per_iteration": 2.558044910430908
    },
    {
      "auxiliary_loss_clip": 0.0117894,
      "auxiliary_loss_mlp": 0.0103307,
      "balance_loss_clip": 1.05557072,
      "balance_loss_mlp": 1.02211475,
      "epoch": 0.1689412613479228,
      "flos": 26468785900800.0,
      "grad_norm": 2.401140848384728,
      "language_loss": 0.69371951,
      "learning_rate": 3.801019346250224e-06,
      "loss": 0.71583956,
      "num_input_tokens_seen": 29641420,
      "step": 1405,
      "time_per_iteration": 2.7142367362976074
    },
    {
      "auxiliary_loss_clip": 0.01208297,
      "auxiliary_loss_mlp": 0.01036994,
      "balance_loss_clip": 1.06167889,
      "balance_loss_mlp": 1.0270822,
      "epoch": 0.1690615042385619,
      "flos": 21138852337920.0,
      "grad_norm": 2.400461623334689,
      "language_loss": 0.84560919,
      "learning_rate": 3.8006804847533395e-06,
      "loss": 0.86806208,
      "num_input_tokens_seen": 29660935,
      "step": 1406,
      "time_per_iteration": 2.641300916671753
    },
    {
      "auxiliary_loss_clip": 0.01221457,
      "auxiliary_loss_mlp": 0.01034935,
      "balance_loss_clip": 1.06121802,
      "balance_loss_mlp": 1.02442074,
      "epoch": 0.16918174712920098,
      "flos": 20849340718080.0,
      "grad_norm": 1.8681371461371108,
      "language_loss": 0.85504961,
      "learning_rate": 3.8003413500946556e-06,
      "loss": 0.87761354,
      "num_input_tokens_seen": 29681045,
      "step": 1407,
      "time_per_iteration": 2.667349338531494
    },
    {
      "auxiliary_loss_clip": 0.01193893,
      "auxiliary_loss_mlp": 0.01033331,
      "balance_loss_clip": 1.05993736,
      "balance_loss_mlp": 1.02290034,
      "epoch": 0.1693019900198401,
      "flos": 16983270028800.0,
      "grad_norm": 2.6274459154397167,
      "language_loss": 0.8316946,
      "learning_rate": 3.8000019423256216e-06,
      "loss": 0.85396683,
      "num_input_tokens_seen": 29698810,
      "step": 1408,
      "time_per_iteration": 2.668830156326294
    },
    {
      "auxiliary_loss_clip": 0.01180333,
      "auxiliary_loss_mlp": 0.01036578,
      "balance_loss_clip": 1.05659294,
      "balance_loss_mlp": 1.02693415,
      "epoch": 0.16942223291047917,
      "flos": 26796901662720.0,
      "grad_norm": 1.843323211139918,
      "language_loss": 0.88209116,
      "learning_rate": 3.7996622614977234e-06,
      "loss": 0.90426028,
      "num_input_tokens_seen": 29720000,
      "step": 1409,
      "time_per_iteration": 2.7244889736175537
    },
    {
      "auxiliary_loss_clip": 0.01194381,
      "auxiliary_loss_mlp": 0.0104115,
      "balance_loss_clip": 1.06346202,
      "balance_loss_mlp": 1.03089762,
      "epoch": 0.16954247580111825,
      "flos": 18583700411520.0,
      "grad_norm": 2.2052345873315344,
      "language_loss": 0.79239404,
      "learning_rate": 3.799322307662492e-06,
      "loss": 0.81474942,
      "num_input_tokens_seen": 29737820,
      "step": 1410,
      "time_per_iteration": 2.6510775089263916
    },
    {
      "auxiliary_loss_clip": 0.01168568,
      "auxiliary_loss_mlp": 0.01038519,
      "balance_loss_clip": 1.05618262,
      "balance_loss_mlp": 1.02807581,
      "epoch": 0.16966271869175734,
      "flos": 13983651210240.0,
      "grad_norm": 2.7323244795110364,
      "language_loss": 0.83720154,
      "learning_rate": 3.798982080871496e-06,
      "loss": 0.85927236,
      "num_input_tokens_seen": 29752960,
      "step": 1411,
      "time_per_iteration": 2.728882074356079
    },
    {
      "auxiliary_loss_clip": 0.01224397,
      "auxiliary_loss_mlp": 0.01038605,
      "balance_loss_clip": 1.06291604,
      "balance_loss_mlp": 1.02749431,
      "epoch": 0.16978296158239645,
      "flos": 37487328284160.0,
      "grad_norm": 2.162222253477683,
      "language_loss": 0.67522776,
      "learning_rate": 3.798641581176349e-06,
      "loss": 0.69785774,
      "num_input_tokens_seen": 29775240,
      "step": 1412,
      "time_per_iteration": 2.7433390617370605
    },
    {
      "auxiliary_loss_clip": 0.01190962,
      "auxiliary_loss_mlp": 0.01043901,
      "balance_loss_clip": 1.05623341,
      "balance_loss_mlp": 1.03322601,
      "epoch": 0.16990320447303553,
      "flos": 28328958506880.0,
      "grad_norm": 1.813039688217963,
      "language_loss": 0.74242407,
      "learning_rate": 3.7983008086287044e-06,
      "loss": 0.76477265,
      "num_input_tokens_seen": 29796560,
      "step": 1413,
      "time_per_iteration": 2.7895894050598145
    },
    {
      "auxiliary_loss_clip": 0.01191416,
      "auxiliary_loss_mlp": 0.01034845,
      "balance_loss_clip": 1.05716228,
      "balance_loss_mlp": 1.02407432,
      "epoch": 0.1700234473636746,
      "flos": 20188189031040.0,
      "grad_norm": 2.4660180421609534,
      "language_loss": 0.79757679,
      "learning_rate": 3.797959763280257e-06,
      "loss": 0.81983948,
      "num_input_tokens_seen": 29815245,
      "step": 1414,
      "time_per_iteration": 2.7089343070983887
    },
    {
      "auxiliary_loss_clip": 0.01212121,
      "auxiliary_loss_mlp": 0.01035697,
      "balance_loss_clip": 1.06163597,
      "balance_loss_mlp": 1.02522457,
      "epoch": 0.17014369025431372,
      "flos": 24858658846080.0,
      "grad_norm": 2.567424336397936,
      "language_loss": 0.79042554,
      "learning_rate": 3.797618445182743e-06,
      "loss": 0.8129037,
      "num_input_tokens_seen": 29836640,
      "step": 1415,
      "time_per_iteration": 2.7094528675079346
    },
    {
      "auxiliary_loss_clip": 0.01156095,
      "auxiliary_loss_mlp": 0.01040327,
      "balance_loss_clip": 1.05098486,
      "balance_loss_mlp": 1.02933562,
      "epoch": 0.1702639331449528,
      "flos": 16467233287680.0,
      "grad_norm": 2.087124864082911,
      "language_loss": 0.84926105,
      "learning_rate": 3.79727685438794e-06,
      "loss": 0.8712253,
      "num_input_tokens_seen": 29850830,
      "step": 1416,
      "time_per_iteration": 2.6894283294677734
    },
    {
      "auxiliary_loss_clip": 0.01118992,
      "auxiliary_loss_mlp": 0.01003965,
      "balance_loss_clip": 1.03284335,
      "balance_loss_mlp": 1.00122356,
      "epoch": 0.1703841760355919,
      "flos": 52508870979840.0,
      "grad_norm": 0.982372001611575,
      "language_loss": 0.61682397,
      "learning_rate": 3.796934990947667e-06,
      "loss": 0.6380536,
      "num_input_tokens_seen": 29912515,
      "step": 1417,
      "time_per_iteration": 3.237032413482666
    },
    {
      "auxiliary_loss_clip": 0.01117238,
      "auxiliary_loss_mlp": 0.01004314,
      "balance_loss_clip": 1.0318315,
      "balance_loss_mlp": 1.00147653,
      "epoch": 0.170504418926231,
      "flos": 49370637576960.0,
      "grad_norm": 0.8700664927492163,
      "language_loss": 0.6251235,
      "learning_rate": 3.7965928549137854e-06,
      "loss": 0.64633894,
      "num_input_tokens_seen": 29969330,
      "step": 1418,
      "time_per_iteration": 3.188380002975464
    },
    {
      "auxiliary_loss_clip": 0.01179961,
      "auxiliary_loss_mlp": 0.01043999,
      "balance_loss_clip": 1.05102825,
      "balance_loss_mlp": 1.03204262,
      "epoch": 0.17062466181687008,
      "flos": 25849219184640.0,
      "grad_norm": 3.2081277692011425,
      "language_loss": 0.77560937,
      "learning_rate": 3.7962504463381953e-06,
      "loss": 0.797849,
      "num_input_tokens_seen": 29990820,
      "step": 1419,
      "time_per_iteration": 3.668259620666504
    },
    {
      "auxiliary_loss_clip": 0.01186827,
      "auxiliary_loss_mlp": 0.00765818,
      "balance_loss_clip": 1.0609566,
      "balance_loss_mlp": 1.00066209,
      "epoch": 0.17074490470750917,
      "flos": 20960412549120.0,
      "grad_norm": 3.878299203601888,
      "language_loss": 0.78973007,
      "learning_rate": 3.7959077652728412e-06,
      "loss": 0.80925655,
      "num_input_tokens_seen": 30009275,
      "step": 1420,
      "time_per_iteration": 2.7301025390625
    },
    {
      "auxiliary_loss_clip": 0.01194278,
      "auxiliary_loss_mlp": 0.01033218,
      "balance_loss_clip": 1.05802667,
      "balance_loss_mlp": 1.02266157,
      "epoch": 0.17086514759814825,
      "flos": 20959766104320.0,
      "grad_norm": 2.2269097204284054,
      "language_loss": 0.77199525,
      "learning_rate": 3.795564811769707e-06,
      "loss": 0.79427022,
      "num_input_tokens_seen": 30027630,
      "step": 1421,
      "time_per_iteration": 2.7083685398101807
    },
    {
      "auxiliary_loss_clip": 0.01194398,
      "auxiliary_loss_mlp": 0.01035742,
      "balance_loss_clip": 1.06180477,
      "balance_loss_mlp": 1.02484608,
      "epoch": 0.17098539048878736,
      "flos": 28474073452800.0,
      "grad_norm": 2.5341188234589964,
      "language_loss": 0.78203297,
      "learning_rate": 3.795221585880818e-06,
      "loss": 0.8043344,
      "num_input_tokens_seen": 30048310,
      "step": 1422,
      "time_per_iteration": 2.7298927307128906
    },
    {
      "auxiliary_loss_clip": 0.01178549,
      "auxiliary_loss_mlp": 0.01037526,
      "balance_loss_clip": 1.06067371,
      "balance_loss_mlp": 1.02769709,
      "epoch": 0.17110563337942644,
      "flos": 16290014561280.0,
      "grad_norm": 2.072994167385411,
      "language_loss": 0.91689312,
      "learning_rate": 3.794878087658242e-06,
      "loss": 0.93905389,
      "num_input_tokens_seen": 30066080,
      "step": 1423,
      "time_per_iteration": 2.6999073028564453
    },
    {
      "auxiliary_loss_clip": 0.01211418,
      "auxiliary_loss_mlp": 0.01037824,
      "balance_loss_clip": 1.06128097,
      "balance_loss_mlp": 1.02709484,
      "epoch": 0.17122587627006552,
      "flos": 29674207693440.0,
      "grad_norm": 1.9607214443753314,
      "language_loss": 0.78631586,
      "learning_rate": 3.7945343171540873e-06,
      "loss": 0.80880833,
      "num_input_tokens_seen": 30086955,
      "step": 1424,
      "time_per_iteration": 4.568922281265259
    },
    {
      "auxiliary_loss_clip": 0.01222907,
      "auxiliary_loss_mlp": 0.01038125,
      "balance_loss_clip": 1.06172287,
      "balance_loss_mlp": 1.02758694,
      "epoch": 0.17134611916070464,
      "flos": 25338389915520.0,
      "grad_norm": 2.3551049204746812,
      "language_loss": 0.79356331,
      "learning_rate": 3.7941902744205033e-06,
      "loss": 0.81617361,
      "num_input_tokens_seen": 30107990,
      "step": 1425,
      "time_per_iteration": 3.605212926864624
    },
    {
      "auxiliary_loss_clip": 0.01197834,
      "auxiliary_loss_mlp": 0.01039326,
      "balance_loss_clip": 1.05930197,
      "balance_loss_mlp": 1.02856147,
      "epoch": 0.17146636205134372,
      "flos": 13953845900160.0,
      "grad_norm": 2.101180685369853,
      "language_loss": 0.83530927,
      "learning_rate": 3.7938459595096817e-06,
      "loss": 0.85768092,
      "num_input_tokens_seen": 30126535,
      "step": 1426,
      "time_per_iteration": 2.6255178451538086
    },
    {
      "auxiliary_loss_clip": 0.01212844,
      "auxiliary_loss_mlp": 0.01041164,
      "balance_loss_clip": 1.05943108,
      "balance_loss_mlp": 1.03039932,
      "epoch": 0.1715866049419828,
      "flos": 23915214172800.0,
      "grad_norm": 1.8441442114866078,
      "language_loss": 0.86305904,
      "learning_rate": 3.7935013724738545e-06,
      "loss": 0.88559908,
      "num_input_tokens_seen": 30147035,
      "step": 1427,
      "time_per_iteration": 2.6195647716522217
    },
    {
      "auxiliary_loss_clip": 0.01204013,
      "auxiliary_loss_mlp": 0.01039149,
      "balance_loss_clip": 1.0600667,
      "balance_loss_mlp": 1.0288676,
      "epoch": 0.17170684783262188,
      "flos": 22709369669760.0,
      "grad_norm": 1.8435586132463884,
      "language_loss": 0.78312886,
      "learning_rate": 3.7931565133652945e-06,
      "loss": 0.80556041,
      "num_input_tokens_seen": 30167110,
      "step": 1428,
      "time_per_iteration": 2.708631992340088
    },
    {
      "auxiliary_loss_clip": 0.01223486,
      "auxiliary_loss_mlp": 0.01042278,
      "balance_loss_clip": 1.06147885,
      "balance_loss_mlp": 1.03186464,
      "epoch": 0.171827090723261,
      "flos": 26613290315520.0,
      "grad_norm": 2.8771705462660306,
      "language_loss": 0.67884421,
      "learning_rate": 3.792811382236317e-06,
      "loss": 0.70150185,
      "num_input_tokens_seen": 30185620,
      "step": 1429,
      "time_per_iteration": 2.633237838745117
    },
    {
      "auxiliary_loss_clip": 0.01213982,
      "auxiliary_loss_mlp": 0.01041157,
      "balance_loss_clip": 1.06077766,
      "balance_loss_mlp": 1.02987969,
      "epoch": 0.17194733361390008,
      "flos": 28148507556480.0,
      "grad_norm": 2.079458280323583,
      "language_loss": 0.78143734,
      "learning_rate": 3.792465979139279e-06,
      "loss": 0.80398875,
      "num_input_tokens_seen": 30208225,
      "step": 1430,
      "time_per_iteration": 2.7056972980499268
    },
    {
      "auxiliary_loss_clip": 0.01088954,
      "auxiliary_loss_mlp": 0.01005981,
      "balance_loss_clip": 1.02841711,
      "balance_loss_mlp": 1.00326335,
      "epoch": 0.17206757650453916,
      "flos": 65530689753600.0,
      "grad_norm": 0.9525094892736684,
      "language_loss": 0.65678722,
      "learning_rate": 3.792120304126576e-06,
      "loss": 0.67773658,
      "num_input_tokens_seen": 30271600,
      "step": 1431,
      "time_per_iteration": 3.2865302562713623
    },
    {
      "auxiliary_loss_clip": 0.0113345,
      "auxiliary_loss_mlp": 0.01033877,
      "balance_loss_clip": 1.0499686,
      "balance_loss_mlp": 1.02333331,
      "epoch": 0.17218781939517827,
      "flos": 22273486128000.0,
      "grad_norm": 1.9197448995605524,
      "language_loss": 0.83603144,
      "learning_rate": 3.791774357250649e-06,
      "loss": 0.85770464,
      "num_input_tokens_seen": 30290430,
      "step": 1432,
      "time_per_iteration": 2.7867259979248047
    },
    {
      "auxiliary_loss_clip": 0.01187536,
      "auxiliary_loss_mlp": 0.01038714,
      "balance_loss_clip": 1.05668092,
      "balance_loss_mlp": 1.02852201,
      "epoch": 0.17230806228581735,
      "flos": 14137313592960.0,
      "grad_norm": 2.130523236315261,
      "language_loss": 0.78688818,
      "learning_rate": 3.7914281385639757e-06,
      "loss": 0.8091507,
      "num_input_tokens_seen": 30308305,
      "step": 1433,
      "time_per_iteration": 2.6444058418273926
    },
    {
      "auxiliary_loss_clip": 0.0120583,
      "auxiliary_loss_mlp": 0.01033073,
      "balance_loss_clip": 1.05666518,
      "balance_loss_mlp": 1.02267206,
      "epoch": 0.17242830517645644,
      "flos": 20704836303360.0,
      "grad_norm": 2.3925042493742197,
      "language_loss": 0.79963535,
      "learning_rate": 3.7910816481190784e-06,
      "loss": 0.82202435,
      "num_input_tokens_seen": 30328120,
      "step": 1434,
      "time_per_iteration": 2.6217660903930664
    },
    {
      "auxiliary_loss_clip": 0.01181486,
      "auxiliary_loss_mlp": 0.01031999,
      "balance_loss_clip": 1.05512881,
      "balance_loss_mlp": 1.02124023,
      "epoch": 0.17254854806709552,
      "flos": 30774582887040.0,
      "grad_norm": 2.3126469802090344,
      "language_loss": 0.75050175,
      "learning_rate": 3.7907348859685193e-06,
      "loss": 0.77263653,
      "num_input_tokens_seen": 30349825,
      "step": 1435,
      "time_per_iteration": 2.75966215133667
    },
    {
      "auxiliary_loss_clip": 0.01206303,
      "auxiliary_loss_mlp": 0.01032868,
      "balance_loss_clip": 1.06273782,
      "balance_loss_mlp": 1.02244961,
      "epoch": 0.17266879095773463,
      "flos": 26614726859520.0,
      "grad_norm": 2.043725839902142,
      "language_loss": 0.80907905,
      "learning_rate": 3.790387852164902e-06,
      "loss": 0.83147079,
      "num_input_tokens_seen": 30370555,
      "step": 1436,
      "time_per_iteration": 2.661001682281494
    },
    {
      "auxiliary_loss_clip": 0.01207801,
      "auxiliary_loss_mlp": 0.01036431,
      "balance_loss_clip": 1.06041312,
      "balance_loss_mlp": 1.02564812,
      "epoch": 0.1727890338483737,
      "flos": 20266295155200.0,
      "grad_norm": 2.014579748757546,
      "language_loss": 0.76593566,
      "learning_rate": 3.7900405467608707e-06,
      "loss": 0.788378,
      "num_input_tokens_seen": 30390100,
      "step": 1437,
      "time_per_iteration": 2.675593137741089
    },
    {
      "auxiliary_loss_clip": 0.01148599,
      "auxiliary_loss_mlp": 0.010357,
      "balance_loss_clip": 1.04925644,
      "balance_loss_mlp": 1.02411282,
      "epoch": 0.1729092767390128,
      "flos": 18179812909440.0,
      "grad_norm": 3.590798098147219,
      "language_loss": 0.79428625,
      "learning_rate": 3.7896929698091114e-06,
      "loss": 0.81612927,
      "num_input_tokens_seen": 30402915,
      "step": 1438,
      "time_per_iteration": 2.682413339614868
    },
    {
      "auxiliary_loss_clip": 0.0122842,
      "auxiliary_loss_mlp": 0.01051644,
      "balance_loss_clip": 1.06611288,
      "balance_loss_mlp": 1.03979445,
      "epoch": 0.1730295196296519,
      "flos": 26759518583040.0,
      "grad_norm": 2.768274637306085,
      "language_loss": 0.68118602,
      "learning_rate": 3.7893451213623518e-06,
      "loss": 0.70398664,
      "num_input_tokens_seen": 30420145,
      "step": 1439,
      "time_per_iteration": 2.6361167430877686
    },
    {
      "auxiliary_loss_clip": 0.01208547,
      "auxiliary_loss_mlp": 0.00765922,
      "balance_loss_clip": 1.06347561,
      "balance_loss_mlp": 1.00058746,
      "epoch": 0.173149762520291,
      "flos": 23842531002240.0,
      "grad_norm": 2.3066697891852668,
      "language_loss": 0.82222337,
      "learning_rate": 3.7889970014733606e-06,
      "loss": 0.84196806,
      "num_input_tokens_seen": 30439250,
      "step": 1440,
      "time_per_iteration": 2.7188098430633545
    },
    {
      "auxiliary_loss_clip": 0.0114482,
      "auxiliary_loss_mlp": 0.0103851,
      "balance_loss_clip": 1.04918766,
      "balance_loss_mlp": 1.02732229,
      "epoch": 0.17327000541093007,
      "flos": 23368186972800.0,
      "grad_norm": 1.694784205306711,
      "language_loss": 0.78321624,
      "learning_rate": 3.7886486101949463e-06,
      "loss": 0.80504954,
      "num_input_tokens_seen": 30460430,
      "step": 1441,
      "time_per_iteration": 2.748162269592285
    },
    {
      "auxiliary_loss_clip": 0.01149221,
      "auxiliary_loss_mlp": 0.01038761,
      "balance_loss_clip": 1.04993248,
      "balance_loss_mlp": 1.02721012,
      "epoch": 0.17339024830156918,
      "flos": 18221290139520.0,
      "grad_norm": 2.0628141629239454,
      "language_loss": 0.8801896,
      "learning_rate": 3.7882999475799594e-06,
      "loss": 0.90206939,
      "num_input_tokens_seen": 30478465,
      "step": 1442,
      "time_per_iteration": 2.7650859355926514
    },
    {
      "auxiliary_loss_clip": 0.01142461,
      "auxiliary_loss_mlp": 0.01041925,
      "balance_loss_clip": 1.05152965,
      "balance_loss_mlp": 1.03154206,
      "epoch": 0.17351049119220827,
      "flos": 23332024955520.0,
      "grad_norm": 1.8386888472608776,
      "language_loss": 0.81268907,
      "learning_rate": 3.787951013681293e-06,
      "loss": 0.83453298,
      "num_input_tokens_seen": 30496510,
      "step": 1443,
      "time_per_iteration": 2.8288912773132324
    },
    {
      "auxiliary_loss_clip": 0.01208011,
      "auxiliary_loss_mlp": 0.01037803,
      "balance_loss_clip": 1.0587374,
      "balance_loss_mlp": 1.02719367,
      "epoch": 0.17363073408284735,
      "flos": 23803495896960.0,
      "grad_norm": 2.047541340593442,
      "language_loss": 0.7761966,
      "learning_rate": 3.787601808551879e-06,
      "loss": 0.79865479,
      "num_input_tokens_seen": 30516325,
      "step": 1444,
      "time_per_iteration": 2.7429821491241455
    },
    {
      "auxiliary_loss_clip": 0.01181325,
      "auxiliary_loss_mlp": 0.01043104,
      "balance_loss_clip": 1.0566895,
      "balance_loss_mlp": 1.03154659,
      "epoch": 0.17375097697348643,
      "flos": 18515290959360.0,
      "grad_norm": 3.2698178571771495,
      "language_loss": 0.84157097,
      "learning_rate": 3.7872523322446926e-06,
      "loss": 0.86381531,
      "num_input_tokens_seen": 30535210,
      "step": 1445,
      "time_per_iteration": 2.6699202060699463
    },
    {
      "auxiliary_loss_clip": 0.01166652,
      "auxiliary_loss_mlp": 0.01033224,
      "balance_loss_clip": 1.05013978,
      "balance_loss_mlp": 1.02255452,
      "epoch": 0.17387121986412554,
      "flos": 38877897456000.0,
      "grad_norm": 1.6526688703505965,
      "language_loss": 0.60292137,
      "learning_rate": 3.7869025848127478e-06,
      "loss": 0.62492019,
      "num_input_tokens_seen": 30559405,
      "step": 1446,
      "time_per_iteration": 3.788430690765381
    },
    {
      "auxiliary_loss_clip": 0.01209474,
      "auxiliary_loss_mlp": 0.01036914,
      "balance_loss_clip": 1.05833697,
      "balance_loss_mlp": 1.02670348,
      "epoch": 0.17399146275476463,
      "flos": 20375714960640.0,
      "grad_norm": 2.6239609962461423,
      "language_loss": 0.80768383,
      "learning_rate": 3.786552566309102e-06,
      "loss": 0.83014774,
      "num_input_tokens_seen": 30577615,
      "step": 1447,
      "time_per_iteration": 2.601665735244751
    },
    {
      "auxiliary_loss_clip": 0.0119255,
      "auxiliary_loss_mlp": 0.0076562,
      "balance_loss_clip": 1.06085896,
      "balance_loss_mlp": 1.00049806,
      "epoch": 0.1741117056454037,
      "flos": 19164339763200.0,
      "grad_norm": 2.8120628978777464,
      "language_loss": 0.86157525,
      "learning_rate": 3.7862022767868517e-06,
      "loss": 0.88115692,
      "num_input_tokens_seen": 30595205,
      "step": 1448,
      "time_per_iteration": 2.703667640686035
    },
    {
      "auxiliary_loss_clip": 0.0117648,
      "auxiliary_loss_mlp": 0.0104016,
      "balance_loss_clip": 1.06227255,
      "balance_loss_mlp": 1.02893603,
      "epoch": 0.17423194853604282,
      "flos": 25374300537600.0,
      "grad_norm": 2.806840894651503,
      "language_loss": 0.84633338,
      "learning_rate": 3.7858517162991367e-06,
      "loss": 0.86849976,
      "num_input_tokens_seen": 30615280,
      "step": 1449,
      "time_per_iteration": 2.731600284576416
    },
    {
      "auxiliary_loss_clip": 0.01179586,
      "auxiliary_loss_mlp": 0.01037421,
      "balance_loss_clip": 1.05336297,
      "balance_loss_mlp": 1.02658534,
      "epoch": 0.1743521914266819,
      "flos": 25191874339200.0,
      "grad_norm": 2.6047806528481043,
      "language_loss": 0.61037409,
      "learning_rate": 3.7855008848991363e-06,
      "loss": 0.63254416,
      "num_input_tokens_seen": 30633485,
      "step": 1450,
      "time_per_iteration": 5.538543224334717
    },
    {
      "auxiliary_loss_clip": 0.01186003,
      "auxiliary_loss_mlp": 0.01032997,
      "balance_loss_clip": 1.05597258,
      "balance_loss_mlp": 1.02235806,
      "epoch": 0.17447243431732098,
      "flos": 25666577504640.0,
      "grad_norm": 1.8930947032185024,
      "language_loss": 0.77847755,
      "learning_rate": 3.7851497826400714e-06,
      "loss": 0.80066752,
      "num_input_tokens_seen": 30653625,
      "step": 1451,
      "time_per_iteration": 2.7068231105804443
    },
    {
      "auxiliary_loss_clip": 0.01225303,
      "auxiliary_loss_mlp": 0.01037223,
      "balance_loss_clip": 1.06441963,
      "balance_loss_mlp": 1.02571297,
      "epoch": 0.17459267720796007,
      "flos": 36281950657920.0,
      "grad_norm": 2.190051188089824,
      "language_loss": 0.76136565,
      "learning_rate": 3.7847984095752034e-06,
      "loss": 0.78399086,
      "num_input_tokens_seen": 30677080,
      "step": 1452,
      "time_per_iteration": 2.819262981414795
    },
    {
      "auxiliary_loss_clip": 0.01223189,
      "auxiliary_loss_mlp": 0.01041839,
      "balance_loss_clip": 1.06170464,
      "balance_loss_mlp": 1.03094912,
      "epoch": 0.17471292009859918,
      "flos": 20011113959040.0,
      "grad_norm": 2.1789644755244386,
      "language_loss": 0.80741262,
      "learning_rate": 3.784446765757836e-06,
      "loss": 0.83006287,
      "num_input_tokens_seen": 30695725,
      "step": 1453,
      "time_per_iteration": 2.548492193222046
    },
    {
      "auxiliary_loss_clip": 0.01162756,
      "auxiliary_loss_mlp": 0.01035257,
      "balance_loss_clip": 1.05346417,
      "balance_loss_mlp": 1.02431345,
      "epoch": 0.17483316298923826,
      "flos": 27819242559360.0,
      "grad_norm": 2.1830890684036377,
      "language_loss": 0.77905548,
      "learning_rate": 3.7840948512413133e-06,
      "loss": 0.80103564,
      "num_input_tokens_seen": 30713310,
      "step": 1454,
      "time_per_iteration": 2.7616875171661377
    },
    {
      "auxiliary_loss_clip": 0.0117651,
      "auxiliary_loss_mlp": 0.01036416,
      "balance_loss_clip": 1.05978966,
      "balance_loss_mlp": 1.02546036,
      "epoch": 0.17495340587987734,
      "flos": 44017934791680.0,
      "grad_norm": 2.0477412226349414,
      "language_loss": 0.79007757,
      "learning_rate": 3.7837426660790196e-06,
      "loss": 0.81220686,
      "num_input_tokens_seen": 30734725,
      "step": 1455,
      "time_per_iteration": 2.8776140213012695
    },
    {
      "auxiliary_loss_clip": 0.0122028,
      "auxiliary_loss_mlp": 0.01037465,
      "balance_loss_clip": 1.06004345,
      "balance_loss_mlp": 1.02677786,
      "epoch": 0.17507364877051645,
      "flos": 20885825957760.0,
      "grad_norm": 2.0319704277145676,
      "language_loss": 0.8212356,
      "learning_rate": 3.783390210324382e-06,
      "loss": 0.84381306,
      "num_input_tokens_seen": 30754450,
      "step": 1456,
      "time_per_iteration": 2.6259491443634033
    },
    {
      "auxiliary_loss_clip": 0.01177036,
      "auxiliary_loss_mlp": 0.01033826,
      "balance_loss_clip": 1.05782819,
      "balance_loss_mlp": 1.02311468,
      "epoch": 0.17519389166115554,
      "flos": 24717602136960.0,
      "grad_norm": 8.358997140889132,
      "language_loss": 0.72720116,
      "learning_rate": 3.7830374840308676e-06,
      "loss": 0.74930978,
      "num_input_tokens_seen": 30774605,
      "step": 1457,
      "time_per_iteration": 2.73349928855896
    },
    {
      "auxiliary_loss_clip": 0.01210574,
      "auxiliary_loss_mlp": 0.01036559,
      "balance_loss_clip": 1.06397033,
      "balance_loss_mlp": 1.02533507,
      "epoch": 0.17531413455179462,
      "flos": 23798144770560.0,
      "grad_norm": 2.944344584648095,
      "language_loss": 0.82482076,
      "learning_rate": 3.7826844872519842e-06,
      "loss": 0.84729207,
      "num_input_tokens_seen": 30792460,
      "step": 1458,
      "time_per_iteration": 2.65874981880188
    },
    {
      "auxiliary_loss_clip": 0.01193872,
      "auxiliary_loss_mlp": 0.0103752,
      "balance_loss_clip": 1.06271482,
      "balance_loss_mlp": 1.02718413,
      "epoch": 0.1754343774424337,
      "flos": 24572379450240.0,
      "grad_norm": 3.7897461846972207,
      "language_loss": 0.72351772,
      "learning_rate": 3.782331220041282e-06,
      "loss": 0.74583161,
      "num_input_tokens_seen": 30812525,
      "step": 1459,
      "time_per_iteration": 2.7126193046569824
    },
    {
      "auxiliary_loss_clip": 0.01186052,
      "auxiliary_loss_mlp": 0.01036455,
      "balance_loss_clip": 1.05572891,
      "balance_loss_mlp": 1.02604222,
      "epoch": 0.17555462033307281,
      "flos": 18114599767680.0,
      "grad_norm": 2.9777343610132925,
      "language_loss": 0.83277702,
      "learning_rate": 3.7819776824523504e-06,
      "loss": 0.85500205,
      "num_input_tokens_seen": 30830390,
      "step": 1460,
      "time_per_iteration": 2.6637237071990967
    },
    {
      "auxiliary_loss_clip": 0.01203757,
      "auxiliary_loss_mlp": 0.01038733,
      "balance_loss_clip": 1.06007195,
      "balance_loss_mlp": 1.02743793,
      "epoch": 0.1756748632237119,
      "flos": 28366018364160.0,
      "grad_norm": 1.8995387074856196,
      "language_loss": 0.83891654,
      "learning_rate": 3.7816238745388213e-06,
      "loss": 0.86134142,
      "num_input_tokens_seen": 30849935,
      "step": 1461,
      "time_per_iteration": 2.715343713760376
    },
    {
      "auxiliary_loss_clip": 0.01195339,
      "auxiliary_loss_mlp": 0.01039226,
      "balance_loss_clip": 1.05519998,
      "balance_loss_mlp": 1.0279603,
      "epoch": 0.17579510611435098,
      "flos": 25732939881600.0,
      "grad_norm": 2.06976280443509,
      "language_loss": 0.87413156,
      "learning_rate": 3.781269796354367e-06,
      "loss": 0.89647722,
      "num_input_tokens_seen": 30869555,
      "step": 1462,
      "time_per_iteration": 2.7079031467437744
    },
    {
      "auxiliary_loss_clip": 0.0120052,
      "auxiliary_loss_mlp": 0.01041077,
      "balance_loss_clip": 1.06368864,
      "balance_loss_mlp": 1.03033602,
      "epoch": 0.1759153490049901,
      "flos": 18588081870720.0,
      "grad_norm": 1.7366994528358626,
      "language_loss": 0.8589142,
      "learning_rate": 3.7809154479527006e-06,
      "loss": 0.88133013,
      "num_input_tokens_seen": 30888760,
      "step": 1463,
      "time_per_iteration": 2.664198875427246
    },
    {
      "auxiliary_loss_clip": 0.01170275,
      "auxiliary_loss_mlp": 0.01032458,
      "balance_loss_clip": 1.05361068,
      "balance_loss_mlp": 1.02175856,
      "epoch": 0.17603559189562917,
      "flos": 18619323724800.0,
      "grad_norm": 2.300537578986749,
      "language_loss": 0.84274334,
      "learning_rate": 3.780560829387577e-06,
      "loss": 0.86477065,
      "num_input_tokens_seen": 30907260,
      "step": 1464,
      "time_per_iteration": 2.6615333557128906
    },
    {
      "auxiliary_loss_clip": 0.01113461,
      "auxiliary_loss_mlp": 0.01008904,
      "balance_loss_clip": 1.03028893,
      "balance_loss_mlp": 1.00634134,
      "epoch": 0.17615583478626826,
      "flos": 60530775373440.0,
      "grad_norm": 0.856271415963087,
      "language_loss": 0.57956278,
      "learning_rate": 3.7802059407127915e-06,
      "loss": 0.60078645,
      "num_input_tokens_seen": 30965810,
      "step": 1465,
      "time_per_iteration": 3.1868064403533936
    },
    {
      "auxiliary_loss_clip": 0.01185199,
      "auxiliary_loss_mlp": 0.0104106,
      "balance_loss_clip": 1.05337477,
      "balance_loss_mlp": 1.02999187,
      "epoch": 0.17627607767690734,
      "flos": 23616221362560.0,
      "grad_norm": 2.22115736069406,
      "language_loss": 0.86102527,
      "learning_rate": 3.7798507819821797e-06,
      "loss": 0.88328791,
      "num_input_tokens_seen": 30982935,
      "step": 1466,
      "time_per_iteration": 2.671494960784912
    },
    {
      "auxiliary_loss_clip": 0.01167466,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.0556314,
      "balance_loss_mlp": 1.01792693,
      "epoch": 0.17639632056754645,
      "flos": 17639070589440.0,
      "grad_norm": 2.4798735935630187,
      "language_loss": 0.79383594,
      "learning_rate": 3.7794953532496197e-06,
      "loss": 0.81579888,
      "num_input_tokens_seen": 30998840,
      "step": 1467,
      "time_per_iteration": 2.6757991313934326
    },
    {
      "auxiliary_loss_clip": 0.01087377,
      "auxiliary_loss_mlp": 0.00757265,
      "balance_loss_clip": 1.04324389,
      "balance_loss_mlp": 1.00107861,
      "epoch": 0.17651656345818553,
      "flos": 57932604910080.0,
      "grad_norm": 0.8577574351629216,
      "language_loss": 0.5798983,
      "learning_rate": 3.7791396545690295e-06,
      "loss": 0.59834474,
      "num_input_tokens_seen": 31060075,
      "step": 1468,
      "time_per_iteration": 3.2522215843200684
    },
    {
      "auxiliary_loss_clip": 0.01208372,
      "auxiliary_loss_mlp": 0.01040576,
      "balance_loss_clip": 1.06046963,
      "balance_loss_mlp": 1.02988267,
      "epoch": 0.17663680634882462,
      "flos": 22929502170240.0,
      "grad_norm": 2.0834266025202073,
      "language_loss": 0.80879736,
      "learning_rate": 3.7787836859943685e-06,
      "loss": 0.83128685,
      "num_input_tokens_seen": 31078800,
      "step": 1469,
      "time_per_iteration": 2.6371004581451416
    },
    {
      "auxiliary_loss_clip": 0.01203308,
      "auxiliary_loss_mlp": 0.01033729,
      "balance_loss_clip": 1.05575204,
      "balance_loss_mlp": 1.02227914,
      "epoch": 0.17675704923946373,
      "flos": 22637979388800.0,
      "grad_norm": 2.4691593755137333,
      "language_loss": 0.78554195,
      "learning_rate": 3.7784274475796363e-06,
      "loss": 0.80791235,
      "num_input_tokens_seen": 31097430,
      "step": 1470,
      "time_per_iteration": 2.6697990894317627
    },
    {
      "auxiliary_loss_clip": 0.0118212,
      "auxiliary_loss_mlp": 0.01040622,
      "balance_loss_clip": 1.05588055,
      "balance_loss_mlp": 1.02847469,
      "epoch": 0.1768772921301028,
      "flos": 27126525795840.0,
      "grad_norm": 2.7366067321924734,
      "language_loss": 0.75790048,
      "learning_rate": 3.7780709393788745e-06,
      "loss": 0.78012794,
      "num_input_tokens_seen": 31117905,
      "step": 1471,
      "time_per_iteration": 2.7280287742614746
    },
    {
      "auxiliary_loss_clip": 0.01220801,
      "auxiliary_loss_mlp": 0.01038116,
      "balance_loss_clip": 1.06104171,
      "balance_loss_mlp": 1.02708316,
      "epoch": 0.1769975350207419,
      "flos": 19172133014400.0,
      "grad_norm": 2.0482150438751665,
      "language_loss": 0.75249445,
      "learning_rate": 3.777714161446165e-06,
      "loss": 0.77508354,
      "num_input_tokens_seen": 31137610,
      "step": 1472,
      "time_per_iteration": 3.607952117919922
    },
    {
      "auxiliary_loss_clip": 0.01210978,
      "auxiliary_loss_mlp": 0.01034546,
      "balance_loss_clip": 1.06048012,
      "balance_loss_mlp": 1.02381182,
      "epoch": 0.177117777911381,
      "flos": 36134932291200.0,
      "grad_norm": 3.482897521573406,
      "language_loss": 0.6954968,
      "learning_rate": 3.7773571138356304e-06,
      "loss": 0.71795207,
      "num_input_tokens_seen": 31157780,
      "step": 1473,
      "time_per_iteration": 2.8020639419555664
    },
    {
      "auxiliary_loss_clip": 0.01152826,
      "auxiliary_loss_mlp": 0.01042225,
      "balance_loss_clip": 1.0558846,
      "balance_loss_mlp": 1.03165662,
      "epoch": 0.17723802080202009,
      "flos": 22090593052800.0,
      "grad_norm": 2.819462449038129,
      "language_loss": 0.89291161,
      "learning_rate": 3.776999796601435e-06,
      "loss": 0.9148621,
      "num_input_tokens_seen": 31176540,
      "step": 1474,
      "time_per_iteration": 2.712747812271118
    },
    {
      "auxiliary_loss_clip": 0.01216427,
      "auxiliary_loss_mlp": 0.01040648,
      "balance_loss_clip": 1.06424701,
      "balance_loss_mlp": 1.02925122,
      "epoch": 0.17735826369265917,
      "flos": 30222671437440.0,
      "grad_norm": 2.2350172850720815,
      "language_loss": 0.72904682,
      "learning_rate": 3.776642209797783e-06,
      "loss": 0.75161755,
      "num_input_tokens_seen": 31198370,
      "step": 1475,
      "time_per_iteration": 2.7424850463867188
    },
    {
      "auxiliary_loss_clip": 0.01200332,
      "auxiliary_loss_mlp": 0.01046608,
      "balance_loss_clip": 1.05749202,
      "balance_loss_mlp": 1.03491926,
      "epoch": 0.17747850658329825,
      "flos": 21397588980480.0,
      "grad_norm": 2.4739380629783407,
      "language_loss": 0.78113347,
      "learning_rate": 3.7762843534789205e-06,
      "loss": 0.80360293,
      "num_input_tokens_seen": 31217120,
      "step": 1476,
      "time_per_iteration": 4.455347537994385
    },
    {
      "auxiliary_loss_clip": 0.0119609,
      "auxiliary_loss_mlp": 0.01036031,
      "balance_loss_clip": 1.05695081,
      "balance_loss_mlp": 1.02432466,
      "epoch": 0.17759874947393736,
      "flos": 16983341856000.0,
      "grad_norm": 3.9846752132317707,
      "language_loss": 0.88412607,
      "learning_rate": 3.7759262276991343e-06,
      "loss": 0.90644723,
      "num_input_tokens_seen": 31234730,
      "step": 1477,
      "time_per_iteration": 2.6873996257781982
    },
    {
      "auxiliary_loss_clip": 0.01200224,
      "auxiliary_loss_mlp": 0.01041776,
      "balance_loss_clip": 1.06019711,
      "balance_loss_mlp": 1.03039777,
      "epoch": 0.17771899236457644,
      "flos": 11546107390080.0,
      "grad_norm": 3.4652024160164223,
      "language_loss": 0.80727243,
      "learning_rate": 3.7755678325127506e-06,
      "loss": 0.82969242,
      "num_input_tokens_seen": 31252410,
      "step": 1478,
      "time_per_iteration": 2.6408019065856934
    },
    {
      "auxiliary_loss_clip": 0.01157659,
      "auxiliary_loss_mlp": 0.01038433,
      "balance_loss_clip": 1.05624568,
      "balance_loss_mlp": 1.02881265,
      "epoch": 0.17783923525521553,
      "flos": 18807747494400.0,
      "grad_norm": 1.8664464248934465,
      "language_loss": 0.75644612,
      "learning_rate": 3.7752091679741393e-06,
      "loss": 0.77840698,
      "num_input_tokens_seen": 31270200,
      "step": 1479,
      "time_per_iteration": 2.7341856956481934
    },
    {
      "auxiliary_loss_clip": 0.01205086,
      "auxiliary_loss_mlp": 0.01037014,
      "balance_loss_clip": 1.05916214,
      "balance_loss_mlp": 1.02570081,
      "epoch": 0.17795947814585464,
      "flos": 30408365773440.0,
      "grad_norm": 5.07705131741926,
      "language_loss": 0.77805638,
      "learning_rate": 3.774850234137708e-06,
      "loss": 0.80047739,
      "num_input_tokens_seen": 31287495,
      "step": 1480,
      "time_per_iteration": 2.721266269683838
    },
    {
      "auxiliary_loss_clip": 0.01204305,
      "auxiliary_loss_mlp": 0.01040749,
      "balance_loss_clip": 1.05854452,
      "balance_loss_mlp": 1.02960896,
      "epoch": 0.17807972103649372,
      "flos": 24389055411840.0,
      "grad_norm": 3.202785772095777,
      "language_loss": 0.83118099,
      "learning_rate": 3.7744910310579076e-06,
      "loss": 0.85363156,
      "num_input_tokens_seen": 31306420,
      "step": 1481,
      "time_per_iteration": 2.6455836296081543
    },
    {
      "auxiliary_loss_clip": 0.01222852,
      "auxiliary_loss_mlp": 0.01029552,
      "balance_loss_clip": 1.06354475,
      "balance_loss_mlp": 1.01946104,
      "epoch": 0.1781999639271328,
      "flos": 20301559332480.0,
      "grad_norm": 2.183064579157,
      "language_loss": 0.85431188,
      "learning_rate": 3.774131558789229e-06,
      "loss": 0.87683594,
      "num_input_tokens_seen": 31325750,
      "step": 1482,
      "time_per_iteration": 2.637542724609375
    },
    {
      "auxiliary_loss_clip": 0.01221943,
      "auxiliary_loss_mlp": 0.00765529,
      "balance_loss_clip": 1.06272209,
      "balance_loss_mlp": 1.00032067,
      "epoch": 0.1783202068177719,
      "flos": 15924479806080.0,
      "grad_norm": 2.9361214873853685,
      "language_loss": 0.70261681,
      "learning_rate": 3.773771817386203e-06,
      "loss": 0.7224915,
      "num_input_tokens_seen": 31343080,
      "step": 1483,
      "time_per_iteration": 2.561711072921753
    },
    {
      "auxiliary_loss_clip": 0.01192224,
      "auxiliary_loss_mlp": 0.01042113,
      "balance_loss_clip": 1.05820155,
      "balance_loss_mlp": 1.0317241,
      "epoch": 0.178440449708411,
      "flos": 20631758083200.0,
      "grad_norm": 1.6507507940809993,
      "language_loss": 0.79399836,
      "learning_rate": 3.773411806903403e-06,
      "loss": 0.81634176,
      "num_input_tokens_seen": 31362160,
      "step": 1484,
      "time_per_iteration": 2.7249412536621094
    },
    {
      "auxiliary_loss_clip": 0.01156171,
      "auxiliary_loss_mlp": 0.01044214,
      "balance_loss_clip": 1.05597425,
      "balance_loss_mlp": 1.03334165,
      "epoch": 0.17856069259905008,
      "flos": 21686059105920.0,
      "grad_norm": 1.8652448150744427,
      "language_loss": 0.94738078,
      "learning_rate": 3.7730515273954415e-06,
      "loss": 0.96938455,
      "num_input_tokens_seen": 31380770,
      "step": 1485,
      "time_per_iteration": 2.755969524383545
    },
    {
      "auxiliary_loss_clip": 0.01220911,
      "auxiliary_loss_mlp": 0.01036151,
      "balance_loss_clip": 1.06308913,
      "balance_loss_mlp": 1.02505898,
      "epoch": 0.17868093548968916,
      "flos": 26572962320640.0,
      "grad_norm": 2.4590614620788918,
      "language_loss": 0.85089135,
      "learning_rate": 3.772690978916973e-06,
      "loss": 0.87346196,
      "num_input_tokens_seen": 31400525,
      "step": 1486,
      "time_per_iteration": 2.6572811603546143
    },
    {
      "auxiliary_loss_clip": 0.01206889,
      "auxiliary_loss_mlp": 0.01041186,
      "balance_loss_clip": 1.06037462,
      "balance_loss_mlp": 1.02993226,
      "epoch": 0.17880117838032827,
      "flos": 18581006891520.0,
      "grad_norm": 2.0196853401237127,
      "language_loss": 0.86559904,
      "learning_rate": 3.772330161522693e-06,
      "loss": 0.88807976,
      "num_input_tokens_seen": 31418435,
      "step": 1487,
      "time_per_iteration": 2.62556529045105
    },
    {
      "auxiliary_loss_clip": 0.01195675,
      "auxiliary_loss_mlp": 0.01033323,
      "balance_loss_clip": 1.06579161,
      "balance_loss_mlp": 1.02231443,
      "epoch": 0.17892142127096736,
      "flos": 26541217676160.0,
      "grad_norm": 2.22740113039234,
      "language_loss": 0.79661804,
      "learning_rate": 3.7719690752673365e-06,
      "loss": 0.81890798,
      "num_input_tokens_seen": 31439230,
      "step": 1488,
      "time_per_iteration": 2.6927714347839355
    },
    {
      "auxiliary_loss_clip": 0.01181992,
      "auxiliary_loss_mlp": 0.01038525,
      "balance_loss_clip": 1.06125891,
      "balance_loss_mlp": 1.02842784,
      "epoch": 0.17904166416160644,
      "flos": 23872623621120.0,
      "grad_norm": 2.1764372132252636,
      "language_loss": 0.78681922,
      "learning_rate": 3.7716077202056796e-06,
      "loss": 0.80902445,
      "num_input_tokens_seen": 31457705,
      "step": 1489,
      "time_per_iteration": 2.76432204246521
    },
    {
      "auxiliary_loss_clip": 0.01180209,
      "auxiliary_loss_mlp": 0.01036492,
      "balance_loss_clip": 1.05742669,
      "balance_loss_mlp": 1.02562642,
      "epoch": 0.17916190705224552,
      "flos": 19134426712320.0,
      "grad_norm": 2.7845375015612728,
      "language_loss": 0.93584228,
      "learning_rate": 3.7712460963925404e-06,
      "loss": 0.95800936,
      "num_input_tokens_seen": 31473645,
      "step": 1490,
      "time_per_iteration": 2.6645596027374268
    },
    {
      "auxiliary_loss_clip": 0.01180206,
      "auxiliary_loss_mlp": 0.01032008,
      "balance_loss_clip": 1.05305052,
      "balance_loss_mlp": 1.02136254,
      "epoch": 0.17928214994288463,
      "flos": 25152120961920.0,
      "grad_norm": 2.104899015224723,
      "language_loss": 0.75887668,
      "learning_rate": 3.7708842038827775e-06,
      "loss": 0.78099883,
      "num_input_tokens_seen": 31492605,
      "step": 1491,
      "time_per_iteration": 2.6988956928253174
    },
    {
      "auxiliary_loss_clip": 0.01206247,
      "auxiliary_loss_mlp": 0.01043209,
      "balance_loss_clip": 1.05870342,
      "balance_loss_mlp": 1.03233719,
      "epoch": 0.17940239283352372,
      "flos": 22384629786240.0,
      "grad_norm": 1.837977430375674,
      "language_loss": 0.85780579,
      "learning_rate": 3.770522042731288e-06,
      "loss": 0.88030034,
      "num_input_tokens_seen": 31514500,
      "step": 1492,
      "time_per_iteration": 2.6633012294769287
    },
    {
      "auxiliary_loss_clip": 0.01153884,
      "auxiliary_loss_mlp": 0.01034693,
      "balance_loss_clip": 1.05431116,
      "balance_loss_mlp": 1.02268255,
      "epoch": 0.1795226357241628,
      "flos": 23178685795200.0,
      "grad_norm": 3.4223667252981294,
      "language_loss": 0.88099819,
      "learning_rate": 3.7701596129930122e-06,
      "loss": 0.90288401,
      "num_input_tokens_seen": 31533225,
      "step": 1493,
      "time_per_iteration": 2.7647504806518555
    },
    {
      "auxiliary_loss_clip": 0.01185472,
      "auxiliary_loss_mlp": 0.01035223,
      "balance_loss_clip": 1.05919743,
      "balance_loss_mlp": 1.0241785,
      "epoch": 0.1796428786148019,
      "flos": 22090413484800.0,
      "grad_norm": 1.9985352966411212,
      "language_loss": 0.73371691,
      "learning_rate": 3.7697969147229315e-06,
      "loss": 0.75592387,
      "num_input_tokens_seen": 31551385,
      "step": 1494,
      "time_per_iteration": 2.6852943897247314
    },
    {
      "auxiliary_loss_clip": 0.01203908,
      "auxiliary_loss_mlp": 0.01034932,
      "balance_loss_clip": 1.05812299,
      "balance_loss_mlp": 1.0242691,
      "epoch": 0.179763121505441,
      "flos": 21324618501120.0,
      "grad_norm": 3.422576437416537,
      "language_loss": 0.85315645,
      "learning_rate": 3.7694339479760647e-06,
      "loss": 0.87554485,
      "num_input_tokens_seen": 31570415,
      "step": 1495,
      "time_per_iteration": 2.730390787124634
    },
    {
      "auxiliary_loss_clip": 0.01102571,
      "auxiliary_loss_mlp": 0.01005075,
      "balance_loss_clip": 1.03228998,
      "balance_loss_mlp": 1.00238132,
      "epoch": 0.17988336439608008,
      "flos": 68161864815360.0,
      "grad_norm": 0.8011476154917179,
      "language_loss": 0.5735482,
      "learning_rate": 3.769070712807476e-06,
      "loss": 0.59462464,
      "num_input_tokens_seen": 31632445,
      "step": 1496,
      "time_per_iteration": 3.2969655990600586
    },
    {
      "auxiliary_loss_clip": 0.01137501,
      "auxiliary_loss_mlp": 0.01037596,
      "balance_loss_clip": 1.0543561,
      "balance_loss_mlp": 1.02743888,
      "epoch": 0.18000360728671919,
      "flos": 21945047143680.0,
      "grad_norm": 2.2346381953774044,
      "language_loss": 0.78763902,
      "learning_rate": 3.768707209272266e-06,
      "loss": 0.80939001,
      "num_input_tokens_seen": 31652575,
      "step": 1497,
      "time_per_iteration": 3.7097339630126953
    },
    {
      "auxiliary_loss_clip": 0.01185112,
      "auxiliary_loss_mlp": 0.01034492,
      "balance_loss_clip": 1.05497885,
      "balance_loss_mlp": 1.02333403,
      "epoch": 0.18012385017735827,
      "flos": 18986330937600.0,
      "grad_norm": 2.30493783304258,
      "language_loss": 0.7696833,
      "learning_rate": 3.768343437425579e-06,
      "loss": 0.7918793,
      "num_input_tokens_seen": 31671145,
      "step": 1498,
      "time_per_iteration": 2.685978651046753
    },
    {
      "auxiliary_loss_clip": 0.01125444,
      "auxiliary_loss_mlp": 0.01040795,
      "balance_loss_clip": 1.05189931,
      "balance_loss_mlp": 1.02929139,
      "epoch": 0.18024409306799735,
      "flos": 19748103598080.0,
      "grad_norm": 2.785100245837132,
      "language_loss": 0.86319327,
      "learning_rate": 3.7679793973225987e-06,
      "loss": 0.88485563,
      "num_input_tokens_seen": 31686955,
      "step": 1499,
      "time_per_iteration": 2.784003496170044
    },
    {
      "auxiliary_loss_clip": 0.01071333,
      "auxiliary_loss_mlp": 0.01007019,
      "balance_loss_clip": 1.02887845,
      "balance_loss_mlp": 1.00417006,
      "epoch": 0.18036433595863643,
      "flos": 67227183060480.0,
      "grad_norm": 0.8802582100401489,
      "language_loss": 0.61638987,
      "learning_rate": 3.767615089018549e-06,
      "loss": 0.63717341,
      "num_input_tokens_seen": 31749300,
      "step": 1500,
      "time_per_iteration": 3.3119330406188965
    },
    {
      "auxiliary_loss_clip": 0.01184901,
      "auxiliary_loss_mlp": 0.01045123,
      "balance_loss_clip": 1.05675173,
      "balance_loss_mlp": 1.03363132,
      "epoch": 0.18048457884927555,
      "flos": 18181464935040.0,
      "grad_norm": 4.720955942267611,
      "language_loss": 0.85996592,
      "learning_rate": 3.7672505125686966e-06,
      "loss": 0.88226616,
      "num_input_tokens_seen": 31765665,
      "step": 1501,
      "time_per_iteration": 3.5245425701141357
    },
    {
      "auxiliary_loss_clip": 0.01159389,
      "auxiliary_loss_mlp": 0.01041649,
      "balance_loss_clip": 1.05420756,
      "balance_loss_mlp": 1.03018737,
      "epoch": 0.18060482173991463,
      "flos": 15813767111040.0,
      "grad_norm": 4.09909658600406,
      "language_loss": 0.84476846,
      "learning_rate": 3.7668856680283455e-06,
      "loss": 0.86677885,
      "num_input_tokens_seen": 31782690,
      "step": 1502,
      "time_per_iteration": 3.6115450859069824
    },
    {
      "auxiliary_loss_clip": 0.01199486,
      "auxiliary_loss_mlp": 0.01035956,
      "balance_loss_clip": 1.06021976,
      "balance_loss_mlp": 1.02517378,
      "epoch": 0.1807250646305537,
      "flos": 18587399512320.0,
      "grad_norm": 1.9679725395845071,
      "language_loss": 0.82734519,
      "learning_rate": 3.7665205554528437e-06,
      "loss": 0.84969962,
      "num_input_tokens_seen": 31802045,
      "step": 1503,
      "time_per_iteration": 3.576866388320923
    },
    {
      "auxiliary_loss_clip": 0.01190657,
      "auxiliary_loss_mlp": 0.01038424,
      "balance_loss_clip": 1.06180382,
      "balance_loss_mlp": 1.02703989,
      "epoch": 0.18084530752119282,
      "flos": 23149131880320.0,
      "grad_norm": 2.1429393656805775,
      "language_loss": 0.74432653,
      "learning_rate": 3.7661551748975782e-06,
      "loss": 0.76661736,
      "num_input_tokens_seen": 31820220,
      "step": 1504,
      "time_per_iteration": 2.7332825660705566
    },
    {
      "auxiliary_loss_clip": 0.01095017,
      "auxiliary_loss_mlp": 0.01004178,
      "balance_loss_clip": 1.02560365,
      "balance_loss_mlp": 1.00150764,
      "epoch": 0.1809655504118319,
      "flos": 59803153568640.0,
      "grad_norm": 0.8113945235702145,
      "language_loss": 0.60501766,
      "learning_rate": 3.7657895264179772e-06,
      "loss": 0.62600958,
      "num_input_tokens_seen": 31876195,
      "step": 1505,
      "time_per_iteration": 3.258424758911133
    },
    {
      "auxiliary_loss_clip": 0.01187805,
      "auxiliary_loss_mlp": 0.01039831,
      "balance_loss_clip": 1.05833924,
      "balance_loss_mlp": 1.02833903,
      "epoch": 0.181085793302471,
      "flos": 44201941188480.0,
      "grad_norm": 1.9671106367414704,
      "language_loss": 0.74751341,
      "learning_rate": 3.765423610069509e-06,
      "loss": 0.76978976,
      "num_input_tokens_seen": 31901585,
      "step": 1506,
      "time_per_iteration": 2.9830660820007324
    },
    {
      "auxiliary_loss_clip": 0.01195583,
      "auxiliary_loss_mlp": 0.01039337,
      "balance_loss_clip": 1.06159282,
      "balance_loss_mlp": 1.02775586,
      "epoch": 0.18120603619311007,
      "flos": 34898384638080.0,
      "grad_norm": 1.8856338638125494,
      "language_loss": 0.73137265,
      "learning_rate": 3.765057425907683e-06,
      "loss": 0.75372183,
      "num_input_tokens_seen": 31923045,
      "step": 1507,
      "time_per_iteration": 2.8265762329101562
    },
    {
      "auxiliary_loss_clip": 0.01210152,
      "auxiliary_loss_mlp": 0.01035303,
      "balance_loss_clip": 1.05947614,
      "balance_loss_mlp": 1.02478266,
      "epoch": 0.18132627908374918,
      "flos": 21506757390720.0,
      "grad_norm": 1.8928962531001756,
      "language_loss": 0.78825259,
      "learning_rate": 3.764690973988048e-06,
      "loss": 0.81070709,
      "num_input_tokens_seen": 31943385,
      "step": 1508,
      "time_per_iteration": 2.6553759574890137
    },
    {
      "auxiliary_loss_clip": 0.01180872,
      "auxiliary_loss_mlp": 0.01040105,
      "balance_loss_clip": 1.05789423,
      "balance_loss_mlp": 1.02956736,
      "epoch": 0.18144652197438826,
      "flos": 29057693633280.0,
      "grad_norm": 2.0271499772919004,
      "language_loss": 0.74016738,
      "learning_rate": 3.7643242543661967e-06,
      "loss": 0.76237714,
      "num_input_tokens_seen": 31966045,
      "step": 1509,
      "time_per_iteration": 2.767719030380249
    },
    {
      "auxiliary_loss_clip": 0.01092667,
      "auxiliary_loss_mlp": 0.0100254,
      "balance_loss_clip": 1.02933884,
      "balance_loss_mlp": 0.99984568,
      "epoch": 0.18156676486502735,
      "flos": 68675064382080.0,
      "grad_norm": 0.815735386778952,
      "language_loss": 0.60423523,
      "learning_rate": 3.7639572670977573e-06,
      "loss": 0.62518728,
      "num_input_tokens_seen": 32021540,
      "step": 1510,
      "time_per_iteration": 3.145909547805786
    },
    {
      "auxiliary_loss_clip": 0.01179225,
      "auxiliary_loss_mlp": 0.01032884,
      "balance_loss_clip": 1.05631661,
      "balance_loss_mlp": 1.02303767,
      "epoch": 0.18168700775566646,
      "flos": 26471515334400.0,
      "grad_norm": 2.269369433723992,
      "language_loss": 0.76507246,
      "learning_rate": 3.7635900122384042e-06,
      "loss": 0.78719354,
      "num_input_tokens_seen": 32044535,
      "step": 1511,
      "time_per_iteration": 2.770695447921753
    },
    {
      "auxiliary_loss_clip": 0.01199179,
      "auxiliary_loss_mlp": 0.01043718,
      "balance_loss_clip": 1.06047273,
      "balance_loss_mlp": 1.03192186,
      "epoch": 0.18180725064630554,
      "flos": 15005668884480.0,
      "grad_norm": 2.2065914506343343,
      "language_loss": 0.86825395,
      "learning_rate": 3.7632224898438477e-06,
      "loss": 0.89068294,
      "num_input_tokens_seen": 32061010,
      "step": 1512,
      "time_per_iteration": 2.642801523208618
    },
    {
      "auxiliary_loss_clip": 0.01177396,
      "auxiliary_loss_mlp": 0.01036003,
      "balance_loss_clip": 1.0540669,
      "balance_loss_mlp": 1.02597713,
      "epoch": 0.18192749353694462,
      "flos": 19682387665920.0,
      "grad_norm": 3.6415886196113703,
      "language_loss": 0.79210877,
      "learning_rate": 3.762854699969842e-06,
      "loss": 0.81424278,
      "num_input_tokens_seen": 32081520,
      "step": 1513,
      "time_per_iteration": 2.7427663803100586
    },
    {
      "auxiliary_loss_clip": 0.01205837,
      "auxiliary_loss_mlp": 0.01034402,
      "balance_loss_clip": 1.06225657,
      "balance_loss_mlp": 1.02408433,
      "epoch": 0.1820477364275837,
      "flos": 20702717400960.0,
      "grad_norm": 1.9737186271089855,
      "language_loss": 0.73094189,
      "learning_rate": 3.762486642672179e-06,
      "loss": 0.7533443,
      "num_input_tokens_seen": 32098460,
      "step": 1514,
      "time_per_iteration": 2.6408891677856445
    },
    {
      "auxiliary_loss_clip": 0.01197735,
      "auxiliary_loss_mlp": 0.01032479,
      "balance_loss_clip": 1.06167412,
      "balance_loss_mlp": 1.02184582,
      "epoch": 0.18216797931822282,
      "flos": 17128708197120.0,
      "grad_norm": 2.1596216626296076,
      "language_loss": 0.87096071,
      "learning_rate": 3.7621183180066946e-06,
      "loss": 0.89326286,
      "num_input_tokens_seen": 32116420,
      "step": 1515,
      "time_per_iteration": 2.679025888442993
    },
    {
      "auxiliary_loss_clip": 0.01186124,
      "auxiliary_loss_mlp": 0.01035362,
      "balance_loss_clip": 1.05201066,
      "balance_loss_mlp": 1.02447271,
      "epoch": 0.1822882222088619,
      "flos": 29242561956480.0,
      "grad_norm": 1.8549359114350574,
      "language_loss": 0.74037498,
      "learning_rate": 3.7617497260292625e-06,
      "loss": 0.76258981,
      "num_input_tokens_seen": 32138475,
      "step": 1516,
      "time_per_iteration": 2.7294561862945557
    },
    {
      "auxiliary_loss_clip": 0.01188572,
      "auxiliary_loss_mlp": 0.0104525,
      "balance_loss_clip": 1.05934072,
      "balance_loss_mlp": 1.03350782,
      "epoch": 0.18240846509950098,
      "flos": 17702739446400.0,
      "grad_norm": 3.2758223001493367,
      "language_loss": 0.7857042,
      "learning_rate": 3.7613808667957967e-06,
      "loss": 0.80804235,
      "num_input_tokens_seen": 32151165,
      "step": 1517,
      "time_per_iteration": 2.6694765090942383
    },
    {
      "auxiliary_loss_clip": 0.01194802,
      "auxiliary_loss_mlp": 0.01037697,
      "balance_loss_clip": 1.05845201,
      "balance_loss_mlp": 1.02621698,
      "epoch": 0.1825287079901401,
      "flos": 14790025584000.0,
      "grad_norm": 2.0204416511851178,
      "language_loss": 0.91021287,
      "learning_rate": 3.7610117403622547e-06,
      "loss": 0.93253791,
      "num_input_tokens_seen": 32167725,
      "step": 1518,
      "time_per_iteration": 2.6257219314575195
    },
    {
      "auxiliary_loss_clip": 0.01165623,
      "auxiliary_loss_mlp": 0.01033741,
      "balance_loss_clip": 1.05126762,
      "balance_loss_mlp": 1.02327466,
      "epoch": 0.18264895088077918,
      "flos": 21946232292480.0,
      "grad_norm": 1.6522070271233864,
      "language_loss": 0.90037608,
      "learning_rate": 3.7606423467846313e-06,
      "loss": 0.92236972,
      "num_input_tokens_seen": 32187330,
      "step": 1519,
      "time_per_iteration": 2.7517755031585693
    },
    {
      "auxiliary_loss_clip": 0.01185996,
      "auxiliary_loss_mlp": 0.01038386,
      "balance_loss_clip": 1.06087494,
      "balance_loss_mlp": 1.02791953,
      "epoch": 0.18276919377141826,
      "flos": 20886759711360.0,
      "grad_norm": 2.0708404567846137,
      "language_loss": 0.79767936,
      "learning_rate": 3.760272686118964e-06,
      "loss": 0.81992316,
      "num_input_tokens_seen": 32205550,
      "step": 1520,
      "time_per_iteration": 2.7002108097076416
    },
    {
      "auxiliary_loss_clip": 0.01192408,
      "auxiliary_loss_mlp": 0.01037946,
      "balance_loss_clip": 1.05708814,
      "balance_loss_mlp": 1.02709842,
      "epoch": 0.18288943666205737,
      "flos": 21469877101440.0,
      "grad_norm": 2.24835767019247,
      "language_loss": 0.92644173,
      "learning_rate": 3.7599027584213297e-06,
      "loss": 0.94874531,
      "num_input_tokens_seen": 32224430,
      "step": 1521,
      "time_per_iteration": 2.7223899364471436
    },
    {
      "auxiliary_loss_clip": 0.01210897,
      "auxiliary_loss_mlp": 0.01040177,
      "balance_loss_clip": 1.05953503,
      "balance_loss_mlp": 1.02923357,
      "epoch": 0.18300967955269645,
      "flos": 21539363961600.0,
      "grad_norm": 2.1166431969046267,
      "language_loss": 0.78154182,
      "learning_rate": 3.7595325637478465e-06,
      "loss": 0.80405259,
      "num_input_tokens_seen": 32242455,
      "step": 1522,
      "time_per_iteration": 2.650888204574585
    },
    {
      "auxiliary_loss_clip": 0.01185476,
      "auxiliary_loss_mlp": 0.01045204,
      "balance_loss_clip": 1.06077313,
      "balance_loss_mlp": 1.03399825,
      "epoch": 0.18312992244333554,
      "flos": 28876237102080.0,
      "grad_norm": 1.8451627073144232,
      "language_loss": 0.81564701,
      "learning_rate": 3.7591621021546723e-06,
      "loss": 0.83795381,
      "num_input_tokens_seen": 32264450,
      "step": 1523,
      "time_per_iteration": 2.71543025970459
    },
    {
      "auxiliary_loss_clip": 0.01198556,
      "auxiliary_loss_mlp": 0.01042405,
      "balance_loss_clip": 1.05639613,
      "balance_loss_mlp": 1.03150368,
      "epoch": 0.18325016533397462,
      "flos": 20120102801280.0,
      "grad_norm": 2.006656340185675,
      "language_loss": 0.81693959,
      "learning_rate": 3.7587913736980062e-06,
      "loss": 0.83934915,
      "num_input_tokens_seen": 32284090,
      "step": 1524,
      "time_per_iteration": 3.5757951736450195
    },
    {
      "auxiliary_loss_clip": 0.0113251,
      "auxiliary_loss_mlp": 0.01043693,
      "balance_loss_clip": 1.04866862,
      "balance_loss_mlp": 1.03220153,
      "epoch": 0.18337040822461373,
      "flos": 23329187781120.0,
      "grad_norm": 1.628198307359135,
      "language_loss": 0.84508044,
      "learning_rate": 3.7584203784340865e-06,
      "loss": 0.86684251,
      "num_input_tokens_seen": 32303260,
      "step": 1525,
      "time_per_iteration": 2.761080265045166
    },
    {
      "auxiliary_loss_clip": 0.01187347,
      "auxiliary_loss_mlp": 0.01035232,
      "balance_loss_clip": 1.05584908,
      "balance_loss_mlp": 1.0243777,
      "epoch": 0.1834906511152528,
      "flos": 25009555881600.0,
      "grad_norm": 2.305725598596235,
      "language_loss": 0.85479438,
      "learning_rate": 3.7580491164191938e-06,
      "loss": 0.87702012,
      "num_input_tokens_seen": 32321570,
      "step": 1526,
      "time_per_iteration": 2.6950886249542236
    },
    {
      "auxiliary_loss_clip": 0.01108378,
      "auxiliary_loss_mlp": 0.01003146,
      "balance_loss_clip": 1.02621531,
      "balance_loss_mlp": 1.00057113,
      "epoch": 0.1836108940058919,
      "flos": 67251493589760.0,
      "grad_norm": 0.753394429682065,
      "language_loss": 0.61300087,
      "learning_rate": 3.757677587709648e-06,
      "loss": 0.63411605,
      "num_input_tokens_seen": 32384835,
      "step": 1527,
      "time_per_iteration": 4.208168029785156
    },
    {
      "auxiliary_loss_clip": 0.01171193,
      "auxiliary_loss_mlp": 0.0105292,
      "balance_loss_clip": 1.05639935,
      "balance_loss_mlp": 1.04199457,
      "epoch": 0.183731136896531,
      "flos": 25738721971200.0,
      "grad_norm": 2.7629933578772383,
      "language_loss": 0.75684947,
      "learning_rate": 3.7573057923618095e-06,
      "loss": 0.77909058,
      "num_input_tokens_seen": 32404930,
      "step": 1528,
      "time_per_iteration": 4.504167318344116
    },
    {
      "auxiliary_loss_clip": 0.01158715,
      "auxiliary_loss_mlp": 0.01032511,
      "balance_loss_clip": 1.05071008,
      "balance_loss_mlp": 1.02214611,
      "epoch": 0.1838513797871701,
      "flos": 20449403712000.0,
      "grad_norm": 2.2030850904400006,
      "language_loss": 0.74380654,
      "learning_rate": 3.7569337304320793e-06,
      "loss": 0.76571882,
      "num_input_tokens_seen": 32424515,
      "step": 1529,
      "time_per_iteration": 2.7050156593322754
    },
    {
      "auxiliary_loss_clip": 0.01090254,
      "auxiliary_loss_mlp": 0.0100435,
      "balance_loss_clip": 1.02225065,
      "balance_loss_mlp": 1.0017159,
      "epoch": 0.18397162267780917,
      "flos": 68565141786240.0,
      "grad_norm": 0.834205629065132,
      "language_loss": 0.64479637,
      "learning_rate": 3.756561401976899e-06,
      "loss": 0.6657424,
      "num_input_tokens_seen": 32484220,
      "step": 1530,
      "time_per_iteration": 3.1571104526519775
    },
    {
      "auxiliary_loss_clip": 0.01224498,
      "auxiliary_loss_mlp": 0.0103767,
      "balance_loss_clip": 1.06284559,
      "balance_loss_mlp": 1.02692962,
      "epoch": 0.18409186556844825,
      "flos": 31941104976000.0,
      "grad_norm": 2.370360550615901,
      "language_loss": 0.82795393,
      "learning_rate": 3.7561888070527514e-06,
      "loss": 0.85057563,
      "num_input_tokens_seen": 32506260,
      "step": 1531,
      "time_per_iteration": 2.6579456329345703
    },
    {
      "auxiliary_loss_clip": 0.01163816,
      "auxiliary_loss_mlp": 0.00765957,
      "balance_loss_clip": 1.05538225,
      "balance_loss_mlp": 1.00040245,
      "epoch": 0.18421210845908736,
      "flos": 20120533764480.0,
      "grad_norm": 2.2651025913241933,
      "language_loss": 0.80405438,
      "learning_rate": 3.7558159457161577e-06,
      "loss": 0.8233521,
      "num_input_tokens_seen": 32524225,
      "step": 1532,
      "time_per_iteration": 2.703951120376587
    },
    {
      "auxiliary_loss_clip": 0.01189495,
      "auxiliary_loss_mlp": 0.00765644,
      "balance_loss_clip": 1.05694175,
      "balance_loss_mlp": 1.00048661,
      "epoch": 0.18433235134972645,
      "flos": 23110491824640.0,
      "grad_norm": 3.1661083724776273,
      "language_loss": 0.78293765,
      "learning_rate": 3.755442818023681e-06,
      "loss": 0.80248904,
      "num_input_tokens_seen": 32543850,
      "step": 1533,
      "time_per_iteration": 2.690779685974121
    },
    {
      "auxiliary_loss_clip": 0.01173701,
      "auxiliary_loss_mlp": 0.01033258,
      "balance_loss_clip": 1.05628777,
      "balance_loss_mlp": 1.02308321,
      "epoch": 0.18445259424036553,
      "flos": 18291351617280.0,
      "grad_norm": 1.8765192831108266,
      "language_loss": 0.75920582,
      "learning_rate": 3.7550694240319246e-06,
      "loss": 0.78127539,
      "num_input_tokens_seen": 32561725,
      "step": 1534,
      "time_per_iteration": 2.7708969116210938
    },
    {
      "auxiliary_loss_clip": 0.01208527,
      "auxiliary_loss_mlp": 0.01034327,
      "balance_loss_clip": 1.05859756,
      "balance_loss_mlp": 1.02318668,
      "epoch": 0.18457283713100464,
      "flos": 21324079797120.0,
      "grad_norm": 2.031560077333653,
      "language_loss": 0.76692456,
      "learning_rate": 3.7546957637975326e-06,
      "loss": 0.78935307,
      "num_input_tokens_seen": 32579135,
      "step": 1535,
      "time_per_iteration": 2.6517465114593506
    },
    {
      "auxiliary_loss_clip": 0.01132797,
      "auxiliary_loss_mlp": 0.01035702,
      "balance_loss_clip": 1.04492617,
      "balance_loss_mlp": 1.02594435,
      "epoch": 0.18469308002164372,
      "flos": 20375679047040.0,
      "grad_norm": 1.4989304484593549,
      "language_loss": 0.74119532,
      "learning_rate": 3.7543218373771873e-06,
      "loss": 0.76288033,
      "num_input_tokens_seen": 32598460,
      "step": 1536,
      "time_per_iteration": 2.7783422470092773
    },
    {
      "auxiliary_loss_clip": 0.01136601,
      "auxiliary_loss_mlp": 0.00765261,
      "balance_loss_clip": 1.049263,
      "balance_loss_mlp": 1.0003624,
      "epoch": 0.1848133229122828,
      "flos": 26435892021120.0,
      "grad_norm": 1.405667770186295,
      "language_loss": 0.78065598,
      "learning_rate": 3.753947644827615e-06,
      "loss": 0.79967463,
      "num_input_tokens_seen": 32621920,
      "step": 1537,
      "time_per_iteration": 2.8490183353424072
    },
    {
      "auxiliary_loss_clip": 0.01092945,
      "auxiliary_loss_mlp": 0.01004962,
      "balance_loss_clip": 1.02077401,
      "balance_loss_mlp": 1.0023396,
      "epoch": 0.1849335658029219,
      "flos": 70547447612160.0,
      "grad_norm": 0.9478098100613206,
      "language_loss": 0.5716722,
      "learning_rate": 3.753573186205579e-06,
      "loss": 0.59265125,
      "num_input_tokens_seen": 32690040,
      "step": 1538,
      "time_per_iteration": 3.357631206512451
    },
    {
      "auxiliary_loss_clip": 0.01179199,
      "auxiliary_loss_mlp": 0.00765653,
      "balance_loss_clip": 1.05256522,
      "balance_loss_mlp": 1.00034702,
      "epoch": 0.185053808693561,
      "flos": 17384140788480.0,
      "grad_norm": 2.293041159548234,
      "language_loss": 0.7808972,
      "learning_rate": 3.753198461567885e-06,
      "loss": 0.80034578,
      "num_input_tokens_seen": 32707285,
      "step": 1539,
      "time_per_iteration": 2.7164037227630615
    },
    {
      "auxiliary_loss_clip": 0.01168278,
      "auxiliary_loss_mlp": 0.01035678,
      "balance_loss_clip": 1.05676031,
      "balance_loss_mlp": 1.02593303,
      "epoch": 0.18517405158420008,
      "flos": 28986159697920.0,
      "grad_norm": 2.3343906622686115,
      "language_loss": 0.92079973,
      "learning_rate": 3.7528234709713783e-06,
      "loss": 0.94283932,
      "num_input_tokens_seen": 32730030,
      "step": 1540,
      "time_per_iteration": 2.754969835281372
    },
    {
      "auxiliary_loss_clip": 0.01207359,
      "auxiliary_loss_mlp": 0.01042898,
      "balance_loss_clip": 1.05953884,
      "balance_loss_mlp": 1.03229988,
      "epoch": 0.18529429447483917,
      "flos": 26794962328320.0,
      "grad_norm": 2.2285643729722455,
      "language_loss": 0.8446945,
      "learning_rate": 3.7524482144729447e-06,
      "loss": 0.86719698,
      "num_input_tokens_seen": 32749485,
      "step": 1541,
      "time_per_iteration": 2.727708339691162
    },
    {
      "auxiliary_loss_clip": 0.01166277,
      "auxiliary_loss_mlp": 0.01044505,
      "balance_loss_clip": 1.05021858,
      "balance_loss_mlp": 1.03355527,
      "epoch": 0.18541453736547828,
      "flos": 13581595301760.0,
      "grad_norm": 2.0523139261393357,
      "language_loss": 0.83921075,
      "learning_rate": 3.7520726921295106e-06,
      "loss": 0.86131859,
      "num_input_tokens_seen": 32766205,
      "step": 1542,
      "time_per_iteration": 2.719616174697876
    },
    {
      "auxiliary_loss_clip": 0.01200944,
      "auxiliary_loss_mlp": 0.01039634,
      "balance_loss_clip": 1.05758786,
      "balance_loss_mlp": 1.02935207,
      "epoch": 0.18553478025611736,
      "flos": 24025424077440.0,
      "grad_norm": 2.7878874223455066,
      "language_loss": 0.72619563,
      "learning_rate": 3.751696903998042e-06,
      "loss": 0.74860144,
      "num_input_tokens_seen": 32784840,
      "step": 1543,
      "time_per_iteration": 2.6694929599761963
    },
    {
      "auxiliary_loss_clip": 0.01203687,
      "auxiliary_loss_mlp": 0.01045195,
      "balance_loss_clip": 1.06130791,
      "balance_loss_mlp": 1.03419209,
      "epoch": 0.18565502314675644,
      "flos": 25885165720320.0,
      "grad_norm": 1.8064544834521932,
      "language_loss": 0.7034775,
      "learning_rate": 3.7513208501355456e-06,
      "loss": 0.72596633,
      "num_input_tokens_seen": 32805945,
      "step": 1544,
      "time_per_iteration": 2.730578660964966
    },
    {
      "auxiliary_loss_clip": 0.01186007,
      "auxiliary_loss_mlp": 0.01036003,
      "balance_loss_clip": 1.05677509,
      "balance_loss_mlp": 1.02549517,
      "epoch": 0.18577526603739553,
      "flos": 19610063631360.0,
      "grad_norm": 1.8913219813158655,
      "language_loss": 0.83862758,
      "learning_rate": 3.750944530599069e-06,
      "loss": 0.86084765,
      "num_input_tokens_seen": 32825515,
      "step": 1545,
      "time_per_iteration": 2.6332736015319824
    },
    {
      "auxiliary_loss_clip": 0.0121367,
      "auxiliary_loss_mlp": 0.01038006,
      "balance_loss_clip": 1.06261134,
      "balance_loss_mlp": 1.02702105,
      "epoch": 0.18589550892803464,
      "flos": 18474891137280.0,
      "grad_norm": 2.638061196974815,
      "language_loss": 0.81366014,
      "learning_rate": 3.7505679454456992e-06,
      "loss": 0.83617693,
      "num_input_tokens_seen": 32842125,
      "step": 1546,
      "time_per_iteration": 2.647264003753662
    },
    {
      "auxiliary_loss_clip": 0.01125197,
      "auxiliary_loss_mlp": 0.01038666,
      "balance_loss_clip": 1.04923642,
      "balance_loss_mlp": 1.02783608,
      "epoch": 0.18601575181867372,
      "flos": 23549966726400.0,
      "grad_norm": 1.863593080330065,
      "language_loss": 0.70353127,
      "learning_rate": 3.750191094732564e-06,
      "loss": 0.72516984,
      "num_input_tokens_seen": 32862990,
      "step": 1547,
      "time_per_iteration": 2.80033278465271
    },
    {
      "auxiliary_loss_clip": 0.01124016,
      "auxiliary_loss_mlp": 0.00765497,
      "balance_loss_clip": 1.04614568,
      "balance_loss_mlp": 1.0004034,
      "epoch": 0.1861359947093128,
      "flos": 26360192108160.0,
      "grad_norm": 2.0613546795768576,
      "language_loss": 0.75225663,
      "learning_rate": 3.7498139785168313e-06,
      "loss": 0.77115178,
      "num_input_tokens_seen": 32883595,
      "step": 1548,
      "time_per_iteration": 2.8836395740509033
    },
    {
      "auxiliary_loss_clip": 0.01203837,
      "auxiliary_loss_mlp": 0.01036111,
      "balance_loss_clip": 1.05984747,
      "balance_loss_mlp": 1.02578807,
      "epoch": 0.1862562375999519,
      "flos": 23331198942720.0,
      "grad_norm": 1.6620364607766418,
      "language_loss": 0.775769,
      "learning_rate": 3.749436596855709e-06,
      "loss": 0.79816842,
      "num_input_tokens_seen": 32902895,
      "step": 1549,
      "time_per_iteration": 2.713435411453247
    },
    {
      "auxiliary_loss_clip": 0.01199839,
      "auxiliary_loss_mlp": 0.01043537,
      "balance_loss_clip": 1.05910087,
      "balance_loss_mlp": 1.03218269,
      "epoch": 0.186376480490591,
      "flos": 16648222942080.0,
      "grad_norm": 2.1044541381393436,
      "language_loss": 0.90375984,
      "learning_rate": 3.749058949806446e-06,
      "loss": 0.92619359,
      "num_input_tokens_seen": 32919620,
      "step": 1550,
      "time_per_iteration": 3.5943896770477295
    },
    {
      "auxiliary_loss_clip": 0.01204935,
      "auxiliary_loss_mlp": 0.01043959,
      "balance_loss_clip": 1.05878937,
      "balance_loss_mlp": 1.03376043,
      "epoch": 0.18649672338123008,
      "flos": 21468656039040.0,
      "grad_norm": 1.8869258175938082,
      "language_loss": 0.84575939,
      "learning_rate": 3.748681037426331e-06,
      "loss": 0.86824834,
      "num_input_tokens_seen": 32938830,
      "step": 1551,
      "time_per_iteration": 2.675835609436035
    },
    {
      "auxiliary_loss_clip": 0.01222203,
      "auxiliary_loss_mlp": 0.0103791,
      "balance_loss_clip": 1.06256175,
      "balance_loss_mlp": 1.02635837,
      "epoch": 0.1866169662718692,
      "flos": 12312728386560.0,
      "grad_norm": 2.138547454651232,
      "language_loss": 0.9206028,
      "learning_rate": 3.7483028597726936e-06,
      "loss": 0.94320393,
      "num_input_tokens_seen": 32955600,
      "step": 1552,
      "time_per_iteration": 2.5795977115631104
    },
    {
      "auxiliary_loss_clip": 0.01177532,
      "auxiliary_loss_mlp": 0.0103912,
      "balance_loss_clip": 1.05959272,
      "balance_loss_mlp": 1.02797341,
      "epoch": 0.18673720916250827,
      "flos": 23581280407680.0,
      "grad_norm": 1.7485933997125052,
      "language_loss": 0.62804615,
      "learning_rate": 3.7479244169029017e-06,
      "loss": 0.65021276,
      "num_input_tokens_seen": 32975390,
      "step": 1553,
      "time_per_iteration": 3.667445421218872
    },
    {
      "auxiliary_loss_clip": 0.01205208,
      "auxiliary_loss_mlp": 0.01024101,
      "balance_loss_clip": 1.05507171,
      "balance_loss_mlp": 1.01315725,
      "epoch": 0.18685745205314735,
      "flos": 19718370115200.0,
      "grad_norm": 2.2766063790769855,
      "language_loss": 0.73617238,
      "learning_rate": 3.7475457088743658e-06,
      "loss": 0.75846547,
      "num_input_tokens_seen": 32992640,
      "step": 1554,
      "time_per_iteration": 4.511739492416382
    },
    {
      "auxiliary_loss_clip": 0.01181066,
      "auxiliary_loss_mlp": 0.01032778,
      "balance_loss_clip": 1.05635643,
      "balance_loss_mlp": 1.0224309,
      "epoch": 0.18697769494378644,
      "flos": 34204123589760.0,
      "grad_norm": 2.1464264030806066,
      "language_loss": 0.74841517,
      "learning_rate": 3.7471667357445348e-06,
      "loss": 0.77055365,
      "num_input_tokens_seen": 33012470,
      "step": 1555,
      "time_per_iteration": 2.8026514053344727
    },
    {
      "auxiliary_loss_clip": 0.01149765,
      "auxiliary_loss_mlp": 0.01037529,
      "balance_loss_clip": 1.05454826,
      "balance_loss_mlp": 1.02707982,
      "epoch": 0.18709793783442555,
      "flos": 34241327101440.0,
      "grad_norm": 2.1731735162998476,
      "language_loss": 0.72486848,
      "learning_rate": 3.7467874975709e-06,
      "loss": 0.74674141,
      "num_input_tokens_seen": 33033275,
      "step": 1556,
      "time_per_iteration": 2.9177565574645996
    },
    {
      "auxiliary_loss_clip": 0.01211216,
      "auxiliary_loss_mlp": 0.01041531,
      "balance_loss_clip": 1.06325042,
      "balance_loss_mlp": 1.03126121,
      "epoch": 0.18721818072506463,
      "flos": 40734550529280.0,
      "grad_norm": 2.049736313733444,
      "language_loss": 0.7841152,
      "learning_rate": 3.7464079944109904e-06,
      "loss": 0.80664271,
      "num_input_tokens_seen": 33055135,
      "step": 1557,
      "time_per_iteration": 2.792910575866699
    },
    {
      "auxiliary_loss_clip": 0.01177893,
      "auxiliary_loss_mlp": 0.01035192,
      "balance_loss_clip": 1.0555048,
      "balance_loss_mlp": 1.0245347,
      "epoch": 0.18733842361570371,
      "flos": 22157386392960.0,
      "grad_norm": 1.9534016767966933,
      "language_loss": 0.772452,
      "learning_rate": 3.746028226322376e-06,
      "loss": 0.79458284,
      "num_input_tokens_seen": 33071015,
      "step": 1558,
      "time_per_iteration": 2.7582898139953613
    },
    {
      "auxiliary_loss_clip": 0.01187852,
      "auxiliary_loss_mlp": 0.01034534,
      "balance_loss_clip": 1.05721664,
      "balance_loss_mlp": 1.0240612,
      "epoch": 0.18745866650634282,
      "flos": 18914940656640.0,
      "grad_norm": 2.0698175452531005,
      "language_loss": 0.75213742,
      "learning_rate": 3.745648193362669e-06,
      "loss": 0.77436125,
      "num_input_tokens_seen": 33090370,
      "step": 1559,
      "time_per_iteration": 2.752578020095825
    },
    {
      "auxiliary_loss_clip": 0.01187528,
      "auxiliary_loss_mlp": 0.01038606,
      "balance_loss_clip": 1.05856013,
      "balance_loss_mlp": 1.02921855,
      "epoch": 0.1875789093969819,
      "flos": 19314626267520.0,
      "grad_norm": 1.9585756823152098,
      "language_loss": 0.7227543,
      "learning_rate": 3.745267895589518e-06,
      "loss": 0.74501562,
      "num_input_tokens_seen": 33108910,
      "step": 1560,
      "time_per_iteration": 2.6389782428741455
    },
    {
      "auxiliary_loss_clip": 0.01190118,
      "auxiliary_loss_mlp": 0.01038619,
      "balance_loss_clip": 1.05855727,
      "balance_loss_mlp": 1.0280925,
      "epoch": 0.187699152287621,
      "flos": 17018965169280.0,
      "grad_norm": 2.4000054849976897,
      "language_loss": 0.8227638,
      "learning_rate": 3.7448873330606154e-06,
      "loss": 0.84505117,
      "num_input_tokens_seen": 33126680,
      "step": 1561,
      "time_per_iteration": 2.676196575164795
    },
    {
      "auxiliary_loss_clip": 0.01165676,
      "auxiliary_loss_mlp": 0.01040402,
      "balance_loss_clip": 1.05524015,
      "balance_loss_mlp": 1.02898169,
      "epoch": 0.18781939517826007,
      "flos": 22346384780160.0,
      "grad_norm": 1.9902395232447103,
      "language_loss": 0.87343127,
      "learning_rate": 3.7445065058336914e-06,
      "loss": 0.89549208,
      "num_input_tokens_seen": 33145550,
      "step": 1562,
      "time_per_iteration": 2.705169439315796
    },
    {
      "auxiliary_loss_clip": 0.01147191,
      "auxiliary_loss_mlp": 0.01036849,
      "balance_loss_clip": 1.05046642,
      "balance_loss_mlp": 1.02679372,
      "epoch": 0.18793963806889918,
      "flos": 14611478054400.0,
      "grad_norm": 1.909371913225208,
      "language_loss": 0.86465353,
      "learning_rate": 3.7441254139665176e-06,
      "loss": 0.88649392,
      "num_input_tokens_seen": 33161735,
      "step": 1563,
      "time_per_iteration": 2.7222046852111816
    },
    {
      "auxiliary_loss_clip": 0.01223611,
      "auxiliary_loss_mlp": 0.01032982,
      "balance_loss_clip": 1.06577659,
      "balance_loss_mlp": 1.02301657,
      "epoch": 0.18805988095953827,
      "flos": 17457075354240.0,
      "grad_norm": 1.6763679024808564,
      "language_loss": 0.82419741,
      "learning_rate": 3.743744057516905e-06,
      "loss": 0.84676337,
      "num_input_tokens_seen": 33179795,
      "step": 1564,
      "time_per_iteration": 2.580401659011841
    },
    {
      "auxiliary_loss_clip": 0.01160257,
      "auxiliary_loss_mlp": 0.01037641,
      "balance_loss_clip": 1.05504799,
      "balance_loss_mlp": 1.02694821,
      "epoch": 0.18818012385017735,
      "flos": 15043877976960.0,
      "grad_norm": 5.793394314832816,
      "language_loss": 0.87681663,
      "learning_rate": 3.743362436542706e-06,
      "loss": 0.8987956,
      "num_input_tokens_seen": 33194485,
      "step": 1565,
      "time_per_iteration": 2.7685439586639404
    },
    {
      "auxiliary_loss_clip": 0.012181,
      "auxiliary_loss_mlp": 0.01035666,
      "balance_loss_clip": 1.06113791,
      "balance_loss_mlp": 1.02571225,
      "epoch": 0.18830036674081646,
      "flos": 47551975136640.0,
      "grad_norm": 2.1582880945463363,
      "language_loss": 0.76736325,
      "learning_rate": 3.7429805511018115e-06,
      "loss": 0.7899009,
      "num_input_tokens_seen": 33216145,
      "step": 1566,
      "time_per_iteration": 2.832289695739746
    },
    {
      "auxiliary_loss_clip": 0.01173628,
      "auxiliary_loss_mlp": 0.00764729,
      "balance_loss_clip": 1.0582372,
      "balance_loss_mlp": 1.00038528,
      "epoch": 0.18842060963145554,
      "flos": 30044626698240.0,
      "grad_norm": 2.216693082363248,
      "language_loss": 0.78352576,
      "learning_rate": 3.7425984012521524e-06,
      "loss": 0.80290937,
      "num_input_tokens_seen": 33236345,
      "step": 1567,
      "time_per_iteration": 2.8060951232910156
    },
    {
      "auxiliary_loss_clip": 0.01073331,
      "auxiliary_loss_mlp": 0.00756162,
      "balance_loss_clip": 1.02222025,
      "balance_loss_mlp": 1.00105119,
      "epoch": 0.18854085252209463,
      "flos": 70318372625280.0,
      "grad_norm": 0.7425552400340839,
      "language_loss": 0.60485101,
      "learning_rate": 3.7422159870517025e-06,
      "loss": 0.62314594,
      "num_input_tokens_seen": 33301600,
      "step": 1568,
      "time_per_iteration": 3.280465602874756
    },
    {
      "auxiliary_loss_clip": 0.01185657,
      "auxiliary_loss_mlp": 0.01035624,
      "balance_loss_clip": 1.05711293,
      "balance_loss_mlp": 1.02600336,
      "epoch": 0.1886610954127337,
      "flos": 21289318410240.0,
      "grad_norm": 1.8437193005744055,
      "language_loss": 0.78954947,
      "learning_rate": 3.7418333085584717e-06,
      "loss": 0.81176227,
      "num_input_tokens_seen": 33322785,
      "step": 1569,
      "time_per_iteration": 2.7407495975494385
    },
    {
      "auxiliary_loss_clip": 0.0117684,
      "auxiliary_loss_mlp": 0.01035543,
      "balance_loss_clip": 1.05848289,
      "balance_loss_mlp": 1.02524924,
      "epoch": 0.18878133830337282,
      "flos": 17266819991040.0,
      "grad_norm": 2.1849065105825303,
      "language_loss": 0.90973103,
      "learning_rate": 3.7414503658305128e-06,
      "loss": 0.93185484,
      "num_input_tokens_seen": 33340020,
      "step": 1570,
      "time_per_iteration": 2.7343006134033203
    },
    {
      "auxiliary_loss_clip": 0.01166993,
      "auxiliary_loss_mlp": 0.01038624,
      "balance_loss_clip": 1.05254006,
      "balance_loss_mlp": 1.02765059,
      "epoch": 0.1889015811940119,
      "flos": 25775207210880.0,
      "grad_norm": 2.3411829510861315,
      "language_loss": 0.77812481,
      "learning_rate": 3.7410671589259185e-06,
      "loss": 0.80018103,
      "num_input_tokens_seen": 33358620,
      "step": 1571,
      "time_per_iteration": 2.7734787464141846
    },
    {
      "auxiliary_loss_clip": 0.01220217,
      "auxiliary_loss_mlp": 0.01038748,
      "balance_loss_clip": 1.06310534,
      "balance_loss_mlp": 1.02805507,
      "epoch": 0.18902182408465099,
      "flos": 21032197879680.0,
      "grad_norm": 2.1706876567441715,
      "language_loss": 0.79648364,
      "learning_rate": 3.7406836879028205e-06,
      "loss": 0.81907332,
      "num_input_tokens_seen": 33378845,
      "step": 1572,
      "time_per_iteration": 2.6460819244384766
    },
    {
      "auxiliary_loss_clip": 0.01199143,
      "auxiliary_loss_mlp": 0.01034739,
      "balance_loss_clip": 1.05855775,
      "balance_loss_mlp": 1.02451706,
      "epoch": 0.1891420669752901,
      "flos": 22272121411200.0,
      "grad_norm": 4.345703891062653,
      "language_loss": 0.76719952,
      "learning_rate": 3.7402999528193907e-06,
      "loss": 0.78953826,
      "num_input_tokens_seen": 33398345,
      "step": 1573,
      "time_per_iteration": 2.629659414291382
    },
    {
      "auxiliary_loss_clip": 0.01162275,
      "auxiliary_loss_mlp": 0.00764961,
      "balance_loss_clip": 1.05411887,
      "balance_loss_mlp": 1.00051212,
      "epoch": 0.18926230986592918,
      "flos": 22017802141440.0,
      "grad_norm": 5.380342744314285,
      "language_loss": 0.8582077,
      "learning_rate": 3.739915953733842e-06,
      "loss": 0.87748003,
      "num_input_tokens_seen": 33416390,
      "step": 1574,
      "time_per_iteration": 2.768580198287964
    },
    {
      "auxiliary_loss_clip": 0.01217959,
      "auxiliary_loss_mlp": 0.01037479,
      "balance_loss_clip": 1.05993533,
      "balance_loss_mlp": 1.02704859,
      "epoch": 0.18938255275656826,
      "flos": 24462672336000.0,
      "grad_norm": 1.7624967846574133,
      "language_loss": 0.82347304,
      "learning_rate": 3.7395316907044264e-06,
      "loss": 0.84602743,
      "num_input_tokens_seen": 33437175,
      "step": 1575,
      "time_per_iteration": 3.5580410957336426
    },
    {
      "auxiliary_loss_clip": 0.01204102,
      "auxiliary_loss_mlp": 0.01035411,
      "balance_loss_clip": 1.06009257,
      "balance_loss_mlp": 1.02553463,
      "epoch": 0.18950279564720737,
      "flos": 24427049022720.0,
      "grad_norm": 1.6713884459230004,
      "language_loss": 0.79664993,
      "learning_rate": 3.7391471637894364e-06,
      "loss": 0.81904507,
      "num_input_tokens_seen": 33459440,
      "step": 1576,
      "time_per_iteration": 2.676105260848999
    },
    {
      "auxiliary_loss_clip": 0.01172574,
      "auxiliary_loss_mlp": 0.01032558,
      "balance_loss_clip": 1.05108213,
      "balance_loss_mlp": 1.02276516,
      "epoch": 0.18962303853784646,
      "flos": 19756291898880.0,
      "grad_norm": 2.007904152228609,
      "language_loss": 0.85011685,
      "learning_rate": 3.738762373047205e-06,
      "loss": 0.87216818,
      "num_input_tokens_seen": 33479360,
      "step": 1577,
      "time_per_iteration": 2.739616632461548
    },
    {
      "auxiliary_loss_clip": 0.01178191,
      "auxiliary_loss_mlp": 0.01038831,
      "balance_loss_clip": 1.05817544,
      "balance_loss_mlp": 1.02928281,
      "epoch": 0.18974328142848554,
      "flos": 21032054225280.0,
      "grad_norm": 1.874922164716666,
      "language_loss": 0.8353647,
      "learning_rate": 3.738377318536103e-06,
      "loss": 0.85753495,
      "num_input_tokens_seen": 33499245,
      "step": 1578,
      "time_per_iteration": 2.706434726715088
    },
    {
      "auxiliary_loss_clip": 0.01215787,
      "auxiliary_loss_mlp": 0.01030252,
      "balance_loss_clip": 1.06143451,
      "balance_loss_mlp": 1.0209949,
      "epoch": 0.18986352431912462,
      "flos": 12966122736000.0,
      "grad_norm": 2.0384051017475584,
      "language_loss": 0.71717292,
      "learning_rate": 3.7379920003145447e-06,
      "loss": 0.73963332,
      "num_input_tokens_seen": 33513520,
      "step": 1579,
      "time_per_iteration": 3.548917531967163
    },
    {
      "auxiliary_loss_clip": 0.01182055,
      "auxiliary_loss_mlp": 0.01041239,
      "balance_loss_clip": 1.05852056,
      "balance_loss_mlp": 1.03079617,
      "epoch": 0.18998376720976373,
      "flos": 23767908497280.0,
      "grad_norm": 1.6697420886148606,
      "language_loss": 0.83789343,
      "learning_rate": 3.7376064184409817e-06,
      "loss": 0.86012638,
      "num_input_tokens_seen": 33533100,
      "step": 1580,
      "time_per_iteration": 3.5903079509735107
    },
    {
      "auxiliary_loss_clip": 0.01186505,
      "auxiliary_loss_mlp": 0.01036881,
      "balance_loss_clip": 1.05727291,
      "balance_loss_mlp": 1.02682018,
      "epoch": 0.19010401010040281,
      "flos": 22966023323520.0,
      "grad_norm": 1.498791558434531,
      "language_loss": 0.8705529,
      "learning_rate": 3.7372205729739063e-06,
      "loss": 0.89278674,
      "num_input_tokens_seen": 33554915,
      "step": 1581,
      "time_per_iteration": 2.695162773132324
    },
    {
      "auxiliary_loss_clip": 0.01209502,
      "auxiliary_loss_mlp": 0.01042128,
      "balance_loss_clip": 1.06037307,
      "balance_loss_mlp": 1.03178096,
      "epoch": 0.1902242529910419,
      "flos": 19135647774720.0,
      "grad_norm": 2.115984785384188,
      "language_loss": 0.71743214,
      "learning_rate": 3.7368344639718514e-06,
      "loss": 0.73994851,
      "num_input_tokens_seen": 33572850,
      "step": 1582,
      "time_per_iteration": 2.7316370010375977
    },
    {
      "auxiliary_loss_clip": 0.01203752,
      "auxiliary_loss_mlp": 0.01038249,
      "balance_loss_clip": 1.05649948,
      "balance_loss_mlp": 1.02812862,
      "epoch": 0.190344495881681,
      "flos": 25483935824640.0,
      "grad_norm": 1.5873846863794812,
      "language_loss": 0.80469447,
      "learning_rate": 3.7364480914933895e-06,
      "loss": 0.82711452,
      "num_input_tokens_seen": 33593090,
      "step": 1583,
      "time_per_iteration": 2.6865155696868896
    },
    {
      "auxiliary_loss_clip": 0.01155495,
      "auxiliary_loss_mlp": 0.00764898,
      "balance_loss_clip": 1.05501914,
      "balance_loss_mlp": 1.0003345,
      "epoch": 0.1904647387723201,
      "flos": 26792843425920.0,
      "grad_norm": 2.0179144475329407,
      "language_loss": 0.80900228,
      "learning_rate": 3.7360614555971325e-06,
      "loss": 0.82820618,
      "num_input_tokens_seen": 33612745,
      "step": 1584,
      "time_per_iteration": 2.7985048294067383
    },
    {
      "auxiliary_loss_clip": 0.01204705,
      "auxiliary_loss_mlp": 0.00764631,
      "balance_loss_clip": 1.06078291,
      "balance_loss_mlp": 1.0003469,
      "epoch": 0.19058498166295917,
      "flos": 23987753688960.0,
      "grad_norm": 2.141482612512167,
      "language_loss": 0.85261488,
      "learning_rate": 3.735674556341733e-06,
      "loss": 0.87230819,
      "num_input_tokens_seen": 33632360,
      "step": 1585,
      "time_per_iteration": 2.681871175765991
    },
    {
      "auxiliary_loss_clip": 0.01187675,
      "auxiliary_loss_mlp": 0.01035638,
      "balance_loss_clip": 1.05938864,
      "balance_loss_mlp": 1.02486181,
      "epoch": 0.19070522455359826,
      "flos": 28293299280000.0,
      "grad_norm": 4.068689563343202,
      "language_loss": 0.8290869,
      "learning_rate": 3.7352873937858835e-06,
      "loss": 0.85132003,
      "num_input_tokens_seen": 33653895,
      "step": 1586,
      "time_per_iteration": 2.7686574459075928
    },
    {
      "auxiliary_loss_clip": 0.01170021,
      "auxiliary_loss_mlp": 0.00764413,
      "balance_loss_clip": 1.0569011,
      "balance_loss_mlp": 1.00036395,
      "epoch": 0.19082546744423737,
      "flos": 25660220797440.0,
      "grad_norm": 1.974835517063537,
      "language_loss": 0.72060329,
      "learning_rate": 3.734899967988316e-06,
      "loss": 0.73994762,
      "num_input_tokens_seen": 33672075,
      "step": 1587,
      "time_per_iteration": 2.745598554611206
    },
    {
      "auxiliary_loss_clip": 0.01167828,
      "auxiliary_loss_mlp": 0.01035026,
      "balance_loss_clip": 1.05303645,
      "balance_loss_mlp": 1.02517295,
      "epoch": 0.19094571033487645,
      "flos": 19719483436800.0,
      "grad_norm": 1.8850345534381805,
      "language_loss": 0.8374157,
      "learning_rate": 3.7345122790078026e-06,
      "loss": 0.85944414,
      "num_input_tokens_seen": 33689640,
      "step": 1588,
      "time_per_iteration": 2.72975492477417
    },
    {
      "auxiliary_loss_clip": 0.01202308,
      "auxiliary_loss_mlp": 0.01036479,
      "balance_loss_clip": 1.05875206,
      "balance_loss_mlp": 1.02590537,
      "epoch": 0.19106595322551553,
      "flos": 21616320850560.0,
      "grad_norm": 2.803734634021549,
      "language_loss": 0.93232375,
      "learning_rate": 3.7341243269031556e-06,
      "loss": 0.95471156,
      "num_input_tokens_seen": 33708630,
      "step": 1589,
      "time_per_iteration": 2.628833055496216
    },
    {
      "auxiliary_loss_clip": 0.01181731,
      "auxiliary_loss_mlp": 0.01033928,
      "balance_loss_clip": 1.05804682,
      "balance_loss_mlp": 1.02414083,
      "epoch": 0.19118619611615464,
      "flos": 29896890059520.0,
      "grad_norm": 2.6499761573150264,
      "language_loss": 0.77434117,
      "learning_rate": 3.7337361117332275e-06,
      "loss": 0.7964977,
      "num_input_tokens_seen": 33730370,
      "step": 1590,
      "time_per_iteration": 2.7652671337127686
    },
    {
      "auxiliary_loss_clip": 0.01177619,
      "auxiliary_loss_mlp": 0.01039383,
      "balance_loss_clip": 1.0554738,
      "balance_loss_mlp": 1.02883291,
      "epoch": 0.19130643900679373,
      "flos": 17273428093440.0,
      "grad_norm": 2.02621404764273,
      "language_loss": 0.76929939,
      "learning_rate": 3.7333476335569087e-06,
      "loss": 0.7914694,
      "num_input_tokens_seen": 33748370,
      "step": 1591,
      "time_per_iteration": 2.7320308685302734
    },
    {
      "auxiliary_loss_clip": 0.01189601,
      "auxiliary_loss_mlp": 0.01035784,
      "balance_loss_clip": 1.05841565,
      "balance_loss_mlp": 1.02534735,
      "epoch": 0.1914266818974328,
      "flos": 24826339584000.0,
      "grad_norm": 6.864705202975076,
      "language_loss": 0.66954768,
      "learning_rate": 3.7329588924331325e-06,
      "loss": 0.69180149,
      "num_input_tokens_seen": 33769575,
      "step": 1592,
      "time_per_iteration": 2.6901040077209473
    },
    {
      "auxiliary_loss_clip": 0.01167635,
      "auxiliary_loss_mlp": 0.01033345,
      "balance_loss_clip": 1.0541029,
      "balance_loss_mlp": 1.02349818,
      "epoch": 0.1915469247880719,
      "flos": 18952467390720.0,
      "grad_norm": 1.8818494644286625,
      "language_loss": 0.8277123,
      "learning_rate": 3.732569888420871e-06,
      "loss": 0.84972215,
      "num_input_tokens_seen": 33789110,
      "step": 1593,
      "time_per_iteration": 2.828733444213867
    },
    {
      "auxiliary_loss_clip": 0.01223582,
      "auxiliary_loss_mlp": 0.01038814,
      "balance_loss_clip": 1.05992627,
      "balance_loss_mlp": 1.02851427,
      "epoch": 0.191667167678711,
      "flos": 21032952065280.0,
      "grad_norm": 3.0410573350660997,
      "language_loss": 0.8227663,
      "learning_rate": 3.732180621579134e-06,
      "loss": 0.8453902,
      "num_input_tokens_seen": 33808325,
      "step": 1594,
      "time_per_iteration": 2.6834306716918945
    },
    {
      "auxiliary_loss_clip": 0.01191656,
      "auxiliary_loss_mlp": 0.01044208,
      "balance_loss_clip": 1.0617007,
      "balance_loss_mlp": 1.03338969,
      "epoch": 0.1917874105693501,
      "flos": 34237663914240.0,
      "grad_norm": 1.951970570191408,
      "language_loss": 0.81202137,
      "learning_rate": 3.7317910919669745e-06,
      "loss": 0.83437997,
      "num_input_tokens_seen": 33829520,
      "step": 1595,
      "time_per_iteration": 2.8777410984039307
    },
    {
      "auxiliary_loss_clip": 0.01205437,
      "auxiliary_loss_mlp": 0.01046728,
      "balance_loss_clip": 1.06299031,
      "balance_loss_mlp": 1.03521812,
      "epoch": 0.19190765345998917,
      "flos": 23550613171200.0,
      "grad_norm": 2.3180453556524134,
      "language_loss": 0.75839764,
      "learning_rate": 3.7314012996434826e-06,
      "loss": 0.78091931,
      "num_input_tokens_seen": 33848250,
      "step": 1596,
      "time_per_iteration": 2.6536946296691895
    },
    {
      "auxiliary_loss_clip": 0.01191873,
      "auxiliary_loss_mlp": 0.01037822,
      "balance_loss_clip": 1.05964255,
      "balance_loss_mlp": 1.02804041,
      "epoch": 0.19202789635062828,
      "flos": 19861330245120.0,
      "grad_norm": 1.870282944765852,
      "language_loss": 0.81197864,
      "learning_rate": 3.7310112446677907e-06,
      "loss": 0.8342756,
      "num_input_tokens_seen": 33866160,
      "step": 1597,
      "time_per_iteration": 2.710824489593506
    },
    {
      "auxiliary_loss_clip": 0.012236,
      "auxiliary_loss_mlp": 0.01041152,
      "balance_loss_clip": 1.064816,
      "balance_loss_mlp": 1.03011894,
      "epoch": 0.19214813924126736,
      "flos": 20922957642240.0,
      "grad_norm": 3.753689376995639,
      "language_loss": 0.69051981,
      "learning_rate": 3.7306209270990695e-06,
      "loss": 0.71316731,
      "num_input_tokens_seen": 33884165,
      "step": 1598,
      "time_per_iteration": 2.621387004852295
    },
    {
      "auxiliary_loss_clip": 0.01193857,
      "auxiliary_loss_mlp": 0.01043922,
      "balance_loss_clip": 1.06226587,
      "balance_loss_mlp": 1.03323483,
      "epoch": 0.19226838213190645,
      "flos": 26359725231360.0,
      "grad_norm": 2.035298772948818,
      "language_loss": 0.86845267,
      "learning_rate": 3.7302303469965292e-06,
      "loss": 0.8908304,
      "num_input_tokens_seen": 33903705,
      "step": 1599,
      "time_per_iteration": 2.737128734588623
    },
    {
      "auxiliary_loss_clip": 0.0120864,
      "auxiliary_loss_mlp": 0.01035745,
      "balance_loss_clip": 1.06500137,
      "balance_loss_mlp": 1.02584434,
      "epoch": 0.19238862502254553,
      "flos": 20850525866880.0,
      "grad_norm": 1.762670257314,
      "language_loss": 0.70751715,
      "learning_rate": 3.7298395044194206e-06,
      "loss": 0.72996098,
      "num_input_tokens_seen": 33922515,
      "step": 1600,
      "time_per_iteration": 2.620255470275879
    },
    {
      "auxiliary_loss_clip": 0.01222166,
      "auxiliary_loss_mlp": 0.01034266,
      "balance_loss_clip": 1.06487823,
      "balance_loss_mlp": 1.02424037,
      "epoch": 0.19250886791318464,
      "flos": 21726063878400.0,
      "grad_norm": 2.7357127338932568,
      "language_loss": 0.94362319,
      "learning_rate": 3.7294483994270356e-06,
      "loss": 0.96618748,
      "num_input_tokens_seen": 33940840,
      "step": 1601,
      "time_per_iteration": 2.6158266067504883
    },
    {
      "auxiliary_loss_clip": 0.01152619,
      "auxiliary_loss_mlp": 0.01039455,
      "balance_loss_clip": 1.05232263,
      "balance_loss_mlp": 1.02971578,
      "epoch": 0.19262911080382372,
      "flos": 23367827836800.0,
      "grad_norm": 2.7854940335818568,
      "language_loss": 0.7802276,
      "learning_rate": 3.7290570320787033e-06,
      "loss": 0.80214834,
      "num_input_tokens_seen": 33960420,
      "step": 1602,
      "time_per_iteration": 3.7063117027282715
    },
    {
      "auxiliary_loss_clip": 0.0120658,
      "auxiliary_loss_mlp": 0.01036808,
      "balance_loss_clip": 1.06370831,
      "balance_loss_mlp": 1.02691936,
      "epoch": 0.1927493536944628,
      "flos": 21943502858880.0,
      "grad_norm": 2.0727468896605106,
      "language_loss": 0.71819353,
      "learning_rate": 3.728665402433793e-06,
      "loss": 0.74062741,
      "num_input_tokens_seen": 33978990,
      "step": 1603,
      "time_per_iteration": 2.6120762825012207
    },
    {
      "auxiliary_loss_clip": 0.01188657,
      "auxiliary_loss_mlp": 0.01031847,
      "balance_loss_clip": 1.0603888,
      "balance_loss_mlp": 1.02200651,
      "epoch": 0.19286959658510192,
      "flos": 16545590807040.0,
      "grad_norm": 2.4437673649739304,
      "language_loss": 0.86346799,
      "learning_rate": 3.7282735105517164e-06,
      "loss": 0.88567299,
      "num_input_tokens_seen": 33997115,
      "step": 1604,
      "time_per_iteration": 2.690920114517212
    },
    {
      "auxiliary_loss_clip": 0.01165931,
      "auxiliary_loss_mlp": 0.01043488,
      "balance_loss_clip": 1.05529737,
      "balance_loss_mlp": 1.03249681,
      "epoch": 0.192989839475741,
      "flos": 21616967295360.0,
      "grad_norm": 2.1402701626119898,
      "language_loss": 0.6792804,
      "learning_rate": 3.727881356491922e-06,
      "loss": 0.70137459,
      "num_input_tokens_seen": 34015525,
      "step": 1605,
      "time_per_iteration": 3.6823296546936035
    },
    {
      "auxiliary_loss_clip": 0.01225588,
      "auxiliary_loss_mlp": 0.01042151,
      "balance_loss_clip": 1.06856585,
      "balance_loss_mlp": 1.03245914,
      "epoch": 0.19311008236638008,
      "flos": 19281516906240.0,
      "grad_norm": 1.7567586223534826,
      "language_loss": 0.75727755,
      "learning_rate": 3.7274889403139002e-06,
      "loss": 0.77995497,
      "num_input_tokens_seen": 34033150,
      "step": 1606,
      "time_per_iteration": 4.3374645709991455
    },
    {
      "auxiliary_loss_clip": 0.01156219,
      "auxiliary_loss_mlp": 0.01030953,
      "balance_loss_clip": 1.05780292,
      "balance_loss_mlp": 1.02104712,
      "epoch": 0.1932303252570192,
      "flos": 28652369587200.0,
      "grad_norm": 2.1572398440936387,
      "language_loss": 0.78492785,
      "learning_rate": 3.727096262077179e-06,
      "loss": 0.80679953,
      "num_input_tokens_seen": 34052145,
      "step": 1607,
      "time_per_iteration": 2.8018698692321777
    },
    {
      "auxiliary_loss_clip": 0.01206899,
      "auxiliary_loss_mlp": 0.01036742,
      "balance_loss_clip": 1.06037724,
      "balance_loss_mlp": 1.02663291,
      "epoch": 0.19335056814765827,
      "flos": 18368990864640.0,
      "grad_norm": 1.7816225885615478,
      "language_loss": 0.85400224,
      "learning_rate": 3.7267033218413285e-06,
      "loss": 0.87643868,
      "num_input_tokens_seen": 34069940,
      "step": 1608,
      "time_per_iteration": 2.617530345916748
    },
    {
      "auxiliary_loss_clip": 0.01144398,
      "auxiliary_loss_mlp": 0.01040793,
      "balance_loss_clip": 1.04822755,
      "balance_loss_mlp": 1.03029084,
      "epoch": 0.19347081103829736,
      "flos": 13260877741440.0,
      "grad_norm": 2.0671221688213035,
      "language_loss": 0.81318182,
      "learning_rate": 3.726310119665957e-06,
      "loss": 0.83503377,
      "num_input_tokens_seen": 34086275,
      "step": 1609,
      "time_per_iteration": 2.738497734069824
    },
    {
      "auxiliary_loss_clip": 0.01204686,
      "auxiliary_loss_mlp": 0.0104103,
      "balance_loss_clip": 1.06112373,
      "balance_loss_mlp": 1.03017569,
      "epoch": 0.19359105392893644,
      "flos": 20300122788480.0,
      "grad_norm": 1.8434858420124096,
      "language_loss": 0.85479486,
      "learning_rate": 3.725916655610713e-06,
      "loss": 0.8772521,
      "num_input_tokens_seen": 34105605,
      "step": 1610,
      "time_per_iteration": 2.6593339443206787
    },
    {
      "auxiliary_loss_clip": 0.01183212,
      "auxiliary_loss_mlp": 0.01037507,
      "balance_loss_clip": 1.05775452,
      "balance_loss_mlp": 1.02583039,
      "epoch": 0.19371129681957555,
      "flos": 20484596062080.0,
      "grad_norm": 6.557986718030366,
      "language_loss": 0.75276494,
      "learning_rate": 3.725522929735284e-06,
      "loss": 0.77497208,
      "num_input_tokens_seen": 34122540,
      "step": 1611,
      "time_per_iteration": 2.6925272941589355
    },
    {
      "auxiliary_loss_clip": 0.0119619,
      "auxiliary_loss_mlp": 0.01035871,
      "balance_loss_clip": 1.05735564,
      "balance_loss_mlp": 1.0257262,
      "epoch": 0.19383153971021463,
      "flos": 30445497457920.0,
      "grad_norm": 1.963077797995845,
      "language_loss": 0.74259394,
      "learning_rate": 3.725128942099399e-06,
      "loss": 0.76491457,
      "num_input_tokens_seen": 34142940,
      "step": 1612,
      "time_per_iteration": 2.819523334503174
    },
    {
      "auxiliary_loss_clip": 0.01181183,
      "auxiliary_loss_mlp": 0.01035892,
      "balance_loss_clip": 1.05540347,
      "balance_loss_mlp": 1.0256815,
      "epoch": 0.19395178260085372,
      "flos": 24569937325440.0,
      "grad_norm": 2.1590773267294163,
      "language_loss": 0.79879105,
      "learning_rate": 3.7247346927628245e-06,
      "loss": 0.82096183,
      "num_input_tokens_seen": 34162875,
      "step": 1613,
      "time_per_iteration": 2.7619969844818115
    },
    {
      "auxiliary_loss_clip": 0.0118884,
      "auxiliary_loss_mlp": 0.0076609,
      "balance_loss_clip": 1.05869246,
      "balance_loss_mlp": 1.00037909,
      "epoch": 0.19407202549149283,
      "flos": 28950608211840.0,
      "grad_norm": 1.8465221731833261,
      "language_loss": 0.79274416,
      "learning_rate": 3.7243401817853694e-06,
      "loss": 0.81229347,
      "num_input_tokens_seen": 34183565,
      "step": 1614,
      "time_per_iteration": 2.769099235534668
    },
    {
      "auxiliary_loss_clip": 0.01194108,
      "auxiliary_loss_mlp": 0.01037405,
      "balance_loss_clip": 1.05786669,
      "balance_loss_mlp": 1.02812481,
      "epoch": 0.1941922683821319,
      "flos": 18004497603840.0,
      "grad_norm": 1.8103876307679714,
      "language_loss": 0.71763432,
      "learning_rate": 3.723945409226879e-06,
      "loss": 0.73994946,
      "num_input_tokens_seen": 34202055,
      "step": 1615,
      "time_per_iteration": 2.701265573501587
    },
    {
      "auxiliary_loss_clip": 0.01202262,
      "auxiliary_loss_mlp": 0.01043159,
      "balance_loss_clip": 1.0570035,
      "balance_loss_mlp": 1.03273416,
      "epoch": 0.194312511272771,
      "flos": 9720337034880.0,
      "grad_norm": 2.2771167668968153,
      "language_loss": 0.80291224,
      "learning_rate": 3.723550375147241e-06,
      "loss": 0.8253665,
      "num_input_tokens_seen": 34216830,
      "step": 1616,
      "time_per_iteration": 2.653162956237793
    },
    {
      "auxiliary_loss_clip": 0.01165251,
      "auxiliary_loss_mlp": 0.01038034,
      "balance_loss_clip": 1.05426133,
      "balance_loss_mlp": 1.0279727,
      "epoch": 0.19443275416341008,
      "flos": 27016208150400.0,
      "grad_norm": 2.165658358875777,
      "language_loss": 0.80520159,
      "learning_rate": 3.7231550796063816e-06,
      "loss": 0.82723451,
      "num_input_tokens_seen": 34236840,
      "step": 1617,
      "time_per_iteration": 2.7206828594207764
    },
    {
      "auxiliary_loss_clip": 0.01196668,
      "auxiliary_loss_mlp": 0.01037117,
      "balance_loss_clip": 1.06060588,
      "balance_loss_mlp": 1.02574456,
      "epoch": 0.1945529970540492,
      "flos": 15846625077120.0,
      "grad_norm": 1.9546624966010528,
      "language_loss": 0.64659065,
      "learning_rate": 3.722759522664266e-06,
      "loss": 0.6689285,
      "num_input_tokens_seen": 34254140,
      "step": 1618,
      "time_per_iteration": 2.6727190017700195
    },
    {
      "auxiliary_loss_clip": 0.01160696,
      "auxiliary_loss_mlp": 0.01035625,
      "balance_loss_clip": 1.05421662,
      "balance_loss_mlp": 1.02559912,
      "epoch": 0.19467323994468827,
      "flos": 19314985403520.0,
      "grad_norm": 3.1102726666328198,
      "language_loss": 0.82070327,
      "learning_rate": 3.7223637043809016e-06,
      "loss": 0.84266645,
      "num_input_tokens_seen": 34273120,
      "step": 1619,
      "time_per_iteration": 2.6848127841949463
    },
    {
      "auxiliary_loss_clip": 0.01180828,
      "auxiliary_loss_mlp": 0.01033686,
      "balance_loss_clip": 1.06012321,
      "balance_loss_mlp": 1.02308261,
      "epoch": 0.19479348283532735,
      "flos": 24133227770880.0,
      "grad_norm": 1.8725404678165212,
      "language_loss": 0.86233091,
      "learning_rate": 3.7219676248163322e-06,
      "loss": 0.88447607,
      "num_input_tokens_seen": 34290285,
      "step": 1620,
      "time_per_iteration": 2.7272744178771973
    },
    {
      "auxiliary_loss_clip": 0.01208096,
      "auxiliary_loss_mlp": 0.01033054,
      "balance_loss_clip": 1.06157351,
      "balance_loss_mlp": 1.02289128,
      "epoch": 0.19491372572596646,
      "flos": 25775638174080.0,
      "grad_norm": 1.9724666738993923,
      "language_loss": 0.93326533,
      "learning_rate": 3.721571284030643e-06,
      "loss": 0.95567691,
      "num_input_tokens_seen": 34310095,
      "step": 1621,
      "time_per_iteration": 2.671029806137085
    },
    {
      "auxiliary_loss_clip": 0.01207851,
      "auxiliary_loss_mlp": 0.0103608,
      "balance_loss_clip": 1.06112516,
      "balance_loss_mlp": 1.02604258,
      "epoch": 0.19503396861660555,
      "flos": 19645220067840.0,
      "grad_norm": 2.11034263706293,
      "language_loss": 0.7865411,
      "learning_rate": 3.7211746820839587e-06,
      "loss": 0.80898041,
      "num_input_tokens_seen": 34327190,
      "step": 1622,
      "time_per_iteration": 2.646682024002075
    },
    {
      "auxiliary_loss_clip": 0.01115296,
      "auxiliary_loss_mlp": 0.0103223,
      "balance_loss_clip": 1.04758763,
      "balance_loss_mlp": 1.02154839,
      "epoch": 0.19515421150724463,
      "flos": 21033023892480.0,
      "grad_norm": 1.7202495622638343,
      "language_loss": 0.81000853,
      "learning_rate": 3.7207778190364437e-06,
      "loss": 0.83148378,
      "num_input_tokens_seen": 34345615,
      "step": 1623,
      "time_per_iteration": 2.862776279449463
    },
    {
      "auxiliary_loss_clip": 0.01125254,
      "auxiliary_loss_mlp": 0.01038618,
      "balance_loss_clip": 1.04838061,
      "balance_loss_mlp": 1.02908158,
      "epoch": 0.1952744543978837,
      "flos": 32961255143040.0,
      "grad_norm": 1.7146883622886562,
      "language_loss": 0.74363679,
      "learning_rate": 3.720380694948302e-06,
      "loss": 0.76527548,
      "num_input_tokens_seen": 34368500,
      "step": 1624,
      "time_per_iteration": 2.845585823059082
    },
    {
      "auxiliary_loss_clip": 0.0107259,
      "auxiliary_loss_mlp": 0.01002951,
      "balance_loss_clip": 1.02277088,
      "balance_loss_mlp": 1.0002923,
      "epoch": 0.19539469728852282,
      "flos": 64044312030720.0,
      "grad_norm": 1.0376579821521623,
      "language_loss": 0.71220857,
      "learning_rate": 3.719983309879777e-06,
      "loss": 0.73296392,
      "num_input_tokens_seen": 34428280,
      "step": 1625,
      "time_per_iteration": 3.3515465259552
    },
    {
      "auxiliary_loss_clip": 0.01162774,
      "auxiliary_loss_mlp": 0.01030777,
      "balance_loss_clip": 1.0530616,
      "balance_loss_mlp": 1.02048278,
      "epoch": 0.1955149401791619,
      "flos": 13370908078080.0,
      "grad_norm": 2.1474978154627813,
      "language_loss": 0.77619386,
      "learning_rate": 3.719585663891151e-06,
      "loss": 0.79812938,
      "num_input_tokens_seen": 34445815,
      "step": 1626,
      "time_per_iteration": 2.7452781200408936
    },
    {
      "auxiliary_loss_clip": 0.01148647,
      "auxiliary_loss_mlp": 0.01040302,
      "balance_loss_clip": 1.05319595,
      "balance_loss_mlp": 1.02975178,
      "epoch": 0.195635183069801,
      "flos": 18728887184640.0,
      "grad_norm": 2.0570827351973087,
      "language_loss": 0.78805733,
      "learning_rate": 3.719187757042747e-06,
      "loss": 0.80994684,
      "num_input_tokens_seen": 34463635,
      "step": 1627,
      "time_per_iteration": 3.778257369995117
    },
    {
      "auxiliary_loss_clip": 0.01086681,
      "auxiliary_loss_mlp": 0.01003099,
      "balance_loss_clip": 1.02014399,
      "balance_loss_mlp": 1.00058353,
      "epoch": 0.1957554259604401,
      "flos": 69313952615040.0,
      "grad_norm": 0.7270530549696459,
      "language_loss": 0.54912388,
      "learning_rate": 3.7187895893949275e-06,
      "loss": 0.57002163,
      "num_input_tokens_seen": 34530105,
      "step": 1628,
      "time_per_iteration": 3.36309552192688
    },
    {
      "auxiliary_loss_clip": 0.01145026,
      "auxiliary_loss_mlp": 0.0103547,
      "balance_loss_clip": 1.04988635,
      "balance_loss_mlp": 1.02501535,
      "epoch": 0.19587566885107918,
      "flos": 21069257736960.0,
      "grad_norm": 2.5348951175679564,
      "language_loss": 0.75784409,
      "learning_rate": 3.7183911610080937e-06,
      "loss": 0.77964914,
      "num_input_tokens_seen": 34546970,
      "step": 1629,
      "time_per_iteration": 2.732863664627075
    },
    {
      "auxiliary_loss_clip": 0.01176398,
      "auxiliary_loss_mlp": 0.01038151,
      "balance_loss_clip": 1.05638313,
      "balance_loss_mlp": 1.02742171,
      "epoch": 0.19599591174171827,
      "flos": 22194661731840.0,
      "grad_norm": 3.3058366412703197,
      "language_loss": 0.75441158,
      "learning_rate": 3.7179924719426872e-06,
      "loss": 0.77655709,
      "num_input_tokens_seen": 34564865,
      "step": 1630,
      "time_per_iteration": 3.665860891342163
    },
    {
      "auxiliary_loss_clip": 0.01210037,
      "auxiliary_loss_mlp": 0.01036393,
      "balance_loss_clip": 1.06387007,
      "balance_loss_mlp": 1.02538955,
      "epoch": 0.19611615463235738,
      "flos": 23768375374080.0,
      "grad_norm": 2.9257658447004586,
      "language_loss": 0.76209515,
      "learning_rate": 3.7175935222591885e-06,
      "loss": 0.78455943,
      "num_input_tokens_seen": 34584165,
      "step": 1631,
      "time_per_iteration": 3.63474178314209
    },
    {
      "auxiliary_loss_clip": 0.01196353,
      "auxiliary_loss_mlp": 0.01039639,
      "balance_loss_clip": 1.06337619,
      "balance_loss_mlp": 1.02954149,
      "epoch": 0.19623639752299646,
      "flos": 28618218731520.0,
      "grad_norm": 2.0664447591213055,
      "language_loss": 0.7441802,
      "learning_rate": 3.717194312018118e-06,
      "loss": 0.76654017,
      "num_input_tokens_seen": 34603150,
      "step": 1632,
      "time_per_iteration": 3.656517744064331
    },
    {
      "auxiliary_loss_clip": 0.01205469,
      "auxiliary_loss_mlp": 0.01032213,
      "balance_loss_clip": 1.05960906,
      "balance_loss_mlp": 1.02162683,
      "epoch": 0.19635664041363554,
      "flos": 21032700670080.0,
      "grad_norm": 2.2734319481801646,
      "language_loss": 0.76245272,
      "learning_rate": 3.716794841280036e-06,
      "loss": 0.7848295,
      "num_input_tokens_seen": 34621855,
      "step": 1633,
      "time_per_iteration": 2.607369899749756
    },
    {
      "auxiliary_loss_clip": 0.01212144,
      "auxiliary_loss_mlp": 0.01040858,
      "balance_loss_clip": 1.0621357,
      "balance_loss_mlp": 1.02940226,
      "epoch": 0.19647688330427462,
      "flos": 18879748306560.0,
      "grad_norm": 3.5203457896433736,
      "language_loss": 0.77359742,
      "learning_rate": 3.7163951101055407e-06,
      "loss": 0.79612744,
      "num_input_tokens_seen": 34639915,
      "step": 1634,
      "time_per_iteration": 2.6446104049682617
    },
    {
      "auxiliary_loss_clip": 0.01186148,
      "auxiliary_loss_mlp": 0.01041353,
      "balance_loss_clip": 1.06057644,
      "balance_loss_mlp": 1.03054643,
      "epoch": 0.19659712619491373,
      "flos": 24242503921920.0,
      "grad_norm": 2.008789667085443,
      "language_loss": 0.79184604,
      "learning_rate": 3.715995118555273e-06,
      "loss": 0.81412101,
      "num_input_tokens_seen": 34659890,
      "step": 1635,
      "time_per_iteration": 2.670377492904663
    },
    {
      "auxiliary_loss_clip": 0.01158876,
      "auxiliary_loss_mlp": 0.0104062,
      "balance_loss_clip": 1.05378532,
      "balance_loss_mlp": 1.02953315,
      "epoch": 0.19671736908555282,
      "flos": 24717422568960.0,
      "grad_norm": 2.1760719837035603,
      "language_loss": 0.85765707,
      "learning_rate": 3.71559486668991e-06,
      "loss": 0.87965202,
      "num_input_tokens_seen": 34678750,
      "step": 1636,
      "time_per_iteration": 2.76328444480896
    },
    {
      "auxiliary_loss_clip": 0.01212096,
      "auxiliary_loss_mlp": 0.00764534,
      "balance_loss_clip": 1.06325912,
      "balance_loss_mlp": 1.00035894,
      "epoch": 0.1968376119761919,
      "flos": 23842279607040.0,
      "grad_norm": 1.911921921568929,
      "language_loss": 0.77551758,
      "learning_rate": 3.715194354570169e-06,
      "loss": 0.79528391,
      "num_input_tokens_seen": 34698755,
      "step": 1637,
      "time_per_iteration": 2.6655447483062744
    },
    {
      "auxiliary_loss_clip": 0.01201765,
      "auxiliary_loss_mlp": 0.0103861,
      "balance_loss_clip": 1.06042945,
      "balance_loss_mlp": 1.02773762,
      "epoch": 0.196957854866831,
      "flos": 18113917409280.0,
      "grad_norm": 2.5200842005866777,
      "language_loss": 0.83330292,
      "learning_rate": 3.714793582256809e-06,
      "loss": 0.85570669,
      "num_input_tokens_seen": 34715820,
      "step": 1638,
      "time_per_iteration": 2.6650736331939697
    },
    {
      "auxiliary_loss_clip": 0.01219675,
      "auxiliary_loss_mlp": 0.01034701,
      "balance_loss_clip": 1.06224465,
      "balance_loss_mlp": 1.02437758,
      "epoch": 0.1970780977574701,
      "flos": 21653129312640.0,
      "grad_norm": 2.4702991948462785,
      "language_loss": 0.84924835,
      "learning_rate": 3.7143925498106253e-06,
      "loss": 0.87179208,
      "num_input_tokens_seen": 34734360,
      "step": 1639,
      "time_per_iteration": 2.603489637374878
    },
    {
      "auxiliary_loss_clip": 0.01187866,
      "auxiliary_loss_mlp": 0.01045339,
      "balance_loss_clip": 1.05353928,
      "balance_loss_mlp": 1.03477144,
      "epoch": 0.19719834064810918,
      "flos": 20811813984000.0,
      "grad_norm": 4.130725524126125,
      "language_loss": 0.79318523,
      "learning_rate": 3.7139912572924558e-06,
      "loss": 0.81551731,
      "num_input_tokens_seen": 34753390,
      "step": 1640,
      "time_per_iteration": 2.755589723587036
    },
    {
      "auxiliary_loss_clip": 0.01198583,
      "auxiliary_loss_mlp": 0.01036873,
      "balance_loss_clip": 1.05454624,
      "balance_loss_mlp": 1.0266633,
      "epoch": 0.19731858353874826,
      "flos": 23434800744960.0,
      "grad_norm": 3.1789331643547816,
      "language_loss": 0.80617511,
      "learning_rate": 3.7135897047631744e-06,
      "loss": 0.82852972,
      "num_input_tokens_seen": 34771275,
      "step": 1641,
      "time_per_iteration": 2.6487386226654053
    },
    {
      "auxiliary_loss_clip": 0.0119064,
      "auxiliary_loss_mlp": 0.01035115,
      "balance_loss_clip": 1.0570358,
      "balance_loss_mlp": 1.02526796,
      "epoch": 0.19743882642938737,
      "flos": 23988184652160.0,
      "grad_norm": 1.8979611486045749,
      "language_loss": 0.75852454,
      "learning_rate": 3.713187892283698e-06,
      "loss": 0.78078204,
      "num_input_tokens_seen": 34790885,
      "step": 1642,
      "time_per_iteration": 2.7182199954986572
    },
    {
      "auxiliary_loss_clip": 0.0115662,
      "auxiliary_loss_mlp": 0.01034802,
      "balance_loss_clip": 1.05070972,
      "balance_loss_mlp": 1.02413881,
      "epoch": 0.19755906932002645,
      "flos": 15004340081280.0,
      "grad_norm": 2.242851995114338,
      "language_loss": 0.8746227,
      "learning_rate": 3.71278581991498e-06,
      "loss": 0.89653695,
      "num_input_tokens_seen": 34806745,
      "step": 1643,
      "time_per_iteration": 2.7273337841033936
    },
    {
      "auxiliary_loss_clip": 0.01177831,
      "auxiliary_loss_mlp": 0.00766101,
      "balance_loss_clip": 1.06085277,
      "balance_loss_mlp": 1.00039816,
      "epoch": 0.19767931221066554,
      "flos": 19494466686720.0,
      "grad_norm": 1.820360104805376,
      "language_loss": 0.78936523,
      "learning_rate": 3.712383487718015e-06,
      "loss": 0.80880451,
      "num_input_tokens_seen": 34824985,
      "step": 1644,
      "time_per_iteration": 2.7188034057617188
    },
    {
      "auxiliary_loss_clip": 0.01136896,
      "auxiliary_loss_mlp": 0.01037118,
      "balance_loss_clip": 1.05050242,
      "balance_loss_mlp": 1.026896,
      "epoch": 0.19779955510130465,
      "flos": 25737895958400.0,
      "grad_norm": 2.0979741703832544,
      "language_loss": 0.86931986,
      "learning_rate": 3.7119808957538365e-06,
      "loss": 0.89105999,
      "num_input_tokens_seen": 34843980,
      "step": 1645,
      "time_per_iteration": 2.822990894317627
    },
    {
      "auxiliary_loss_clip": 0.01179385,
      "auxiliary_loss_mlp": 0.0103612,
      "balance_loss_clip": 1.05174506,
      "balance_loss_mlp": 1.02539754,
      "epoch": 0.19791979799194373,
      "flos": 20777699041920.0,
      "grad_norm": 1.9802365886251696,
      "language_loss": 0.80001855,
      "learning_rate": 3.711578044083517e-06,
      "loss": 0.8221736,
      "num_input_tokens_seen": 34860780,
      "step": 1646,
      "time_per_iteration": 2.6890830993652344
    },
    {
      "auxiliary_loss_clip": 0.01186781,
      "auxiliary_loss_mlp": 0.01037959,
      "balance_loss_clip": 1.05419421,
      "balance_loss_mlp": 1.02755785,
      "epoch": 0.1980400408825828,
      "flos": 25589010084480.0,
      "grad_norm": 2.1566918454533126,
      "language_loss": 0.7448898,
      "learning_rate": 3.7111749327681698e-06,
      "loss": 0.76713723,
      "num_input_tokens_seen": 34880815,
      "step": 1647,
      "time_per_iteration": 2.747211217880249
    },
    {
      "auxiliary_loss_clip": 0.01208941,
      "auxiliary_loss_mlp": 0.01036124,
      "balance_loss_clip": 1.06238294,
      "balance_loss_mlp": 1.02525854,
      "epoch": 0.1981602837732219,
      "flos": 23513840622720.0,
      "grad_norm": 2.3378797838211516,
      "language_loss": 0.86383677,
      "learning_rate": 3.7107715618689455e-06,
      "loss": 0.88628745,
      "num_input_tokens_seen": 34899790,
      "step": 1648,
      "time_per_iteration": 2.665029764175415
    },
    {
      "auxiliary_loss_clip": 0.01201609,
      "auxiliary_loss_mlp": 0.01035316,
      "balance_loss_clip": 1.06036925,
      "balance_loss_mlp": 1.02458668,
      "epoch": 0.198280526663861,
      "flos": 23185365724800.0,
      "grad_norm": 1.4542661382610191,
      "language_loss": 0.83546567,
      "learning_rate": 3.710367931447035e-06,
      "loss": 0.85783488,
      "num_input_tokens_seen": 34921570,
      "step": 1649,
      "time_per_iteration": 2.666713237762451
    },
    {
      "auxiliary_loss_clip": 0.01210826,
      "auxiliary_loss_mlp": 0.01040442,
      "balance_loss_clip": 1.06071258,
      "balance_loss_mlp": 1.02879477,
      "epoch": 0.1984007695545001,
      "flos": 21689470897920.0,
      "grad_norm": 2.5420637162641144,
      "language_loss": 0.86742747,
      "learning_rate": 3.70996404156367e-06,
      "loss": 0.88994014,
      "num_input_tokens_seen": 34941205,
      "step": 1650,
      "time_per_iteration": 2.631938934326172
    },
    {
      "auxiliary_loss_clip": 0.01146724,
      "auxiliary_loss_mlp": 0.01037288,
      "balance_loss_clip": 1.05157137,
      "balance_loss_mlp": 1.02735245,
      "epoch": 0.19852101244513917,
      "flos": 36064008887040.0,
      "grad_norm": 2.204740948371221,
      "language_loss": 0.72861457,
      "learning_rate": 3.7095598922801187e-06,
      "loss": 0.75045466,
      "num_input_tokens_seen": 34963280,
      "step": 1651,
      "time_per_iteration": 2.9008846282958984
    },
    {
      "auxiliary_loss_clip": 0.01221802,
      "auxiliary_loss_mlp": 0.01041053,
      "balance_loss_clip": 1.06314492,
      "balance_loss_mlp": 1.03073561,
      "epoch": 0.19864125533577828,
      "flos": 23105894883840.0,
      "grad_norm": 3.8111210650229514,
      "language_loss": 0.75859523,
      "learning_rate": 3.7091554836576914e-06,
      "loss": 0.78122377,
      "num_input_tokens_seen": 34979955,
      "step": 1652,
      "time_per_iteration": 2.6117119789123535
    },
    {
      "auxiliary_loss_clip": 0.01201637,
      "auxiliary_loss_mlp": 0.00764428,
      "balance_loss_clip": 1.06084156,
      "balance_loss_mlp": 1.00034595,
      "epoch": 0.19876149822641737,
      "flos": 24608505553920.0,
      "grad_norm": 3.1427705602844607,
      "language_loss": 0.82674503,
      "learning_rate": 3.708750815757736e-06,
      "loss": 0.84640574,
      "num_input_tokens_seen": 35000725,
      "step": 1653,
      "time_per_iteration": 3.607034206390381
    },
    {
      "auxiliary_loss_clip": 0.01206087,
      "auxiliary_loss_mlp": 0.01037251,
      "balance_loss_clip": 1.05963492,
      "balance_loss_mlp": 1.02586007,
      "epoch": 0.19888174111705645,
      "flos": 32196645308160.0,
      "grad_norm": 2.283091313267767,
      "language_loss": 0.73141235,
      "learning_rate": 3.7083458886416407e-06,
      "loss": 0.75384569,
      "num_input_tokens_seen": 35019920,
      "step": 1654,
      "time_per_iteration": 2.7444164752960205
    },
    {
      "auxiliary_loss_clip": 0.01144953,
      "auxiliary_loss_mlp": 0.01037923,
      "balance_loss_clip": 1.05416512,
      "balance_loss_mlp": 1.02794504,
      "epoch": 0.19900198400769553,
      "flos": 24608469640320.0,
      "grad_norm": 2.399440414759682,
      "language_loss": 0.88255692,
      "learning_rate": 3.707940702370832e-06,
      "loss": 0.90438569,
      "num_input_tokens_seen": 35040765,
      "step": 1655,
      "time_per_iteration": 2.810795783996582
    },
    {
      "auxiliary_loss_clip": 0.01100475,
      "auxiliary_loss_mlp": 0.01006355,
      "balance_loss_clip": 1.02480221,
      "balance_loss_mlp": 1.00398278,
      "epoch": 0.19912222689833464,
      "flos": 67915805673600.0,
      "grad_norm": 0.8792203477426301,
      "language_loss": 0.58308947,
      "learning_rate": 3.707535257006777e-06,
      "loss": 0.60415781,
      "num_input_tokens_seen": 35106390,
      "step": 1656,
      "time_per_iteration": 4.201031923294067
    },
    {
      "auxiliary_loss_clip": 0.01189415,
      "auxiliary_loss_mlp": 0.0103821,
      "balance_loss_clip": 1.05763602,
      "balance_loss_mlp": 1.02688503,
      "epoch": 0.19924246978897373,
      "flos": 15742340916480.0,
      "grad_norm": 2.4644453506577033,
      "language_loss": 0.8831085,
      "learning_rate": 3.707129552610981e-06,
      "loss": 0.90538478,
      "num_input_tokens_seen": 35125040,
      "step": 1657,
      "time_per_iteration": 3.5399820804595947
    },
    {
      "auxiliary_loss_clip": 0.01179449,
      "auxiliary_loss_mlp": 0.01036465,
      "balance_loss_clip": 1.05714822,
      "balance_loss_mlp": 1.02568245,
      "epoch": 0.1993627126796128,
      "flos": 17566566986880.0,
      "grad_norm": 1.9039598106391453,
      "language_loss": 0.73251021,
      "learning_rate": 3.70672358924499e-06,
      "loss": 0.75466931,
      "num_input_tokens_seen": 35144280,
      "step": 1658,
      "time_per_iteration": 3.5388295650482178
    },
    {
      "auxiliary_loss_clip": 0.01176945,
      "auxiliary_loss_mlp": 0.0104295,
      "balance_loss_clip": 1.06166828,
      "balance_loss_mlp": 1.03319836,
      "epoch": 0.19948295557025192,
      "flos": 40843826680320.0,
      "grad_norm": 1.9177668510356336,
      "language_loss": 0.78614283,
      "learning_rate": 3.706317366970386e-06,
      "loss": 0.80834174,
      "num_input_tokens_seen": 35165280,
      "step": 1659,
      "time_per_iteration": 2.958683490753174
    },
    {
      "auxiliary_loss_clip": 0.01218795,
      "auxiliary_loss_mlp": 0.00765849,
      "balance_loss_clip": 1.05849886,
      "balance_loss_mlp": 1.00038064,
      "epoch": 0.199603198460891,
      "flos": 25082418620160.0,
      "grad_norm": 4.0338024809059965,
      "language_loss": 0.83964419,
      "learning_rate": 3.705910885848795e-06,
      "loss": 0.85949063,
      "num_input_tokens_seen": 35183655,
      "step": 1660,
      "time_per_iteration": 2.6872732639312744
    },
    {
      "auxiliary_loss_clip": 0.01200486,
      "auxiliary_loss_mlp": 0.01032206,
      "balance_loss_clip": 1.05850911,
      "balance_loss_mlp": 1.0219481,
      "epoch": 0.19972344135153008,
      "flos": 20084120352000.0,
      "grad_norm": 2.0604614335470437,
      "language_loss": 0.8458153,
      "learning_rate": 3.705504145941879e-06,
      "loss": 0.86814225,
      "num_input_tokens_seen": 35201825,
      "step": 1661,
      "time_per_iteration": 2.6485090255737305
    },
    {
      "auxiliary_loss_clip": 0.01216679,
      "auxiliary_loss_mlp": 0.01036696,
      "balance_loss_clip": 1.05956721,
      "balance_loss_mlp": 1.02624726,
      "epoch": 0.1998436842421692,
      "flos": 23727472761600.0,
      "grad_norm": 6.054430083791726,
      "language_loss": 0.78466845,
      "learning_rate": 3.7050971473113403e-06,
      "loss": 0.80720222,
      "num_input_tokens_seen": 35221600,
      "step": 1662,
      "time_per_iteration": 2.666290760040283
    },
    {
      "auxiliary_loss_clip": 0.01197174,
      "auxiliary_loss_mlp": 0.00765121,
      "balance_loss_clip": 1.0552845,
      "balance_loss_mlp": 1.00030494,
      "epoch": 0.19996392713280828,
      "flos": 36102361633920.0,
      "grad_norm": 1.9929382229563757,
      "language_loss": 0.79739201,
      "learning_rate": 3.7046898900189196e-06,
      "loss": 0.81701493,
      "num_input_tokens_seen": 35245935,
      "step": 1663,
      "time_per_iteration": 2.741389513015747
    },
    {
      "auxiliary_loss_clip": 0.01177984,
      "auxiliary_loss_mlp": 0.0103851,
      "balance_loss_clip": 1.05891132,
      "balance_loss_mlp": 1.02779341,
      "epoch": 0.20008417002344736,
      "flos": 23657662679040.0,
      "grad_norm": 1.84739961850067,
      "language_loss": 0.82973504,
      "learning_rate": 3.704282374126398e-06,
      "loss": 0.85190004,
      "num_input_tokens_seen": 35265615,
      "step": 1664,
      "time_per_iteration": 2.6868438720703125
    },
    {
      "auxiliary_loss_clip": 0.01168726,
      "auxiliary_loss_mlp": 0.01037255,
      "balance_loss_clip": 1.05378294,
      "balance_loss_mlp": 1.02631795,
      "epoch": 0.20020441291408644,
      "flos": 21872076664320.0,
      "grad_norm": 1.7463519401263667,
      "language_loss": 0.87579846,
      "learning_rate": 3.7038745996955954e-06,
      "loss": 0.89785826,
      "num_input_tokens_seen": 35284960,
      "step": 1665,
      "time_per_iteration": 2.809835433959961
    },
    {
      "auxiliary_loss_clip": 0.01179234,
      "auxiliary_loss_mlp": 0.01035143,
      "balance_loss_clip": 1.05551934,
      "balance_loss_mlp": 1.02387178,
      "epoch": 0.20032465580472555,
      "flos": 23179691376000.0,
      "grad_norm": 4.442985304959047,
      "language_loss": 0.71608877,
      "learning_rate": 3.703466566788371e-06,
      "loss": 0.73823249,
      "num_input_tokens_seen": 35304090,
      "step": 1666,
      "time_per_iteration": 2.766836404800415
    },
    {
      "auxiliary_loss_clip": 0.01177856,
      "auxiliary_loss_mlp": 0.01038855,
      "balance_loss_clip": 1.05594087,
      "balance_loss_mlp": 1.02832317,
      "epoch": 0.20044489869536464,
      "flos": 23873521461120.0,
      "grad_norm": 1.9135495055218326,
      "language_loss": 0.74500132,
      "learning_rate": 3.703058275466622e-06,
      "loss": 0.7671684,
      "num_input_tokens_seen": 35323325,
      "step": 1667,
      "time_per_iteration": 2.7509090900421143
    },
    {
      "auxiliary_loss_clip": 0.01187117,
      "auxiliary_loss_mlp": 0.01029837,
      "balance_loss_clip": 1.057423,
      "balance_loss_mlp": 1.02010381,
      "epoch": 0.20056514158600372,
      "flos": 21945226711680.0,
      "grad_norm": 2.3913991522243183,
      "language_loss": 0.77819586,
      "learning_rate": 3.7026497257922877e-06,
      "loss": 0.80036533,
      "num_input_tokens_seen": 35343635,
      "step": 1668,
      "time_per_iteration": 2.7524971961975098
    },
    {
      "auxiliary_loss_clip": 0.01151494,
      "auxiliary_loss_mlp": 0.01029604,
      "balance_loss_clip": 1.04945195,
      "balance_loss_mlp": 1.01946521,
      "epoch": 0.20068538447664283,
      "flos": 23879159896320.0,
      "grad_norm": 9.70644699769287,
      "language_loss": 0.85245621,
      "learning_rate": 3.7022409178273436e-06,
      "loss": 0.87426722,
      "num_input_tokens_seen": 35364615,
      "step": 1669,
      "time_per_iteration": 2.8346471786499023
    },
    {
      "auxiliary_loss_clip": 0.01198425,
      "auxiliary_loss_mlp": 0.01036047,
      "balance_loss_clip": 1.05878019,
      "balance_loss_mlp": 1.02513337,
      "epoch": 0.2008056273672819,
      "flos": 18442823270400.0,
      "grad_norm": 2.100290395354796,
      "language_loss": 0.78446567,
      "learning_rate": 3.7018318516338054e-06,
      "loss": 0.80681038,
      "num_input_tokens_seen": 35383775,
      "step": 1670,
      "time_per_iteration": 2.726123094558716
    },
    {
      "auxiliary_loss_clip": 0.01208043,
      "auxiliary_loss_mlp": 0.01038799,
      "balance_loss_clip": 1.06002617,
      "balance_loss_mlp": 1.02818298,
      "epoch": 0.200925870257921,
      "flos": 23659530186240.0,
      "grad_norm": 3.967887376780115,
      "language_loss": 0.82039654,
      "learning_rate": 3.7014225272737284e-06,
      "loss": 0.84286499,
      "num_input_tokens_seen": 35403000,
      "step": 1671,
      "time_per_iteration": 2.789266347885132
    },
    {
      "auxiliary_loss_clip": 0.01195852,
      "auxiliary_loss_mlp": 0.0103807,
      "balance_loss_clip": 1.05797982,
      "balance_loss_mlp": 1.02740717,
      "epoch": 0.20104611314856008,
      "flos": 16217115909120.0,
      "grad_norm": 2.7160509788082843,
      "language_loss": 0.74009037,
      "learning_rate": 3.701012944809207e-06,
      "loss": 0.76242965,
      "num_input_tokens_seen": 35420115,
      "step": 1672,
      "time_per_iteration": 2.6169464588165283
    },
    {
      "auxiliary_loss_clip": 0.01188796,
      "auxiliary_loss_mlp": 0.00765066,
      "balance_loss_clip": 1.06103432,
      "balance_loss_mlp": 1.0003618,
      "epoch": 0.2011663560391992,
      "flos": 21397373498880.0,
      "grad_norm": 1.9073008496628021,
      "language_loss": 0.784289,
      "learning_rate": 3.700603104302374e-06,
      "loss": 0.80382764,
      "num_input_tokens_seen": 35439925,
      "step": 1673,
      "time_per_iteration": 2.7431938648223877
    },
    {
      "auxiliary_loss_clip": 0.01062004,
      "auxiliary_loss_mlp": 0.01009294,
      "balance_loss_clip": 1.01992321,
      "balance_loss_mlp": 1.00677896,
      "epoch": 0.20128659892983827,
      "flos": 62229459409920.0,
      "grad_norm": 0.9028779273303366,
      "language_loss": 0.5594275,
      "learning_rate": 3.7001930058154027e-06,
      "loss": 0.58014047,
      "num_input_tokens_seen": 35504885,
      "step": 1674,
      "time_per_iteration": 3.295372486114502
    },
    {
      "auxiliary_loss_clip": 0.01170699,
      "auxiliary_loss_mlp": 0.01044722,
      "balance_loss_clip": 1.05583417,
      "balance_loss_mlp": 1.03348613,
      "epoch": 0.20140684182047736,
      "flos": 28438737448320.0,
      "grad_norm": 2.590227625655642,
      "language_loss": 0.79631025,
      "learning_rate": 3.6997826494105037e-06,
      "loss": 0.81846446,
      "num_input_tokens_seen": 35525330,
      "step": 1675,
      "time_per_iteration": 2.783301591873169
    },
    {
      "auxiliary_loss_clip": 0.01184746,
      "auxiliary_loss_mlp": 0.01040066,
      "balance_loss_clip": 1.05573487,
      "balance_loss_mlp": 1.02946877,
      "epoch": 0.20152708471111647,
      "flos": 28074064619520.0,
      "grad_norm": 2.4710316702413055,
      "language_loss": 0.69307631,
      "learning_rate": 3.6993720351499286e-06,
      "loss": 0.71532446,
      "num_input_tokens_seen": 35546455,
      "step": 1676,
      "time_per_iteration": 2.7147305011749268
    },
    {
      "auxiliary_loss_clip": 0.01187529,
      "auxiliary_loss_mlp": 0.01034813,
      "balance_loss_clip": 1.06359935,
      "balance_loss_mlp": 1.02510321,
      "epoch": 0.20164732760175555,
      "flos": 23549751244800.0,
      "grad_norm": 1.7207198602815565,
      "language_loss": 0.77161872,
      "learning_rate": 3.6989611630959666e-06,
      "loss": 0.79384208,
      "num_input_tokens_seen": 35565010,
      "step": 1677,
      "time_per_iteration": 2.723766803741455
    },
    {
      "auxiliary_loss_clip": 0.01100808,
      "auxiliary_loss_mlp": 0.0100347,
      "balance_loss_clip": 1.02627802,
      "balance_loss_mlp": 1.00114572,
      "epoch": 0.20176757049239463,
      "flos": 71100616037760.0,
      "grad_norm": 0.6888361340675919,
      "language_loss": 0.58290714,
      "learning_rate": 3.6985500333109474e-06,
      "loss": 0.6039499,
      "num_input_tokens_seen": 35633340,
      "step": 1678,
      "time_per_iteration": 3.3094279766082764
    },
    {
      "auxiliary_loss_clip": 0.01164952,
      "auxiliary_loss_mlp": 0.01029008,
      "balance_loss_clip": 1.05326629,
      "balance_loss_mlp": 1.01895845,
      "epoch": 0.20188781338303372,
      "flos": 21430159637760.0,
      "grad_norm": 3.254991614295124,
      "language_loss": 0.76923764,
      "learning_rate": 3.6981386458572385e-06,
      "loss": 0.79117727,
      "num_input_tokens_seen": 35651315,
      "step": 1679,
      "time_per_iteration": 3.813613176345825
    },
    {
      "auxiliary_loss_clip": 0.01162272,
      "auxiliary_loss_mlp": 0.01045037,
      "balance_loss_clip": 1.05218637,
      "balance_loss_mlp": 1.03391504,
      "epoch": 0.20200805627367283,
      "flos": 11546215130880.0,
      "grad_norm": 2.4749328846522283,
      "language_loss": 0.75854427,
      "learning_rate": 3.6977270007972468e-06,
      "loss": 0.78061736,
      "num_input_tokens_seen": 35668850,
      "step": 1680,
      "time_per_iteration": 2.853621482849121
    },
    {
      "auxiliary_loss_clip": 0.0119381,
      "auxiliary_loss_mlp": 0.01033535,
      "balance_loss_clip": 1.06105661,
      "balance_loss_mlp": 1.02347946,
      "epoch": 0.2021282991643119,
      "flos": 28545391906560.0,
      "grad_norm": 2.8403465665444285,
      "language_loss": 0.72534204,
      "learning_rate": 3.6973150981934196e-06,
      "loss": 0.74761546,
      "num_input_tokens_seen": 35690080,
      "step": 1681,
      "time_per_iteration": 3.7295126914978027
    },
    {
      "auxiliary_loss_clip": 0.01220908,
      "auxiliary_loss_mlp": 0.01040401,
      "balance_loss_clip": 1.06258249,
      "balance_loss_mlp": 1.02992892,
      "epoch": 0.202248542054951,
      "flos": 17923446564480.0,
      "grad_norm": 2.5666566693481507,
      "language_loss": 0.83531266,
      "learning_rate": 3.6969029381082415e-06,
      "loss": 0.85792577,
      "num_input_tokens_seen": 35706075,
      "step": 1682,
      "time_per_iteration": 2.6280176639556885
    },
    {
      "auxiliary_loss_clip": 0.0118254,
      "auxiliary_loss_mlp": 0.01032295,
      "balance_loss_clip": 1.05650663,
      "balance_loss_mlp": 1.02269852,
      "epoch": 0.2023687849455901,
      "flos": 19864634296320.0,
      "grad_norm": 2.3371718072948053,
      "language_loss": 0.79532433,
      "learning_rate": 3.696490520604237e-06,
      "loss": 0.81747264,
      "num_input_tokens_seen": 35724765,
      "step": 1683,
      "time_per_iteration": 3.653432846069336
    },
    {
      "auxiliary_loss_clip": 0.01195,
      "auxiliary_loss_mlp": 0.01035665,
      "balance_loss_clip": 1.05813158,
      "balance_loss_mlp": 1.02583575,
      "epoch": 0.20248902783622919,
      "flos": 22564721600640.0,
      "grad_norm": 1.8466424481199166,
      "language_loss": 0.80921042,
      "learning_rate": 3.696077845743968e-06,
      "loss": 0.8315171,
      "num_input_tokens_seen": 35744355,
      "step": 1684,
      "time_per_iteration": 3.5204105377197266
    },
    {
      "auxiliary_loss_clip": 0.01215021,
      "auxiliary_loss_mlp": 0.01037451,
      "balance_loss_clip": 1.05734468,
      "balance_loss_mlp": 1.02710903,
      "epoch": 0.20260927072686827,
      "flos": 22709728805760.0,
      "grad_norm": 3.409134916094016,
      "language_loss": 0.73203027,
      "learning_rate": 3.69566491359004e-06,
      "loss": 0.75455499,
      "num_input_tokens_seen": 35761000,
      "step": 1685,
      "time_per_iteration": 2.602459192276001
    },
    {
      "auxiliary_loss_clip": 0.01184047,
      "auxiliary_loss_mlp": 0.01041367,
      "balance_loss_clip": 1.05686069,
      "balance_loss_mlp": 1.03007793,
      "epoch": 0.20272951361750738,
      "flos": 51023998650240.0,
      "grad_norm": 2.3044442012510777,
      "language_loss": 0.69552159,
      "learning_rate": 3.695251724205092e-06,
      "loss": 0.7177757,
      "num_input_tokens_seen": 35785360,
      "step": 1686,
      "time_per_iteration": 2.928880214691162
    },
    {
      "auxiliary_loss_clip": 0.01216673,
      "auxiliary_loss_mlp": 0.01035234,
      "balance_loss_clip": 1.06263804,
      "balance_loss_mlp": 1.02505374,
      "epoch": 0.20284975650814646,
      "flos": 26578133879040.0,
      "grad_norm": 1.6454935353654447,
      "language_loss": 0.86735886,
      "learning_rate": 3.6948382776518054e-06,
      "loss": 0.88987797,
      "num_input_tokens_seen": 35806065,
      "step": 1687,
      "time_per_iteration": 2.675960063934326
    },
    {
      "auxiliary_loss_clip": 0.01176942,
      "auxiliary_loss_mlp": 0.01043648,
      "balance_loss_clip": 1.05519176,
      "balance_loss_mlp": 1.03368163,
      "epoch": 0.20296999939878554,
      "flos": 16034222833920.0,
      "grad_norm": 3.422564855315413,
      "language_loss": 0.7927435,
      "learning_rate": 3.6944245739929e-06,
      "loss": 0.81494939,
      "num_input_tokens_seen": 35822225,
      "step": 1688,
      "time_per_iteration": 2.710784673690796
    },
    {
      "auxiliary_loss_clip": 0.01203204,
      "auxiliary_loss_mlp": 0.01043801,
      "balance_loss_clip": 1.06107306,
      "balance_loss_mlp": 1.03254163,
      "epoch": 0.20309024228942463,
      "flos": 19203374868480.0,
      "grad_norm": 7.121614655897142,
      "language_loss": 0.7185654,
      "learning_rate": 3.6940106132911332e-06,
      "loss": 0.74103546,
      "num_input_tokens_seen": 35839410,
      "step": 1689,
      "time_per_iteration": 2.6182799339294434
    },
    {
      "auxiliary_loss_clip": 0.01204453,
      "auxiliary_loss_mlp": 0.01037627,
      "balance_loss_clip": 1.06064963,
      "balance_loss_mlp": 1.02676725,
      "epoch": 0.20321048518006374,
      "flos": 22821087945600.0,
      "grad_norm": 3.0127473001887743,
      "language_loss": 0.88788921,
      "learning_rate": 3.6935963956093037e-06,
      "loss": 0.91031003,
      "num_input_tokens_seen": 35859495,
      "step": 1690,
      "time_per_iteration": 2.720550775527954
    },
    {
      "auxiliary_loss_clip": 0.01192746,
      "auxiliary_loss_mlp": 0.01038113,
      "balance_loss_clip": 1.05827355,
      "balance_loss_mlp": 1.02748561,
      "epoch": 0.20333072807070282,
      "flos": 19096397187840.0,
      "grad_norm": 1.889931560689134,
      "language_loss": 0.69490588,
      "learning_rate": 3.6931819210102474e-06,
      "loss": 0.71721447,
      "num_input_tokens_seen": 35878890,
      "step": 1691,
      "time_per_iteration": 2.8321774005889893
    },
    {
      "auxiliary_loss_clip": 0.01216064,
      "auxiliary_loss_mlp": 0.01029998,
      "balance_loss_clip": 1.05856526,
      "balance_loss_mlp": 1.01913226,
      "epoch": 0.2034509709613419,
      "flos": 18180962144640.0,
      "grad_norm": 5.4572864688174185,
      "language_loss": 0.8422972,
      "learning_rate": 3.6927671895568402e-06,
      "loss": 0.8647579,
      "num_input_tokens_seen": 35897950,
      "step": 1692,
      "time_per_iteration": 2.588818311691284
    },
    {
      "auxiliary_loss_clip": 0.0121611,
      "auxiliary_loss_mlp": 0.01036414,
      "balance_loss_clip": 1.06152439,
      "balance_loss_mlp": 1.02574492,
      "epoch": 0.20357121385198101,
      "flos": 22923899648640.0,
      "grad_norm": 3.3141171458972605,
      "language_loss": 0.86597788,
      "learning_rate": 3.692352201311996e-06,
      "loss": 0.88850307,
      "num_input_tokens_seen": 35916800,
      "step": 1693,
      "time_per_iteration": 2.8246896266937256
    },
    {
      "auxiliary_loss_clip": 0.01168714,
      "auxiliary_loss_mlp": 0.0103537,
      "balance_loss_clip": 1.05582273,
      "balance_loss_mlp": 1.02456975,
      "epoch": 0.2036914567426201,
      "flos": 20922131629440.0,
      "grad_norm": 10.031485493633443,
      "language_loss": 0.76880246,
      "learning_rate": 3.6919369563386687e-06,
      "loss": 0.79084331,
      "num_input_tokens_seen": 35936600,
      "step": 1694,
      "time_per_iteration": 2.7632648944854736
    },
    {
      "auxiliary_loss_clip": 0.01180218,
      "auxiliary_loss_mlp": 0.01029641,
      "balance_loss_clip": 1.05718493,
      "balance_loss_mlp": 1.01984191,
      "epoch": 0.20381169963325918,
      "flos": 15519155760000.0,
      "grad_norm": 3.695182405151826,
      "language_loss": 0.79577863,
      "learning_rate": 3.69152145469985e-06,
      "loss": 0.81787717,
      "num_input_tokens_seen": 35953645,
      "step": 1695,
      "time_per_iteration": 2.7080001831054688
    },
    {
      "auxiliary_loss_clip": 0.01160321,
      "auxiliary_loss_mlp": 0.01037944,
      "balance_loss_clip": 1.05185795,
      "balance_loss_mlp": 1.02747738,
      "epoch": 0.20393194252389826,
      "flos": 28833143760000.0,
      "grad_norm": 2.136562814459686,
      "language_loss": 0.81943333,
      "learning_rate": 3.691105696458572e-06,
      "loss": 0.841416,
      "num_input_tokens_seen": 35970940,
      "step": 1696,
      "time_per_iteration": 2.7780191898345947
    },
    {
      "auxiliary_loss_clip": 0.01215379,
      "auxiliary_loss_mlp": 0.01034074,
      "balance_loss_clip": 1.0616461,
      "balance_loss_mlp": 1.02426267,
      "epoch": 0.20405218541453737,
      "flos": 22488554810880.0,
      "grad_norm": 2.9553705179796386,
      "language_loss": 0.67877775,
      "learning_rate": 3.690689681677904e-06,
      "loss": 0.70127225,
      "num_input_tokens_seen": 35989410,
      "step": 1697,
      "time_per_iteration": 2.6194007396698
    },
    {
      "auxiliary_loss_clip": 0.01186905,
      "auxiliary_loss_mlp": 0.0103543,
      "balance_loss_clip": 1.05554366,
      "balance_loss_mlp": 1.02510667,
      "epoch": 0.20417242830517646,
      "flos": 25374408278400.0,
      "grad_norm": 1.9613400301558965,
      "language_loss": 0.88823247,
      "learning_rate": 3.690273410420956e-06,
      "loss": 0.91045582,
      "num_input_tokens_seen": 36009175,
      "step": 1698,
      "time_per_iteration": 2.7287704944610596
    },
    {
      "auxiliary_loss_clip": 0.01197688,
      "auxiliary_loss_mlp": 0.01030533,
      "balance_loss_clip": 1.05717802,
      "balance_loss_mlp": 1.01998258,
      "epoch": 0.20429267119581554,
      "flos": 14793078240000.0,
      "grad_norm": 2.752520543902167,
      "language_loss": 0.77102751,
      "learning_rate": 3.689856882750875e-06,
      "loss": 0.79330969,
      "num_input_tokens_seen": 36024375,
      "step": 1699,
      "time_per_iteration": 2.7311878204345703
    },
    {
      "auxiliary_loss_clip": 0.01193948,
      "auxiliary_loss_mlp": 0.01032458,
      "balance_loss_clip": 1.05663562,
      "balance_loss_mlp": 1.02276039,
      "epoch": 0.20441291408645465,
      "flos": 17781851151360.0,
      "grad_norm": 1.832459255344264,
      "language_loss": 0.785676,
      "learning_rate": 3.6894400987308486e-06,
      "loss": 0.80794007,
      "num_input_tokens_seen": 36041895,
      "step": 1700,
      "time_per_iteration": 2.592207670211792
    },
    {
      "auxiliary_loss_clip": 0.01202257,
      "auxiliary_loss_mlp": 0.01039268,
      "balance_loss_clip": 1.05735421,
      "balance_loss_mlp": 1.02747202,
      "epoch": 0.20453315697709373,
      "flos": 16435668211200.0,
      "grad_norm": 4.505765965966685,
      "language_loss": 0.85194415,
      "learning_rate": 3.6890230584241024e-06,
      "loss": 0.87435937,
      "num_input_tokens_seen": 36058825,
      "step": 1701,
      "time_per_iteration": 2.653635263442993
    },
    {
      "auxiliary_loss_clip": 0.01120877,
      "auxiliary_loss_mlp": 0.01013901,
      "balance_loss_clip": 1.03207219,
      "balance_loss_mlp": 1.01164782,
      "epoch": 0.20465339986773282,
      "flos": 66713085653760.0,
      "grad_norm": 1.0746719082313454,
      "language_loss": 0.66490388,
      "learning_rate": 3.6886057618939016e-06,
      "loss": 0.68625164,
      "num_input_tokens_seen": 36121645,
      "step": 1702,
      "time_per_iteration": 3.22568678855896
    },
    {
      "auxiliary_loss_clip": 0.01164045,
      "auxiliary_loss_mlp": 0.01035159,
      "balance_loss_clip": 1.05295658,
      "balance_loss_mlp": 1.02488267,
      "epoch": 0.2047736427583719,
      "flos": 41974114924800.0,
      "grad_norm": 2.25800134615194,
      "language_loss": 0.69488227,
      "learning_rate": 3.6881882092035492e-06,
      "loss": 0.7168743,
      "num_input_tokens_seen": 36143030,
      "step": 1703,
      "time_per_iteration": 2.8601865768432617
    },
    {
      "auxiliary_loss_clip": 0.01092428,
      "auxiliary_loss_mlp": 0.00755837,
      "balance_loss_clip": 1.03160381,
      "balance_loss_mlp": 1.00131416,
      "epoch": 0.204893885649011,
      "flos": 69940878641280.0,
      "grad_norm": 0.9337215248100351,
      "language_loss": 0.61198246,
      "learning_rate": 3.6877704004163873e-06,
      "loss": 0.63046515,
      "num_input_tokens_seen": 36203435,
      "step": 1704,
      "time_per_iteration": 3.44199538230896
    },
    {
      "auxiliary_loss_clip": 0.0121476,
      "auxiliary_loss_mlp": 0.01035604,
      "balance_loss_clip": 1.05945885,
      "balance_loss_mlp": 1.02559614,
      "epoch": 0.2050141285396501,
      "flos": 22200012858240.0,
      "grad_norm": 4.583968677116409,
      "language_loss": 0.77592146,
      "learning_rate": 3.6873523355957984e-06,
      "loss": 0.79842508,
      "num_input_tokens_seen": 36222435,
      "step": 1705,
      "time_per_iteration": 3.5690698623657227
    },
    {
      "auxiliary_loss_clip": 0.01119163,
      "auxiliary_loss_mlp": 0.01002587,
      "balance_loss_clip": 1.03022861,
      "balance_loss_mlp": 1.0002625,
      "epoch": 0.20513437143028918,
      "flos": 46283721730560.0,
      "grad_norm": 0.9892376734725632,
      "language_loss": 0.64065397,
      "learning_rate": 3.686934014805201e-06,
      "loss": 0.66187143,
      "num_input_tokens_seen": 36273065,
      "step": 1706,
      "time_per_iteration": 3.0296056270599365
    },
    {
      "auxiliary_loss_clip": 0.01198525,
      "auxiliary_loss_mlp": 0.0103426,
      "balance_loss_clip": 1.05843735,
      "balance_loss_mlp": 1.02348971,
      "epoch": 0.20525461432092829,
      "flos": 21904324099200.0,
      "grad_norm": 1.859275298104681,
      "language_loss": 0.81331307,
      "learning_rate": 3.6865154381080552e-06,
      "loss": 0.83564091,
      "num_input_tokens_seen": 36293750,
      "step": 1707,
      "time_per_iteration": 2.730319023132324
    },
    {
      "auxiliary_loss_clip": 0.01125023,
      "auxiliary_loss_mlp": 0.01036949,
      "balance_loss_clip": 1.04747486,
      "balance_loss_mlp": 1.02692986,
      "epoch": 0.20537485721156737,
      "flos": 21214264942080.0,
      "grad_norm": 2.039175482377376,
      "language_loss": 0.82293898,
      "learning_rate": 3.6860966055678585e-06,
      "loss": 0.84455872,
      "num_input_tokens_seen": 36310105,
      "step": 1708,
      "time_per_iteration": 3.699664354324341
    },
    {
      "auxiliary_loss_clip": 0.01198678,
      "auxiliary_loss_mlp": 0.0102908,
      "balance_loss_clip": 1.05847085,
      "balance_loss_mlp": 1.0183686,
      "epoch": 0.20549510010220645,
      "flos": 20191205773440.0,
      "grad_norm": 1.8475223525157303,
      "language_loss": 0.86486328,
      "learning_rate": 3.685677517248147e-06,
      "loss": 0.88714087,
      "num_input_tokens_seen": 36328995,
      "step": 1709,
      "time_per_iteration": 4.453352212905884
    },
    {
      "auxiliary_loss_clip": 0.01184796,
      "auxiliary_loss_mlp": 0.00764922,
      "balance_loss_clip": 1.06096375,
      "balance_loss_mlp": 1.00054121,
      "epoch": 0.20561534299284553,
      "flos": 17016702612480.0,
      "grad_norm": 3.2977189738829145,
      "language_loss": 0.8049196,
      "learning_rate": 3.6852581732124967e-06,
      "loss": 0.82441688,
      "num_input_tokens_seen": 36346340,
      "step": 1710,
      "time_per_iteration": 2.7561628818511963
    },
    {
      "auxiliary_loss_clip": 0.01197343,
      "auxiliary_loss_mlp": 0.01032608,
      "balance_loss_clip": 1.05760574,
      "balance_loss_mlp": 1.02190292,
      "epoch": 0.20573558588348465,
      "flos": 22890467064960.0,
      "grad_norm": 2.276397483175922,
      "language_loss": 0.76492828,
      "learning_rate": 3.6848385735245213e-06,
      "loss": 0.78722781,
      "num_input_tokens_seen": 36365430,
      "step": 1711,
      "time_per_iteration": 2.6317827701568604
    },
    {
      "auxiliary_loss_clip": 0.01188072,
      "auxiliary_loss_mlp": 0.01027688,
      "balance_loss_clip": 1.05526495,
      "balance_loss_mlp": 1.01803827,
      "epoch": 0.20585582877412373,
      "flos": 24643123286400.0,
      "grad_norm": 2.190492230157755,
      "language_loss": 0.86416459,
      "learning_rate": 3.6844187182478734e-06,
      "loss": 0.8863222,
      "num_input_tokens_seen": 36386285,
      "step": 1712,
      "time_per_iteration": 2.729926586151123
    },
    {
      "auxiliary_loss_clip": 0.01171509,
      "auxiliary_loss_mlp": 0.01035112,
      "balance_loss_clip": 1.05101156,
      "balance_loss_mlp": 1.02492523,
      "epoch": 0.2059760716647628,
      "flos": 24206952435840.0,
      "grad_norm": 2.4523242675344354,
      "language_loss": 0.75339741,
      "learning_rate": 3.683998607446246e-06,
      "loss": 0.77546358,
      "num_input_tokens_seen": 36404935,
      "step": 1713,
      "time_per_iteration": 2.759671688079834
    },
    {
      "auxiliary_loss_clip": 0.01199785,
      "auxiliary_loss_mlp": 0.01035975,
      "balance_loss_clip": 1.05978942,
      "balance_loss_mlp": 1.02626586,
      "epoch": 0.20609631455540192,
      "flos": 20229522606720.0,
      "grad_norm": 2.265295381206127,
      "language_loss": 0.74892426,
      "learning_rate": 3.6835782411833686e-06,
      "loss": 0.7712819,
      "num_input_tokens_seen": 36424455,
      "step": 1714,
      "time_per_iteration": 2.640052556991577
    },
    {
      "auxiliary_loss_clip": 0.01155996,
      "auxiliary_loss_mlp": 0.0103664,
      "balance_loss_clip": 1.05100727,
      "balance_loss_mlp": 1.0264771,
      "epoch": 0.206216557446041,
      "flos": 19864957518720.0,
      "grad_norm": 1.8692162648044905,
      "language_loss": 0.7410692,
      "learning_rate": 3.68315761952301e-06,
      "loss": 0.7629956,
      "num_input_tokens_seen": 36441685,
      "step": 1715,
      "time_per_iteration": 2.7175722122192383
    },
    {
      "auxiliary_loss_clip": 0.01212391,
      "auxiliary_loss_mlp": 0.01039047,
      "balance_loss_clip": 1.05879593,
      "balance_loss_mlp": 1.02861595,
      "epoch": 0.2063368003366801,
      "flos": 24096311568000.0,
      "grad_norm": 2.301196399137744,
      "language_loss": 0.83216989,
      "learning_rate": 3.6827367425289797e-06,
      "loss": 0.85468429,
      "num_input_tokens_seen": 36461460,
      "step": 1716,
      "time_per_iteration": 2.624420404434204
    },
    {
      "auxiliary_loss_clip": 0.01184779,
      "auxiliary_loss_mlp": 0.01039491,
      "balance_loss_clip": 1.0575155,
      "balance_loss_mlp": 1.02918565,
      "epoch": 0.2064570432273192,
      "flos": 20340163474560.0,
      "grad_norm": 3.0858014415244703,
      "language_loss": 0.72680068,
      "learning_rate": 3.6823156102651225e-06,
      "loss": 0.74904341,
      "num_input_tokens_seen": 36479615,
      "step": 1717,
      "time_per_iteration": 2.7593390941619873
    },
    {
      "auxiliary_loss_clip": 0.01130672,
      "auxiliary_loss_mlp": 0.01031376,
      "balance_loss_clip": 1.05279279,
      "balance_loss_mlp": 1.02186918,
      "epoch": 0.20657728611795828,
      "flos": 20520363029760.0,
      "grad_norm": 3.408372898553658,
      "language_loss": 0.70911294,
      "learning_rate": 3.6818942227953257e-06,
      "loss": 0.73073345,
      "num_input_tokens_seen": 36500160,
      "step": 1718,
      "time_per_iteration": 2.809767723083496
    },
    {
      "auxiliary_loss_clip": 0.0116924,
      "auxiliary_loss_mlp": 0.01035104,
      "balance_loss_clip": 1.05308485,
      "balance_loss_mlp": 1.02509046,
      "epoch": 0.20669752900859736,
      "flos": 21799285752960.0,
      "grad_norm": 1.9109493014703354,
      "language_loss": 0.69148213,
      "learning_rate": 3.681472580183512e-06,
      "loss": 0.71352559,
      "num_input_tokens_seen": 36518810,
      "step": 1719,
      "time_per_iteration": 2.717984914779663
    },
    {
      "auxiliary_loss_clip": 0.0119411,
      "auxiliary_loss_mlp": 0.01032479,
      "balance_loss_clip": 1.05733609,
      "balance_loss_mlp": 1.02352047,
      "epoch": 0.20681777189923645,
      "flos": 15122020014720.0,
      "grad_norm": 2.3698632069511874,
      "language_loss": 0.8635602,
      "learning_rate": 3.6810506824936455e-06,
      "loss": 0.88582611,
      "num_input_tokens_seen": 36536890,
      "step": 1720,
      "time_per_iteration": 2.652461528778076
    },
    {
      "auxiliary_loss_clip": 0.01089195,
      "auxiliary_loss_mlp": 0.01006138,
      "balance_loss_clip": 1.02389336,
      "balance_loss_mlp": 1.00369406,
      "epoch": 0.20693801478987556,
      "flos": 56481021509760.0,
      "grad_norm": 1.1881034499513454,
      "language_loss": 0.62472129,
      "learning_rate": 3.680628529789726e-06,
      "loss": 0.64567459,
      "num_input_tokens_seen": 36589300,
      "step": 1721,
      "time_per_iteration": 3.097033977508545
    },
    {
      "auxiliary_loss_clip": 0.01220505,
      "auxiliary_loss_mlp": 0.0103573,
      "balance_loss_clip": 1.06276131,
      "balance_loss_mlp": 1.02420259,
      "epoch": 0.20705825768051464,
      "flos": 21614201948160.0,
      "grad_norm": 1.8158738435769777,
      "language_loss": 0.86948621,
      "learning_rate": 3.680206122135796e-06,
      "loss": 0.8920486,
      "num_input_tokens_seen": 36609905,
      "step": 1722,
      "time_per_iteration": 2.63529634475708
    },
    {
      "auxiliary_loss_clip": 0.0116041,
      "auxiliary_loss_mlp": 0.01039126,
      "balance_loss_clip": 1.05478978,
      "balance_loss_mlp": 1.0291481,
      "epoch": 0.20717850057115372,
      "flos": 25848895962240.0,
      "grad_norm": 6.789407037999269,
      "language_loss": 0.78595388,
      "learning_rate": 3.6797834595959323e-06,
      "loss": 0.8079493,
      "num_input_tokens_seen": 36629805,
      "step": 1723,
      "time_per_iteration": 2.755037307739258
    },
    {
      "auxiliary_loss_clip": 0.01142468,
      "auxiliary_loss_mlp": 0.01033577,
      "balance_loss_clip": 1.04797077,
      "balance_loss_mlp": 1.02264571,
      "epoch": 0.20729874346179283,
      "flos": 29130807767040.0,
      "grad_norm": 4.367857733743932,
      "language_loss": 0.78193772,
      "learning_rate": 3.679360542234254e-06,
      "loss": 0.80369818,
      "num_input_tokens_seen": 36649150,
      "step": 1724,
      "time_per_iteration": 2.886676788330078
    },
    {
      "auxiliary_loss_clip": 0.01175821,
      "auxiliary_loss_mlp": 0.0076487,
      "balance_loss_clip": 1.05178928,
      "balance_loss_mlp": 1.00048518,
      "epoch": 0.20741898635243192,
      "flos": 29023363209600.0,
      "grad_norm": 1.9048844373106568,
      "language_loss": 0.72284812,
      "learning_rate": 3.678937370114916e-06,
      "loss": 0.74225509,
      "num_input_tokens_seen": 36668955,
      "step": 1725,
      "time_per_iteration": 2.7701542377471924
    },
    {
      "auxiliary_loss_clip": 0.01179496,
      "auxiliary_loss_mlp": 0.01037124,
      "balance_loss_clip": 1.05552888,
      "balance_loss_mlp": 1.02787375,
      "epoch": 0.207539229243071,
      "flos": 15559447841280.0,
      "grad_norm": 1.938857728205364,
      "language_loss": 0.78481722,
      "learning_rate": 3.678513943302114e-06,
      "loss": 0.80698341,
      "num_input_tokens_seen": 36685730,
      "step": 1726,
      "time_per_iteration": 2.700315237045288
    },
    {
      "auxiliary_loss_clip": 0.01212311,
      "auxiliary_loss_mlp": 0.01037629,
      "balance_loss_clip": 1.06042004,
      "balance_loss_mlp": 1.02818167,
      "epoch": 0.20765947213371008,
      "flos": 20521081301760.0,
      "grad_norm": 1.8202160737902255,
      "language_loss": 0.85487711,
      "learning_rate": 3.678090261860082e-06,
      "loss": 0.8773765,
      "num_input_tokens_seen": 36705460,
      "step": 1727,
      "time_per_iteration": 2.643371820449829
    },
    {
      "auxiliary_loss_clip": 0.01168509,
      "auxiliary_loss_mlp": 0.010306,
      "balance_loss_clip": 1.04963374,
      "balance_loss_mlp": 1.02073526,
      "epoch": 0.2077797150243492,
      "flos": 19354415558400.0,
      "grad_norm": 2.0320129293750253,
      "language_loss": 0.77687228,
      "learning_rate": 3.6776663258530906e-06,
      "loss": 0.79886341,
      "num_input_tokens_seen": 36724110,
      "step": 1728,
      "time_per_iteration": 2.7693707942962646
    },
    {
      "auxiliary_loss_clip": 0.01198529,
      "auxiliary_loss_mlp": 0.01032354,
      "balance_loss_clip": 1.05635071,
      "balance_loss_mlp": 1.02203071,
      "epoch": 0.20789995791498828,
      "flos": 21829952989440.0,
      "grad_norm": 1.9668632081977668,
      "language_loss": 0.71253777,
      "learning_rate": 3.6772421353454516e-06,
      "loss": 0.73484665,
      "num_input_tokens_seen": 36742705,
      "step": 1729,
      "time_per_iteration": 2.6397814750671387
    },
    {
      "auxiliary_loss_clip": 0.01197347,
      "auxiliary_loss_mlp": 0.01037826,
      "balance_loss_clip": 1.05975199,
      "balance_loss_mlp": 1.02698362,
      "epoch": 0.20802020080562736,
      "flos": 23148844571520.0,
      "grad_norm": 1.8654733944421404,
      "language_loss": 0.8829729,
      "learning_rate": 3.6768176904015153e-06,
      "loss": 0.90532464,
      "num_input_tokens_seen": 36762510,
      "step": 1730,
      "time_per_iteration": 2.6546359062194824
    },
    {
      "auxiliary_loss_clip": 0.01198077,
      "auxiliary_loss_mlp": 0.01034587,
      "balance_loss_clip": 1.05658317,
      "balance_loss_mlp": 1.02432895,
      "epoch": 0.20814044369626647,
      "flos": 23072677781760.0,
      "grad_norm": 1.919141280248307,
      "language_loss": 0.59853429,
      "learning_rate": 3.6763929910856674e-06,
      "loss": 0.62086093,
      "num_input_tokens_seen": 36780960,
      "step": 1731,
      "time_per_iteration": 3.6072425842285156
    },
    {
      "auxiliary_loss_clip": 0.01198715,
      "auxiliary_loss_mlp": 0.01031528,
      "balance_loss_clip": 1.05895197,
      "balance_loss_mlp": 1.02169299,
      "epoch": 0.20826068658690555,
      "flos": 19608016556160.0,
      "grad_norm": 2.34189815347847,
      "language_loss": 0.77571517,
      "learning_rate": 3.6759680374623365e-06,
      "loss": 0.79801762,
      "num_input_tokens_seen": 36798875,
      "step": 1732,
      "time_per_iteration": 2.633657932281494
    },
    {
      "auxiliary_loss_clip": 0.01214125,
      "auxiliary_loss_mlp": 0.01031684,
      "balance_loss_clip": 1.06242466,
      "balance_loss_mlp": 1.02196276,
      "epoch": 0.20838092947754464,
      "flos": 25374049142400.0,
      "grad_norm": 2.3433147656913462,
      "language_loss": 0.75293887,
      "learning_rate": 3.675542829595986e-06,
      "loss": 0.77539694,
      "num_input_tokens_seen": 36818540,
      "step": 1733,
      "time_per_iteration": 3.5939900875091553
    },
    {
      "auxiliary_loss_clip": 0.01185841,
      "auxiliary_loss_mlp": 0.01033637,
      "balance_loss_clip": 1.05744135,
      "balance_loss_mlp": 1.0234741,
      "epoch": 0.20850117236818372,
      "flos": 24061729749120.0,
      "grad_norm": 1.4719628875557313,
      "language_loss": 0.7935015,
      "learning_rate": 3.6751173675511213e-06,
      "loss": 0.81569636,
      "num_input_tokens_seen": 36840585,
      "step": 1734,
      "time_per_iteration": 2.6695034503936768
    },
    {
      "auxiliary_loss_clip": 0.01179629,
      "auxiliary_loss_mlp": 0.01027226,
      "balance_loss_clip": 1.05138862,
      "balance_loss_mlp": 1.01733112,
      "epoch": 0.20862141525882283,
      "flos": 20077799558400.0,
      "grad_norm": 2.1789433624444032,
      "language_loss": 0.87397063,
      "learning_rate": 3.674691651392283e-06,
      "loss": 0.89603925,
      "num_input_tokens_seen": 36858255,
      "step": 1735,
      "time_per_iteration": 3.539163112640381
    },
    {
      "auxiliary_loss_clip": 0.01188342,
      "auxiliary_loss_mlp": 0.01030565,
      "balance_loss_clip": 1.05802441,
      "balance_loss_mlp": 1.01995552,
      "epoch": 0.2087416581494619,
      "flos": 39015183237120.0,
      "grad_norm": 2.1752788467259085,
      "language_loss": 0.75683272,
      "learning_rate": 3.674265681184053e-06,
      "loss": 0.7790218,
      "num_input_tokens_seen": 36881515,
      "step": 1736,
      "time_per_iteration": 3.7494421005249023
    },
    {
      "auxiliary_loss_clip": 0.01184734,
      "auxiliary_loss_mlp": 0.01032751,
      "balance_loss_clip": 1.05622435,
      "balance_loss_mlp": 1.02240384,
      "epoch": 0.208861901040101,
      "flos": 26101994169600.0,
      "grad_norm": 2.251027785148863,
      "language_loss": 0.86255765,
      "learning_rate": 3.6738394569910504e-06,
      "loss": 0.88473248,
      "num_input_tokens_seen": 36902055,
      "step": 1737,
      "time_per_iteration": 2.6958632469177246
    },
    {
      "auxiliary_loss_clip": 0.0120149,
      "auxiliary_loss_mlp": 0.01034242,
      "balance_loss_clip": 1.06069207,
      "balance_loss_mlp": 1.02358425,
      "epoch": 0.2089821439307401,
      "flos": 28398732675840.0,
      "grad_norm": 5.195665248401253,
      "language_loss": 0.83122218,
      "learning_rate": 3.6734129788779333e-06,
      "loss": 0.85357952,
      "num_input_tokens_seen": 36921230,
      "step": 1738,
      "time_per_iteration": 2.7096691131591797
    },
    {
      "auxiliary_loss_clip": 0.01167545,
      "auxiliary_loss_mlp": 0.01034392,
      "balance_loss_clip": 1.05646694,
      "balance_loss_mlp": 1.02545702,
      "epoch": 0.2091023868213792,
      "flos": 21069616872960.0,
      "grad_norm": 1.7629681885007642,
      "language_loss": 0.90293068,
      "learning_rate": 3.6729862469093976e-06,
      "loss": 0.92495,
      "num_input_tokens_seen": 36940325,
      "step": 1739,
      "time_per_iteration": 2.7805280685424805
    },
    {
      "auxiliary_loss_clip": 0.01171152,
      "auxiliary_loss_mlp": 0.01031732,
      "balance_loss_clip": 1.05247748,
      "balance_loss_mlp": 1.02193928,
      "epoch": 0.20922262971201827,
      "flos": 22455481363200.0,
      "grad_norm": 2.127832818135102,
      "language_loss": 0.82559264,
      "learning_rate": 3.6725592611501782e-06,
      "loss": 0.84762144,
      "num_input_tokens_seen": 36959000,
      "step": 1740,
      "time_per_iteration": 2.7038371562957764
    },
    {
      "auxiliary_loss_clip": 0.01197738,
      "auxiliary_loss_mlp": 0.01037388,
      "balance_loss_clip": 1.05664849,
      "balance_loss_mlp": 1.02729082,
      "epoch": 0.20934287260265738,
      "flos": 27852244179840.0,
      "grad_norm": 1.7592717649007117,
      "language_loss": 0.76531589,
      "learning_rate": 3.6721320216650496e-06,
      "loss": 0.7876671,
      "num_input_tokens_seen": 36979615,
      "step": 1741,
      "time_per_iteration": 2.7016961574554443
    },
    {
      "auxiliary_loss_clip": 0.0118254,
      "auxiliary_loss_mlp": 0.01034075,
      "balance_loss_clip": 1.05787027,
      "balance_loss_mlp": 1.02376366,
      "epoch": 0.20946311549329646,
      "flos": 16435309075200.0,
      "grad_norm": 1.8820783358753295,
      "language_loss": 0.83724016,
      "learning_rate": 3.6717045285188215e-06,
      "loss": 0.85940635,
      "num_input_tokens_seen": 36997310,
      "step": 1742,
      "time_per_iteration": 2.648798942565918
    },
    {
      "auxiliary_loss_clip": 0.01143482,
      "auxiliary_loss_mlp": 0.01036963,
      "balance_loss_clip": 1.04817343,
      "balance_loss_mlp": 1.02705038,
      "epoch": 0.20958335838393555,
      "flos": 22492720788480.0,
      "grad_norm": 2.358981630364755,
      "language_loss": 0.86743903,
      "learning_rate": 3.671276781776346e-06,
      "loss": 0.88924348,
      "num_input_tokens_seen": 37015965,
      "step": 1743,
      "time_per_iteration": 2.781121015548706
    },
    {
      "auxiliary_loss_clip": 0.01177577,
      "auxiliary_loss_mlp": 0.01030095,
      "balance_loss_clip": 1.05394745,
      "balance_loss_mlp": 1.01983714,
      "epoch": 0.20970360127457463,
      "flos": 25224768218880.0,
      "grad_norm": 3.6156926356358157,
      "language_loss": 0.67300093,
      "learning_rate": 3.6708487815025128e-06,
      "loss": 0.69507772,
      "num_input_tokens_seen": 37036545,
      "step": 1744,
      "time_per_iteration": 2.7319209575653076
    },
    {
      "auxiliary_loss_clip": 0.01166208,
      "auxiliary_loss_mlp": 0.01037588,
      "balance_loss_clip": 1.05317521,
      "balance_loss_mlp": 1.02743173,
      "epoch": 0.20982384416521374,
      "flos": 18479164855680.0,
      "grad_norm": 2.3701408808397373,
      "language_loss": 0.74224412,
      "learning_rate": 3.6704205277622463e-06,
      "loss": 0.76428211,
      "num_input_tokens_seen": 37054985,
      "step": 1745,
      "time_per_iteration": 2.7446377277374268
    },
    {
      "auxiliary_loss_clip": 0.01185169,
      "auxiliary_loss_mlp": 0.01034407,
      "balance_loss_clip": 1.05410028,
      "balance_loss_mlp": 1.02463818,
      "epoch": 0.20994408705585282,
      "flos": 25373546352000.0,
      "grad_norm": 2.020786924539658,
      "language_loss": 0.80790091,
      "learning_rate": 3.6699920206205146e-06,
      "loss": 0.8300966,
      "num_input_tokens_seen": 37075725,
      "step": 1746,
      "time_per_iteration": 2.7242655754089355
    },
    {
      "auxiliary_loss_clip": 0.01197023,
      "auxiliary_loss_mlp": 0.01035682,
      "balance_loss_clip": 1.05551553,
      "balance_loss_mlp": 1.02551413,
      "epoch": 0.2100643299464919,
      "flos": 21320955313920.0,
      "grad_norm": 4.152939383314306,
      "language_loss": 0.82030648,
      "learning_rate": 3.669563260142321e-06,
      "loss": 0.84263355,
      "num_input_tokens_seen": 37094615,
      "step": 1747,
      "time_per_iteration": 2.7027008533477783
    },
    {
      "auxiliary_loss_clip": 0.01181268,
      "auxiliary_loss_mlp": 0.01033783,
      "balance_loss_clip": 1.05859828,
      "balance_loss_mlp": 1.0233227,
      "epoch": 0.21018457283713102,
      "flos": 19354379644800.0,
      "grad_norm": 2.299356492945342,
      "language_loss": 0.84258187,
      "learning_rate": 3.6691342463927083e-06,
      "loss": 0.86473238,
      "num_input_tokens_seen": 37113610,
      "step": 1748,
      "time_per_iteration": 2.6373085975646973
    },
    {
      "auxiliary_loss_clip": 0.01173526,
      "auxiliary_loss_mlp": 0.01035403,
      "balance_loss_clip": 1.05430388,
      "balance_loss_mlp": 1.02457297,
      "epoch": 0.2103048157277701,
      "flos": 28330035914880.0,
      "grad_norm": 1.7652732312417194,
      "language_loss": 0.82074702,
      "learning_rate": 3.668704979436758e-06,
      "loss": 0.84283632,
      "num_input_tokens_seen": 37133705,
      "step": 1749,
      "time_per_iteration": 2.741953134536743
    },
    {
      "auxiliary_loss_clip": 0.01178513,
      "auxiliary_loss_mlp": 0.01033948,
      "balance_loss_clip": 1.05456066,
      "balance_loss_mlp": 1.02454281,
      "epoch": 0.21042505861840918,
      "flos": 17457290835840.0,
      "grad_norm": 2.5127510584124977,
      "language_loss": 0.78787661,
      "learning_rate": 3.668275459339588e-06,
      "loss": 0.81000119,
      "num_input_tokens_seen": 37152185,
      "step": 1750,
      "time_per_iteration": 2.6339824199676514
    },
    {
      "auxiliary_loss_clip": 0.01214409,
      "auxiliary_loss_mlp": 0.01037802,
      "balance_loss_clip": 1.06183457,
      "balance_loss_mlp": 1.02819419,
      "epoch": 0.21054530150904827,
      "flos": 14209817195520.0,
      "grad_norm": 2.016458935429845,
      "language_loss": 0.79947603,
      "learning_rate": 3.667845686166358e-06,
      "loss": 0.82199812,
      "num_input_tokens_seen": 37169110,
      "step": 1751,
      "time_per_iteration": 2.621464252471924
    },
    {
      "auxiliary_loss_clip": 0.01153449,
      "auxiliary_loss_mlp": 0.01042917,
      "balance_loss_clip": 1.05179548,
      "balance_loss_mlp": 1.03197348,
      "epoch": 0.21066554439968738,
      "flos": 18618210403200.0,
      "grad_norm": 1.9573107428860472,
      "language_loss": 0.86366576,
      "learning_rate": 3.6674156599822634e-06,
      "loss": 0.88562942,
      "num_input_tokens_seen": 37184905,
      "step": 1752,
      "time_per_iteration": 2.6789586544036865
    },
    {
      "auxiliary_loss_clip": 0.01158539,
      "auxiliary_loss_mlp": 0.01038968,
      "balance_loss_clip": 1.05080712,
      "balance_loss_mlp": 1.02892423,
      "epoch": 0.21078578729032646,
      "flos": 23658883741440.0,
      "grad_norm": 1.8460760120628468,
      "language_loss": 0.81502587,
      "learning_rate": 3.666985380852539e-06,
      "loss": 0.83700097,
      "num_input_tokens_seen": 37203910,
      "step": 1753,
      "time_per_iteration": 2.8073105812072754
    },
    {
      "auxiliary_loss_clip": 0.01182176,
      "auxiliary_loss_mlp": 0.01034425,
      "balance_loss_clip": 1.05574274,
      "balance_loss_mlp": 1.02417254,
      "epoch": 0.21090603018096554,
      "flos": 29346379240320.0,
      "grad_norm": 3.300729141016106,
      "language_loss": 0.7408824,
      "learning_rate": 3.6665548488424576e-06,
      "loss": 0.76304841,
      "num_input_tokens_seen": 37222670,
      "step": 1754,
      "time_per_iteration": 2.734341859817505
    },
    {
      "auxiliary_loss_clip": 0.01219819,
      "auxiliary_loss_mlp": 0.01038548,
      "balance_loss_clip": 1.06279731,
      "balance_loss_mlp": 1.02700293,
      "epoch": 0.21102627307160465,
      "flos": 23261245205760.0,
      "grad_norm": 1.7371224417138964,
      "language_loss": 0.87847388,
      "learning_rate": 3.6661240640173307e-06,
      "loss": 0.9010576,
      "num_input_tokens_seen": 37244140,
      "step": 1755,
      "time_per_iteration": 2.610875368118286
    },
    {
      "auxiliary_loss_clip": 0.01083327,
      "auxiliary_loss_mlp": 0.01005329,
      "balance_loss_clip": 1.02822673,
      "balance_loss_mlp": 1.0028857,
      "epoch": 0.21114651596224374,
      "flos": 54633454577280.0,
      "grad_norm": 0.8591838570941044,
      "language_loss": 0.57903075,
      "learning_rate": 3.6656930264425085e-06,
      "loss": 0.59991729,
      "num_input_tokens_seen": 37308185,
      "step": 1756,
      "time_per_iteration": 3.296066999435425
    },
    {
      "auxiliary_loss_clip": 0.01214418,
      "auxiliary_loss_mlp": 0.01034107,
      "balance_loss_clip": 1.06040132,
      "balance_loss_mlp": 1.02350914,
      "epoch": 0.21126675885288282,
      "flos": 21543314457600.0,
      "grad_norm": 2.161858495790297,
      "language_loss": 0.75665987,
      "learning_rate": 3.665261736183378e-06,
      "loss": 0.77914518,
      "num_input_tokens_seen": 37328220,
      "step": 1757,
      "time_per_iteration": 3.4858505725860596
    },
    {
      "auxiliary_loss_clip": 0.01177423,
      "auxiliary_loss_mlp": 0.010378,
      "balance_loss_clip": 1.06085145,
      "balance_loss_mlp": 1.02665997,
      "epoch": 0.2113870017435219,
      "flos": 10961876678400.0,
      "grad_norm": 2.5292912417616877,
      "language_loss": 0.887061,
      "learning_rate": 3.664830193305366e-06,
      "loss": 0.9092133,
      "num_input_tokens_seen": 37345995,
      "step": 1758,
      "time_per_iteration": 2.6865692138671875
    },
    {
      "auxiliary_loss_clip": 0.01163439,
      "auxiliary_loss_mlp": 0.01037385,
      "balance_loss_clip": 1.05148101,
      "balance_loss_mlp": 1.0252136,
      "epoch": 0.211507244634161,
      "flos": 16653825463680.0,
      "grad_norm": 2.4519806190875686,
      "language_loss": 0.77273959,
      "learning_rate": 3.6643983978739373e-06,
      "loss": 0.79474783,
      "num_input_tokens_seen": 37362610,
      "step": 1759,
      "time_per_iteration": 3.671708345413208
    },
    {
      "auxiliary_loss_clip": 0.01180031,
      "auxiliary_loss_mlp": 0.01037246,
      "balance_loss_clip": 1.05958152,
      "balance_loss_mlp": 1.02655911,
      "epoch": 0.2116274875248001,
      "flos": 20954091755520.0,
      "grad_norm": 2.141732757669399,
      "language_loss": 0.8240115,
      "learning_rate": 3.663966349954596e-06,
      "loss": 0.84618425,
      "num_input_tokens_seen": 37382790,
      "step": 1760,
      "time_per_iteration": 2.6791062355041504
    },
    {
      "auxiliary_loss_clip": 0.01108013,
      "auxiliary_loss_mlp": 0.01002311,
      "balance_loss_clip": 1.02802515,
      "balance_loss_mlp": 0.99974787,
      "epoch": 0.21174773041543918,
      "flos": 68196949424640.0,
      "grad_norm": 0.7923717963250839,
      "language_loss": 0.59683514,
      "learning_rate": 3.6635340496128816e-06,
      "loss": 0.6179384,
      "num_input_tokens_seen": 37439720,
      "step": 1761,
      "time_per_iteration": 4.90225625038147
    },
    {
      "auxiliary_loss_clip": 0.01151439,
      "auxiliary_loss_mlp": 0.01034277,
      "balance_loss_clip": 1.05331182,
      "balance_loss_mlp": 1.02493107,
      "epoch": 0.2118679733060783,
      "flos": 20668315150080.0,
      "grad_norm": 2.625622024097744,
      "language_loss": 0.92715335,
      "learning_rate": 3.6631014969143747e-06,
      "loss": 0.94901049,
      "num_input_tokens_seen": 37459410,
      "step": 1762,
      "time_per_iteration": 2.727818727493286
    },
    {
      "auxiliary_loss_clip": 0.01200936,
      "auxiliary_loss_mlp": 0.01032751,
      "balance_loss_clip": 1.0620116,
      "balance_loss_mlp": 1.02204633,
      "epoch": 0.21198821619671737,
      "flos": 23223431162880.0,
      "grad_norm": 1.8860442645353335,
      "language_loss": 0.88795447,
      "learning_rate": 3.662668691924693e-06,
      "loss": 0.91029125,
      "num_input_tokens_seen": 37480460,
      "step": 1763,
      "time_per_iteration": 2.6984100341796875
    },
    {
      "auxiliary_loss_clip": 0.01169212,
      "auxiliary_loss_mlp": 0.01038787,
      "balance_loss_clip": 1.05488932,
      "balance_loss_mlp": 1.02728963,
      "epoch": 0.21210845908735645,
      "flos": 24498547044480.0,
      "grad_norm": 1.9561302674551908,
      "language_loss": 0.70958638,
      "learning_rate": 3.6622356347094927e-06,
      "loss": 0.73166645,
      "num_input_tokens_seen": 37502025,
      "step": 1764,
      "time_per_iteration": 2.7697904109954834
    },
    {
      "auxiliary_loss_clip": 0.01171342,
      "auxiliary_loss_mlp": 0.01041509,
      "balance_loss_clip": 1.05306935,
      "balance_loss_mlp": 1.02965963,
      "epoch": 0.21222870197799554,
      "flos": 27089789160960.0,
      "grad_norm": 1.9054738172142327,
      "language_loss": 0.78225625,
      "learning_rate": 3.6618023253344684e-06,
      "loss": 0.80438471,
      "num_input_tokens_seen": 37520885,
      "step": 1765,
      "time_per_iteration": 2.73400616645813
    },
    {
      "auxiliary_loss_clip": 0.01198875,
      "auxiliary_loss_mlp": 0.01038019,
      "balance_loss_clip": 1.05706179,
      "balance_loss_mlp": 1.02733743,
      "epoch": 0.21234894486863465,
      "flos": 16873850223360.0,
      "grad_norm": 1.775025470344949,
      "language_loss": 0.83533955,
      "learning_rate": 3.6613687638653527e-06,
      "loss": 0.85770845,
      "num_input_tokens_seen": 37539055,
      "step": 1766,
      "time_per_iteration": 2.6351661682128906
    },
    {
      "auxiliary_loss_clip": 0.01183042,
      "auxiliary_loss_mlp": 0.01030063,
      "balance_loss_clip": 1.06089258,
      "balance_loss_mlp": 1.01999593,
      "epoch": 0.21246918775927373,
      "flos": 23474949171840.0,
      "grad_norm": 2.4108681728164245,
      "language_loss": 0.77835119,
      "learning_rate": 3.660934950367916e-06,
      "loss": 0.80048227,
      "num_input_tokens_seen": 37558300,
      "step": 1767,
      "time_per_iteration": 2.714970588684082
    },
    {
      "auxiliary_loss_clip": 0.01199211,
      "auxiliary_loss_mlp": 0.01034358,
      "balance_loss_clip": 1.05767524,
      "balance_loss_mlp": 1.02420735,
      "epoch": 0.21258943064991281,
      "flos": 22382295402240.0,
      "grad_norm": 2.6204232658594635,
      "language_loss": 0.83423209,
      "learning_rate": 3.660500884907968e-06,
      "loss": 0.85656774,
      "num_input_tokens_seen": 37579040,
      "step": 1768,
      "time_per_iteration": 2.6279985904693604
    },
    {
      "auxiliary_loss_clip": 0.01065435,
      "auxiliary_loss_mlp": 0.01011638,
      "balance_loss_clip": 1.02263594,
      "balance_loss_mlp": 1.00930142,
      "epoch": 0.21270967354055192,
      "flos": 59440168679040.0,
      "grad_norm": 0.8262394958997549,
      "language_loss": 0.60001528,
      "learning_rate": 3.660066567551356e-06,
      "loss": 0.62078607,
      "num_input_tokens_seen": 37639185,
      "step": 1769,
      "time_per_iteration": 3.184523344039917
    },
    {
      "auxiliary_loss_clip": 0.0119885,
      "auxiliary_loss_mlp": 0.0076563,
      "balance_loss_clip": 1.05792427,
      "balance_loss_mlp": 1.00061297,
      "epoch": 0.212829916431191,
      "flos": 21544032729600.0,
      "grad_norm": 4.115122315992354,
      "language_loss": 0.84261167,
      "learning_rate": 3.6596319983639657e-06,
      "loss": 0.86225641,
      "num_input_tokens_seen": 37657765,
      "step": 1770,
      "time_per_iteration": 2.737048625946045
    },
    {
      "auxiliary_loss_clip": 0.01171767,
      "auxiliary_loss_mlp": 0.00765331,
      "balance_loss_clip": 1.05763376,
      "balance_loss_mlp": 1.00067925,
      "epoch": 0.2129501593218301,
      "flos": 28987739896320.0,
      "grad_norm": 1.5651411987687167,
      "language_loss": 0.86200804,
      "learning_rate": 3.6591971774117214e-06,
      "loss": 0.88137901,
      "num_input_tokens_seen": 37680740,
      "step": 1771,
      "time_per_iteration": 2.820146322250366
    },
    {
      "auxiliary_loss_clip": 0.01205188,
      "auxiliary_loss_mlp": 0.01039564,
      "balance_loss_clip": 1.06016827,
      "balance_loss_mlp": 1.02938986,
      "epoch": 0.2130704022124692,
      "flos": 18806993308800.0,
      "grad_norm": 2.2522321738925877,
      "language_loss": 0.80182433,
      "learning_rate": 3.6587621047605833e-06,
      "loss": 0.82427192,
      "num_input_tokens_seen": 37697910,
      "step": 1772,
      "time_per_iteration": 2.6427388191223145
    },
    {
      "auxiliary_loss_clip": 0.01200058,
      "auxiliary_loss_mlp": 0.01038599,
      "balance_loss_clip": 1.05875921,
      "balance_loss_mlp": 1.0283891,
      "epoch": 0.21319064510310828,
      "flos": 13918150759680.0,
      "grad_norm": 2.160865934651858,
      "language_loss": 0.86939013,
      "learning_rate": 3.6583267804765542e-06,
      "loss": 0.89177668,
      "num_input_tokens_seen": 37712245,
      "step": 1773,
      "time_per_iteration": 2.627722978591919
    },
    {
      "auxiliary_loss_clip": 0.01200507,
      "auxiliary_loss_mlp": 0.01037061,
      "balance_loss_clip": 1.05978048,
      "balance_loss_mlp": 1.02560461,
      "epoch": 0.21331088799374737,
      "flos": 20959694277120.0,
      "grad_norm": 2.4920913340649045,
      "language_loss": 0.85753095,
      "learning_rate": 3.6578912046256702e-06,
      "loss": 0.87990665,
      "num_input_tokens_seen": 37730765,
      "step": 1774,
      "time_per_iteration": 2.6075780391693115
    },
    {
      "auxiliary_loss_clip": 0.01163427,
      "auxiliary_loss_mlp": 0.01040636,
      "balance_loss_clip": 1.05191159,
      "balance_loss_mlp": 1.02983594,
      "epoch": 0.21343113088438645,
      "flos": 18624638937600.0,
      "grad_norm": 2.828633540206646,
      "language_loss": 0.76166302,
      "learning_rate": 3.6574553772740083e-06,
      "loss": 0.78370363,
      "num_input_tokens_seen": 37748695,
      "step": 1775,
      "time_per_iteration": 2.780250072479248
    },
    {
      "auxiliary_loss_clip": 0.01126524,
      "auxiliary_loss_mlp": 0.01004296,
      "balance_loss_clip": 1.05086899,
      "balance_loss_mlp": 1.00160217,
      "epoch": 0.21355137377502556,
      "flos": 67413128791680.0,
      "grad_norm": 0.8818462000973489,
      "language_loss": 0.6185379,
      "learning_rate": 3.657019298487684e-06,
      "loss": 0.63984609,
      "num_input_tokens_seen": 37813705,
      "step": 1776,
      "time_per_iteration": 3.2535343170166016
    },
    {
      "auxiliary_loss_clip": 0.01205738,
      "auxiliary_loss_mlp": 0.00764951,
      "balance_loss_clip": 1.05784142,
      "balance_loss_mlp": 1.00067043,
      "epoch": 0.21367161666566464,
      "flos": 34532095697280.0,
      "grad_norm": 1.9899987500529634,
      "language_loss": 0.83646441,
      "learning_rate": 3.6565829683328495e-06,
      "loss": 0.85617125,
      "num_input_tokens_seen": 37836330,
      "step": 1777,
      "time_per_iteration": 2.736564874649048
    },
    {
      "auxiliary_loss_clip": 0.01198493,
      "auxiliary_loss_mlp": 0.01045615,
      "balance_loss_clip": 1.05993319,
      "balance_loss_mlp": 1.03530324,
      "epoch": 0.21379185955630373,
      "flos": 18989347680000.0,
      "grad_norm": 1.9991778811840666,
      "language_loss": 0.86167383,
      "learning_rate": 3.6561463868756965e-06,
      "loss": 0.88411492,
      "num_input_tokens_seen": 37855030,
      "step": 1778,
      "time_per_iteration": 2.649639129638672
    },
    {
      "auxiliary_loss_clip": 0.01201537,
      "auxiliary_loss_mlp": 0.01034728,
      "balance_loss_clip": 1.06206787,
      "balance_loss_mlp": 1.02404094,
      "epoch": 0.21391210244694284,
      "flos": 28218497207040.0,
      "grad_norm": 1.5542390934085208,
      "language_loss": 0.77976918,
      "learning_rate": 3.655709554182452e-06,
      "loss": 0.80213183,
      "num_input_tokens_seen": 37875370,
      "step": 1779,
      "time_per_iteration": 2.692957639694214
    },
    {
      "auxiliary_loss_clip": 0.01205198,
      "auxiliary_loss_mlp": 0.01036121,
      "balance_loss_clip": 1.0592016,
      "balance_loss_mlp": 1.02571368,
      "epoch": 0.21403234533758192,
      "flos": 17455064192640.0,
      "grad_norm": 1.8699029661439395,
      "language_loss": 0.84632224,
      "learning_rate": 3.6552724703193855e-06,
      "loss": 0.86873543,
      "num_input_tokens_seen": 37892560,
      "step": 1780,
      "time_per_iteration": 2.6412971019744873
    },
    {
      "auxiliary_loss_clip": 0.01065691,
      "auxiliary_loss_mlp": 0.01006816,
      "balance_loss_clip": 1.02148199,
      "balance_loss_mlp": 1.00420499,
      "epoch": 0.214152588228221,
      "flos": 51637606686720.0,
      "grad_norm": 0.7951587210304365,
      "language_loss": 0.55895865,
      "learning_rate": 3.654835135352801e-06,
      "loss": 0.57968372,
      "num_input_tokens_seen": 37947370,
      "step": 1781,
      "time_per_iteration": 3.189809799194336
    },
    {
      "auxiliary_loss_clip": 0.01155183,
      "auxiliary_loss_mlp": 0.01034452,
      "balance_loss_clip": 1.04953837,
      "balance_loss_mlp": 1.02385986,
      "epoch": 0.21427283111886009,
      "flos": 19496154625920.0,
      "grad_norm": 1.994049338517953,
      "language_loss": 0.87541991,
      "learning_rate": 3.654397549349043e-06,
      "loss": 0.89731628,
      "num_input_tokens_seen": 37964745,
      "step": 1782,
      "time_per_iteration": 2.7551605701446533
    },
    {
      "auxiliary_loss_clip": 0.01182525,
      "auxiliary_loss_mlp": 0.01043308,
      "balance_loss_clip": 1.05849504,
      "balance_loss_mlp": 1.03213215,
      "epoch": 0.2143930740094992,
      "flos": 20084802710400.0,
      "grad_norm": 2.423530107588807,
      "language_loss": 0.75310594,
      "learning_rate": 3.653959712374491e-06,
      "loss": 0.77536428,
      "num_input_tokens_seen": 37982850,
      "step": 1783,
      "time_per_iteration": 3.6156795024871826
    },
    {
      "auxiliary_loss_clip": 0.01168226,
      "auxiliary_loss_mlp": 0.01039742,
      "balance_loss_clip": 1.05939507,
      "balance_loss_mlp": 1.0295918,
      "epoch": 0.21451331690013828,
      "flos": 21798603394560.0,
      "grad_norm": 1.64401731165414,
      "language_loss": 0.82592952,
      "learning_rate": 3.6535216244955663e-06,
      "loss": 0.84800923,
      "num_input_tokens_seen": 38002745,
      "step": 1784,
      "time_per_iteration": 2.692180633544922
    },
    {
      "auxiliary_loss_clip": 0.01185031,
      "auxiliary_loss_mlp": 0.01035144,
      "balance_loss_clip": 1.05687666,
      "balance_loss_mlp": 1.02440894,
      "epoch": 0.21463355979077736,
      "flos": 32853882412800.0,
      "grad_norm": 7.034347853325655,
      "language_loss": 0.71055663,
      "learning_rate": 3.653083285778726e-06,
      "loss": 0.7327584,
      "num_input_tokens_seen": 38024115,
      "step": 1785,
      "time_per_iteration": 2.826132297515869
    },
    {
      "auxiliary_loss_clip": 0.01202823,
      "auxiliary_loss_mlp": 0.01035837,
      "balance_loss_clip": 1.05730963,
      "balance_loss_mlp": 1.02522182,
      "epoch": 0.21475380268141647,
      "flos": 21543817248000.0,
      "grad_norm": 2.501163561153789,
      "language_loss": 0.81216162,
      "learning_rate": 3.6526446962904653e-06,
      "loss": 0.83454823,
      "num_input_tokens_seen": 38042830,
      "step": 1786,
      "time_per_iteration": 3.622807502746582
    },
    {
      "auxiliary_loss_clip": 0.01198403,
      "auxiliary_loss_mlp": 0.01033478,
      "balance_loss_clip": 1.06112254,
      "balance_loss_mlp": 1.02324426,
      "epoch": 0.21487404557205556,
      "flos": 32159082660480.0,
      "grad_norm": 1.7009124734797032,
      "language_loss": 0.74526447,
      "learning_rate": 3.652205856097318e-06,
      "loss": 0.76758331,
      "num_input_tokens_seen": 38066015,
      "step": 1787,
      "time_per_iteration": 3.643477201461792
    },
    {
      "auxiliary_loss_clip": 0.01182967,
      "auxiliary_loss_mlp": 0.00765432,
      "balance_loss_clip": 1.05683017,
      "balance_loss_mlp": 1.00067806,
      "epoch": 0.21499428846269464,
      "flos": 12673091583360.0,
      "grad_norm": 1.9815090072320514,
      "language_loss": 0.79203022,
      "learning_rate": 3.651766765265856e-06,
      "loss": 0.81151414,
      "num_input_tokens_seen": 38083025,
      "step": 1788,
      "time_per_iteration": 3.6836748123168945
    },
    {
      "auxiliary_loss_clip": 0.0118162,
      "auxiliary_loss_mlp": 0.01037419,
      "balance_loss_clip": 1.05590022,
      "balance_loss_mlp": 1.02655959,
      "epoch": 0.21511453135333372,
      "flos": 23471573293440.0,
      "grad_norm": 2.260235776268623,
      "language_loss": 0.81325787,
      "learning_rate": 3.65132742386269e-06,
      "loss": 0.83544827,
      "num_input_tokens_seen": 38098245,
      "step": 1789,
      "time_per_iteration": 2.6759603023529053
    },
    {
      "auxiliary_loss_clip": 0.01217335,
      "auxiliary_loss_mlp": 0.01036549,
      "balance_loss_clip": 1.06120515,
      "balance_loss_mlp": 1.02639842,
      "epoch": 0.21523477424397283,
      "flos": 26943560893440.0,
      "grad_norm": 1.7301039237285534,
      "language_loss": 0.8429172,
      "learning_rate": 3.6508878319544656e-06,
      "loss": 0.8654561,
      "num_input_tokens_seen": 38118460,
      "step": 1790,
      "time_per_iteration": 2.6598570346832275
    },
    {
      "auxiliary_loss_clip": 0.01182792,
      "auxiliary_loss_mlp": 0.01034613,
      "balance_loss_clip": 1.0588696,
      "balance_loss_mlp": 1.02382469,
      "epoch": 0.21535501713461191,
      "flos": 18916161719040.0,
      "grad_norm": 3.633145880823816,
      "language_loss": 0.82070619,
      "learning_rate": 3.65044798960787e-06,
      "loss": 0.84288025,
      "num_input_tokens_seen": 38136800,
      "step": 1791,
      "time_per_iteration": 2.723137617111206
    },
    {
      "auxiliary_loss_clip": 0.01160693,
      "auxiliary_loss_mlp": 0.01037635,
      "balance_loss_clip": 1.05145836,
      "balance_loss_mlp": 1.02770519,
      "epoch": 0.215475260025251,
      "flos": 17895113712000.0,
      "grad_norm": 2.441683093458962,
      "language_loss": 0.78109038,
      "learning_rate": 3.650007896889627e-06,
      "loss": 0.80307364,
      "num_input_tokens_seen": 38155380,
      "step": 1792,
      "time_per_iteration": 2.72836971282959
    },
    {
      "auxiliary_loss_clip": 0.01213585,
      "auxiliary_loss_mlp": 0.01036118,
      "balance_loss_clip": 1.06141329,
      "balance_loss_mlp": 1.02560985,
      "epoch": 0.2155955029158901,
      "flos": 16654292340480.0,
      "grad_norm": 1.800681664534201,
      "language_loss": 0.8063162,
      "learning_rate": 3.6495675538664974e-06,
      "loss": 0.8288132,
      "num_input_tokens_seen": 38174395,
      "step": 1793,
      "time_per_iteration": 2.6715281009674072
    },
    {
      "auxiliary_loss_clip": 0.01184846,
      "auxiliary_loss_mlp": 0.01034792,
      "balance_loss_clip": 1.0541935,
      "balance_loss_mlp": 1.0249567,
      "epoch": 0.2157157458065292,
      "flos": 23621213352960.0,
      "grad_norm": 2.6696899005338732,
      "language_loss": 0.82647794,
      "learning_rate": 3.649126960605282e-06,
      "loss": 0.8486743,
      "num_input_tokens_seen": 38195380,
      "step": 1794,
      "time_per_iteration": 2.708662748336792
    },
    {
      "auxiliary_loss_clip": 0.01184817,
      "auxiliary_loss_mlp": 0.01036875,
      "balance_loss_clip": 1.05817842,
      "balance_loss_mlp": 1.02605665,
      "epoch": 0.21583598869716827,
      "flos": 22127078292480.0,
      "grad_norm": 3.750168991725045,
      "language_loss": 0.83363801,
      "learning_rate": 3.6486861171728174e-06,
      "loss": 0.85585493,
      "num_input_tokens_seen": 38213775,
      "step": 1795,
      "time_per_iteration": 2.8233165740966797
    },
    {
      "auxiliary_loss_clip": 0.01172224,
      "auxiliary_loss_mlp": 0.01032972,
      "balance_loss_clip": 1.05344927,
      "balance_loss_mlp": 1.02187908,
      "epoch": 0.21595623158780738,
      "flos": 23441229279360.0,
      "grad_norm": 2.7320547901854404,
      "language_loss": 0.78621161,
      "learning_rate": 3.6482450236359803e-06,
      "loss": 0.80826354,
      "num_input_tokens_seen": 38235630,
      "step": 1796,
      "time_per_iteration": 2.6944985389709473
    },
    {
      "auxiliary_loss_clip": 0.01197818,
      "auxiliary_loss_mlp": 0.01030602,
      "balance_loss_clip": 1.06018043,
      "balance_loss_mlp": 1.02004659,
      "epoch": 0.21607647447844647,
      "flos": 26906501036160.0,
      "grad_norm": 2.5024859229128062,
      "language_loss": 0.78098518,
      "learning_rate": 3.647803680061683e-06,
      "loss": 0.80326939,
      "num_input_tokens_seen": 38256045,
      "step": 1797,
      "time_per_iteration": 2.7337169647216797
    },
    {
      "auxiliary_loss_clip": 0.0118774,
      "auxiliary_loss_mlp": 0.01033692,
      "balance_loss_clip": 1.05802274,
      "balance_loss_mlp": 1.02299309,
      "epoch": 0.21619671736908555,
      "flos": 14495378319360.0,
      "grad_norm": 2.558051470818116,
      "language_loss": 0.74255008,
      "learning_rate": 3.6473620865168776e-06,
      "loss": 0.76476437,
      "num_input_tokens_seen": 38272915,
      "step": 1798,
      "time_per_iteration": 2.627615213394165
    },
    {
      "auxiliary_loss_clip": 0.01183354,
      "auxiliary_loss_mlp": 0.01031239,
      "balance_loss_clip": 1.05707574,
      "balance_loss_mlp": 1.02084398,
      "epoch": 0.21631696025972463,
      "flos": 17931096161280.0,
      "grad_norm": 2.2240906913083527,
      "language_loss": 0.81940365,
      "learning_rate": 3.646920243068554e-06,
      "loss": 0.84154958,
      "num_input_tokens_seen": 38290810,
      "step": 1799,
      "time_per_iteration": 2.692553758621216
    },
    {
      "auxiliary_loss_clip": 0.01176273,
      "auxiliary_loss_mlp": 0.01042717,
      "balance_loss_clip": 1.05753756,
      "balance_loss_mlp": 1.03141618,
      "epoch": 0.21643720315036374,
      "flos": 24462385027200.0,
      "grad_norm": 2.0922270667759757,
      "language_loss": 0.74491096,
      "learning_rate": 3.6464781497837384e-06,
      "loss": 0.76710081,
      "num_input_tokens_seen": 38312785,
      "step": 1800,
      "time_per_iteration": 2.691991090774536
    },
    {
      "auxiliary_loss_clip": 0.01187807,
      "auxiliary_loss_mlp": 0.01040894,
      "balance_loss_clip": 1.05440569,
      "balance_loss_mlp": 1.03027844,
      "epoch": 0.21655744604100283,
      "flos": 28474432588800.0,
      "grad_norm": 3.022631748535448,
      "language_loss": 0.73073852,
      "learning_rate": 3.6460358067294965e-06,
      "loss": 0.75302553,
      "num_input_tokens_seen": 38334015,
      "step": 1801,
      "time_per_iteration": 2.7114710807800293
    },
    {
      "auxiliary_loss_clip": 0.01218217,
      "auxiliary_loss_mlp": 0.0103646,
      "balance_loss_clip": 1.05979252,
      "balance_loss_mlp": 1.02537394,
      "epoch": 0.2166776889316419,
      "flos": 20152960767360.0,
      "grad_norm": 2.2965163126684023,
      "language_loss": 0.78137761,
      "learning_rate": 3.645593213972932e-06,
      "loss": 0.80392432,
      "num_input_tokens_seen": 38352920,
      "step": 1802,
      "time_per_iteration": 2.603483200073242
    },
    {
      "auxiliary_loss_clip": 0.0119552,
      "auxiliary_loss_mlp": 0.01037419,
      "balance_loss_clip": 1.0588398,
      "balance_loss_mlp": 1.02690482,
      "epoch": 0.21679793182228102,
      "flos": 15193482122880.0,
      "grad_norm": 2.484984148965709,
      "language_loss": 0.80256057,
      "learning_rate": 3.6451503715811852e-06,
      "loss": 0.82489002,
      "num_input_tokens_seen": 38371230,
      "step": 1803,
      "time_per_iteration": 2.6477551460266113
    },
    {
      "auxiliary_loss_clip": 0.01188069,
      "auxiliary_loss_mlp": 0.01037016,
      "balance_loss_clip": 1.06076527,
      "balance_loss_mlp": 1.02724123,
      "epoch": 0.2169181747129201,
      "flos": 17384464010880.0,
      "grad_norm": 2.2386802417569163,
      "language_loss": 0.80138993,
      "learning_rate": 3.6447072796214345e-06,
      "loss": 0.82364082,
      "num_input_tokens_seen": 38389795,
      "step": 1804,
      "time_per_iteration": 2.671994209289551
    },
    {
      "auxiliary_loss_clip": 0.01064049,
      "auxiliary_loss_mlp": 0.01019514,
      "balance_loss_clip": 1.02499199,
      "balance_loss_mlp": 1.01699913,
      "epoch": 0.21703841760355919,
      "flos": 58760955429120.0,
      "grad_norm": 1.0807393904872657,
      "language_loss": 0.63187885,
      "learning_rate": 3.644263938160898e-06,
      "loss": 0.65271449,
      "num_input_tokens_seen": 38445760,
      "step": 1805,
      "time_per_iteration": 3.1721842288970947
    },
    {
      "auxiliary_loss_clip": 0.01167083,
      "auxiliary_loss_mlp": 0.01036652,
      "balance_loss_clip": 1.05634451,
      "balance_loss_mlp": 1.0255897,
      "epoch": 0.21715866049419827,
      "flos": 22418457419520.0,
      "grad_norm": 1.9907323830796033,
      "language_loss": 0.7196517,
      "learning_rate": 3.6438203472668293e-06,
      "loss": 0.74168909,
      "num_input_tokens_seen": 38465405,
      "step": 1806,
      "time_per_iteration": 2.7546093463897705
    },
    {
      "auxiliary_loss_clip": 0.01190949,
      "auxiliary_loss_mlp": 0.01032793,
      "balance_loss_clip": 1.06079054,
      "balance_loss_mlp": 1.0232507,
      "epoch": 0.21727890338483738,
      "flos": 17237732952960.0,
      "grad_norm": 2.146204592793586,
      "language_loss": 0.81514049,
      "learning_rate": 3.6433765070065206e-06,
      "loss": 0.83737791,
      "num_input_tokens_seen": 38483195,
      "step": 1807,
      "time_per_iteration": 2.6538589000701904
    },
    {
      "auxiliary_loss_clip": 0.0121787,
      "auxiliary_loss_mlp": 0.01034274,
      "balance_loss_clip": 1.06219339,
      "balance_loss_mlp": 1.02356303,
      "epoch": 0.21739914627547646,
      "flos": 13434792416640.0,
      "grad_norm": 4.626795731050862,
      "language_loss": 0.87786114,
      "learning_rate": 3.6429324174473025e-06,
      "loss": 0.90038252,
      "num_input_tokens_seen": 38496735,
      "step": 1808,
      "time_per_iteration": 2.6576898097991943
    },
    {
      "auxiliary_loss_clip": 0.01201369,
      "auxiliary_loss_mlp": 0.01038042,
      "balance_loss_clip": 1.05789018,
      "balance_loss_mlp": 1.02799869,
      "epoch": 0.21751938916611555,
      "flos": 20959514709120.0,
      "grad_norm": 2.5281077565170125,
      "language_loss": 0.8477757,
      "learning_rate": 3.6424880786565425e-06,
      "loss": 0.87016976,
      "num_input_tokens_seen": 38512880,
      "step": 1809,
      "time_per_iteration": 3.558098554611206
    },
    {
      "auxiliary_loss_clip": 0.01155044,
      "auxiliary_loss_mlp": 0.01035655,
      "balance_loss_clip": 1.05775809,
      "balance_loss_mlp": 1.02371073,
      "epoch": 0.21763963205675466,
      "flos": 27599936071680.0,
      "grad_norm": 2.1673442240153644,
      "language_loss": 0.79798645,
      "learning_rate": 3.6420434907016482e-06,
      "loss": 0.81989342,
      "num_input_tokens_seen": 38532570,
      "step": 1810,
      "time_per_iteration": 2.8068535327911377
    },
    {
      "auxiliary_loss_clip": 0.01203231,
      "auxiliary_loss_mlp": 0.0103937,
      "balance_loss_clip": 1.0630368,
      "balance_loss_mlp": 1.02945769,
      "epoch": 0.21775987494739374,
      "flos": 21430411032960.0,
      "grad_norm": 4.843516819571492,
      "language_loss": 0.81219363,
      "learning_rate": 3.6415986536500606e-06,
      "loss": 0.83461964,
      "num_input_tokens_seen": 38550900,
      "step": 1811,
      "time_per_iteration": 2.800466775894165
    },
    {
      "auxiliary_loss_clip": 0.01153202,
      "auxiliary_loss_mlp": 0.01035284,
      "balance_loss_clip": 1.0614264,
      "balance_loss_mlp": 1.02558589,
      "epoch": 0.21788011783803282,
      "flos": 18332972501760.0,
      "grad_norm": 1.7362865144594841,
      "language_loss": 0.8086834,
      "learning_rate": 3.641153567569263e-06,
      "loss": 0.83056831,
      "num_input_tokens_seen": 38569215,
      "step": 1812,
      "time_per_iteration": 3.638063430786133
    },
    {
      "auxiliary_loss_clip": 0.01196622,
      "auxiliary_loss_mlp": 0.01036012,
      "balance_loss_clip": 1.05865788,
      "balance_loss_mlp": 1.0262723,
      "epoch": 0.2180003607286719,
      "flos": 30262748037120.0,
      "grad_norm": 1.9978458423470355,
      "language_loss": 0.9555766,
      "learning_rate": 3.640708232526774e-06,
      "loss": 0.97790295,
      "num_input_tokens_seen": 38587870,
      "step": 1813,
      "time_per_iteration": 3.5974247455596924
    },
    {
      "auxiliary_loss_clip": 0.01139455,
      "auxiliary_loss_mlp": 0.01037194,
      "balance_loss_clip": 1.04905963,
      "balance_loss_mlp": 1.02642906,
      "epoch": 0.21812060361931102,
      "flos": 25480272637440.0,
      "grad_norm": 1.8828328384306876,
      "language_loss": 0.78319716,
      "learning_rate": 3.6402626485901504e-06,
      "loss": 0.80496359,
      "num_input_tokens_seen": 38606965,
      "step": 1814,
      "time_per_iteration": 3.70356822013855
    },
    {
      "auxiliary_loss_clip": 0.01202218,
      "auxiliary_loss_mlp": 0.01035535,
      "balance_loss_clip": 1.06582224,
      "balance_loss_mlp": 1.02595043,
      "epoch": 0.2182408465099501,
      "flos": 21908166854400.0,
      "grad_norm": 2.6568164766562585,
      "language_loss": 0.78020823,
      "learning_rate": 3.639816815826988e-06,
      "loss": 0.80258572,
      "num_input_tokens_seen": 38626290,
      "step": 1815,
      "time_per_iteration": 2.6984000205993652
    },
    {
      "auxiliary_loss_clip": 0.01184383,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 1.05909717,
      "balance_loss_mlp": 1.02156734,
      "epoch": 0.21836108940058918,
      "flos": 23657339456640.0,
      "grad_norm": 2.4899170986942156,
      "language_loss": 0.78029096,
      "learning_rate": 3.6393707343049176e-06,
      "loss": 0.80244708,
      "num_input_tokens_seen": 38646620,
      "step": 1816,
      "time_per_iteration": 2.7015702724456787
    },
    {
      "auxiliary_loss_clip": 0.0120664,
      "auxiliary_loss_mlp": 0.01041559,
      "balance_loss_clip": 1.06096768,
      "balance_loss_mlp": 1.03080606,
      "epoch": 0.2184813322912283,
      "flos": 24681009156480.0,
      "grad_norm": 4.017228392060421,
      "language_loss": 0.73605734,
      "learning_rate": 3.6389244040916104e-06,
      "loss": 0.75853932,
      "num_input_tokens_seen": 38665695,
      "step": 1817,
      "time_per_iteration": 2.6307883262634277
    },
    {
      "auxiliary_loss_clip": 0.01181048,
      "auxiliary_loss_mlp": 0.00765581,
      "balance_loss_clip": 1.06063044,
      "balance_loss_mlp": 1.00070786,
      "epoch": 0.21860157518186737,
      "flos": 26574650259840.0,
      "grad_norm": 3.114167499833823,
      "language_loss": 0.79104865,
      "learning_rate": 3.6384778252547747e-06,
      "loss": 0.81051493,
      "num_input_tokens_seen": 38681575,
      "step": 1818,
      "time_per_iteration": 2.775829315185547
    },
    {
      "auxiliary_loss_clip": 0.01189007,
      "auxiliary_loss_mlp": 0.00765429,
      "balance_loss_clip": 1.06454086,
      "balance_loss_mlp": 1.00079155,
      "epoch": 0.21872181807250646,
      "flos": 20886292834560.0,
      "grad_norm": 3.6375560423266053,
      "language_loss": 0.78562129,
      "learning_rate": 3.638030997862155e-06,
      "loss": 0.80516565,
      "num_input_tokens_seen": 38700510,
      "step": 1819,
      "time_per_iteration": 2.690314292907715
    },
    {
      "auxiliary_loss_clip": 0.01085016,
      "auxiliary_loss_mlp": 0.01004354,
      "balance_loss_clip": 1.02834165,
      "balance_loss_mlp": 1.00193429,
      "epoch": 0.21884206096314554,
      "flos": 61209452897280.0,
      "grad_norm": 0.7593087947221429,
      "language_loss": 0.59430015,
      "learning_rate": 3.6375839219815356e-06,
      "loss": 0.61519384,
      "num_input_tokens_seen": 38758310,
      "step": 1820,
      "time_per_iteration": 3.1798441410064697
    },
    {
      "auxiliary_loss_clip": 0.01220578,
      "auxiliary_loss_mlp": 0.01041838,
      "balance_loss_clip": 1.06383872,
      "balance_loss_mlp": 1.031389,
      "epoch": 0.21896230385378465,
      "flos": 23473835850240.0,
      "grad_norm": 2.1259965951760345,
      "language_loss": 0.82796824,
      "learning_rate": 3.6371365976807375e-06,
      "loss": 0.85059237,
      "num_input_tokens_seen": 38778705,
      "step": 1821,
      "time_per_iteration": 2.6662960052490234
    },
    {
      "auxiliary_loss_clip": 0.01149639,
      "auxiliary_loss_mlp": 0.0103986,
      "balance_loss_clip": 1.05478394,
      "balance_loss_mlp": 1.02978063,
      "epoch": 0.21908254674442373,
      "flos": 25081915829760.0,
      "grad_norm": 1.8358495279171585,
      "language_loss": 0.83423477,
      "learning_rate": 3.6366890250276185e-06,
      "loss": 0.85612977,
      "num_input_tokens_seen": 38799660,
      "step": 1822,
      "time_per_iteration": 2.80729341506958
    },
    {
      "auxiliary_loss_clip": 0.01218015,
      "auxiliary_loss_mlp": 0.01034632,
      "balance_loss_clip": 1.06483722,
      "balance_loss_mlp": 1.02491689,
      "epoch": 0.21920278963506282,
      "flos": 23513768795520.0,
      "grad_norm": 2.8727458937532027,
      "language_loss": 0.90055668,
      "learning_rate": 3.6362412040900764e-06,
      "loss": 0.92308313,
      "num_input_tokens_seen": 38819450,
      "step": 1823,
      "time_per_iteration": 2.6766951084136963
    },
    {
      "auxiliary_loss_clip": 0.01205425,
      "auxiliary_loss_mlp": 0.0103608,
      "balance_loss_clip": 1.06215954,
      "balance_loss_mlp": 1.025828,
      "epoch": 0.21932303252570193,
      "flos": 29242238734080.0,
      "grad_norm": 2.24888374340237,
      "language_loss": 0.80171275,
      "learning_rate": 3.635793134936044e-06,
      "loss": 0.82412779,
      "num_input_tokens_seen": 38840460,
      "step": 1824,
      "time_per_iteration": 2.71860671043396
    },
    {
      "auxiliary_loss_clip": 0.01204379,
      "auxiliary_loss_mlp": 0.01039996,
      "balance_loss_clip": 1.06342423,
      "balance_loss_mlp": 1.0302093,
      "epoch": 0.219443275416341,
      "flos": 20806857907200.0,
      "grad_norm": 1.9938930529238397,
      "language_loss": 0.73317885,
      "learning_rate": 3.635344817633494e-06,
      "loss": 0.75562263,
      "num_input_tokens_seen": 38859775,
      "step": 1825,
      "time_per_iteration": 2.67822265625
    },
    {
      "auxiliary_loss_clip": 0.01197397,
      "auxiliary_loss_mlp": 0.01031107,
      "balance_loss_clip": 1.06078017,
      "balance_loss_mlp": 1.02095604,
      "epoch": 0.2195635183069801,
      "flos": 14501555458560.0,
      "grad_norm": 2.8461982774270522,
      "language_loss": 0.75604427,
      "learning_rate": 3.634896252250436e-06,
      "loss": 0.77832925,
      "num_input_tokens_seen": 38876540,
      "step": 1826,
      "time_per_iteration": 2.6497950553894043
    },
    {
      "auxiliary_loss_clip": 0.01223444,
      "auxiliary_loss_mlp": 0.01038942,
      "balance_loss_clip": 1.06663823,
      "balance_loss_mlp": 1.02822495,
      "epoch": 0.2196837611976192,
      "flos": 24243473589120.0,
      "grad_norm": 2.058806167910548,
      "language_loss": 0.82046247,
      "learning_rate": 3.6344474388549157e-06,
      "loss": 0.84308636,
      "num_input_tokens_seen": 38896195,
      "step": 1827,
      "time_per_iteration": 2.672501564025879
    },
    {
      "auxiliary_loss_clip": 0.01207842,
      "auxiliary_loss_mlp": 0.01034433,
      "balance_loss_clip": 1.06610966,
      "balance_loss_mlp": 1.02369857,
      "epoch": 0.2198040040882583,
      "flos": 18074523168000.0,
      "grad_norm": 2.8822020888149535,
      "language_loss": 0.80662358,
      "learning_rate": 3.6339983775150183e-06,
      "loss": 0.82904631,
      "num_input_tokens_seen": 38912755,
      "step": 1828,
      "time_per_iteration": 2.6545498371124268
    },
    {
      "auxiliary_loss_clip": 0.01203008,
      "auxiliary_loss_mlp": 0.01034707,
      "balance_loss_clip": 1.06423676,
      "balance_loss_mlp": 1.02472353,
      "epoch": 0.21992424697889737,
      "flos": 17784185535360.0,
      "grad_norm": 2.665058451746,
      "language_loss": 0.84304738,
      "learning_rate": 3.6335490682988664e-06,
      "loss": 0.86542451,
      "num_input_tokens_seen": 38928365,
      "step": 1829,
      "time_per_iteration": 2.6310906410217285
    },
    {
      "auxiliary_loss_clip": 0.01133752,
      "auxiliary_loss_mlp": 0.01035736,
      "balance_loss_clip": 1.05035019,
      "balance_loss_mlp": 1.02500141,
      "epoch": 0.22004448986953645,
      "flos": 17638495971840.0,
      "grad_norm": 2.3781619191960974,
      "language_loss": 0.82762015,
      "learning_rate": 3.63309951127462e-06,
      "loss": 0.84931505,
      "num_input_tokens_seen": 38945275,
      "step": 1830,
      "time_per_iteration": 2.841913938522339
    },
    {
      "auxiliary_loss_clip": 0.01173402,
      "auxiliary_loss_mlp": 0.01040144,
      "balance_loss_clip": 1.05869031,
      "balance_loss_mlp": 1.02973652,
      "epoch": 0.22016473276017556,
      "flos": 22275533203200.0,
      "grad_norm": 2.963626019553092,
      "language_loss": 0.75262737,
      "learning_rate": 3.6326497065104757e-06,
      "loss": 0.77476287,
      "num_input_tokens_seen": 38965740,
      "step": 1831,
      "time_per_iteration": 2.75930118560791
    },
    {
      "auxiliary_loss_clip": 0.01205529,
      "auxiliary_loss_mlp": 0.0103985,
      "balance_loss_clip": 1.06212187,
      "balance_loss_mlp": 1.02921653,
      "epoch": 0.22028497565081465,
      "flos": 25556259859200.0,
      "grad_norm": 2.3870122101065574,
      "language_loss": 0.7829532,
      "learning_rate": 3.6321996540746697e-06,
      "loss": 0.80540693,
      "num_input_tokens_seen": 38984815,
      "step": 1832,
      "time_per_iteration": 2.737990617752075
    },
    {
      "auxiliary_loss_clip": 0.01169517,
      "auxiliary_loss_mlp": 0.01045849,
      "balance_loss_clip": 1.05675757,
      "balance_loss_mlp": 1.03504848,
      "epoch": 0.22040521854145373,
      "flos": 36247332925440.0,
      "grad_norm": 1.9085573774356985,
      "language_loss": 0.79970318,
      "learning_rate": 3.6317493540354733e-06,
      "loss": 0.82185686,
      "num_input_tokens_seen": 39008230,
      "step": 1833,
      "time_per_iteration": 2.8235697746276855
    },
    {
      "auxiliary_loss_clip": 0.01194353,
      "auxiliary_loss_mlp": 0.01040195,
      "balance_loss_clip": 1.05787289,
      "balance_loss_mlp": 1.02921546,
      "epoch": 0.22052546143209284,
      "flos": 11838420270720.0,
      "grad_norm": 2.0020641396983048,
      "language_loss": 0.7672444,
      "learning_rate": 3.6312988064611976e-06,
      "loss": 0.78958988,
      "num_input_tokens_seen": 39026540,
      "step": 1834,
      "time_per_iteration": 2.677777051925659
    },
    {
      "auxiliary_loss_clip": 0.011702,
      "auxiliary_loss_mlp": 0.0103056,
      "balance_loss_clip": 1.05270791,
      "balance_loss_mlp": 1.02067721,
      "epoch": 0.22064570432273192,
      "flos": 24209250906240.0,
      "grad_norm": 2.7130221828710943,
      "language_loss": 0.81256092,
      "learning_rate": 3.6308480114201896e-06,
      "loss": 0.8345685,
      "num_input_tokens_seen": 39048460,
      "step": 1835,
      "time_per_iteration": 3.659214496612549
    },
    {
      "auxiliary_loss_clip": 0.01219847,
      "auxiliary_loss_mlp": 0.01036724,
      "balance_loss_clip": 1.06618309,
      "balance_loss_mlp": 1.02655578,
      "epoch": 0.220765947213371,
      "flos": 17931347556480.0,
      "grad_norm": 1.9767578058702866,
      "language_loss": 0.76200628,
      "learning_rate": 3.630396968980835e-06,
      "loss": 0.78457201,
      "num_input_tokens_seen": 39066335,
      "step": 1836,
      "time_per_iteration": 2.6587564945220947
    },
    {
      "auxiliary_loss_clip": 0.0118689,
      "auxiliary_loss_mlp": 0.01031672,
      "balance_loss_clip": 1.05949306,
      "balance_loss_mlp": 1.02174211,
      "epoch": 0.2208861901040101,
      "flos": 26757040544640.0,
      "grad_norm": 2.8360199717239745,
      "language_loss": 0.83389807,
      "learning_rate": 3.6299456792115575e-06,
      "loss": 0.85608375,
      "num_input_tokens_seen": 39087590,
      "step": 1837,
      "time_per_iteration": 2.735236406326294
    },
    {
      "auxiliary_loss_clip": 0.01109269,
      "auxiliary_loss_mlp": 0.01040808,
      "balance_loss_clip": 1.0451498,
      "balance_loss_mlp": 1.03068089,
      "epoch": 0.2210064329946492,
      "flos": 17817977255040.0,
      "grad_norm": 2.1433344494949043,
      "language_loss": 0.80947232,
      "learning_rate": 3.629494142180815e-06,
      "loss": 0.83097315,
      "num_input_tokens_seen": 39106335,
      "step": 1838,
      "time_per_iteration": 3.750500202178955
    },
    {
      "auxiliary_loss_clip": 0.01218689,
      "auxiliary_loss_mlp": 0.01037488,
      "balance_loss_clip": 1.065696,
      "balance_loss_mlp": 1.02802825,
      "epoch": 0.22112667588528828,
      "flos": 17967401832960.0,
      "grad_norm": 2.5282357540140232,
      "language_loss": 0.85351014,
      "learning_rate": 3.6290423579571075e-06,
      "loss": 0.87607193,
      "num_input_tokens_seen": 39122875,
      "step": 1839,
      "time_per_iteration": 4.407745599746704
    },
    {
      "auxiliary_loss_clip": 0.01201212,
      "auxiliary_loss_mlp": 0.01034112,
      "balance_loss_clip": 1.06338716,
      "balance_loss_mlp": 1.02421212,
      "epoch": 0.22124691877592736,
      "flos": 18369206346240.0,
      "grad_norm": 1.740999549317465,
      "language_loss": 0.80091316,
      "learning_rate": 3.6285903266089694e-06,
      "loss": 0.82326639,
      "num_input_tokens_seen": 39142150,
      "step": 1840,
      "time_per_iteration": 2.6872785091400146
    },
    {
      "auxiliary_loss_clip": 0.0118739,
      "auxiliary_loss_mlp": 0.01038179,
      "balance_loss_clip": 1.05883408,
      "balance_loss_mlp": 1.02672875,
      "epoch": 0.22136716166656648,
      "flos": 20813286441600.0,
      "grad_norm": 2.024390396079131,
      "language_loss": 0.77091789,
      "learning_rate": 3.628138048204974e-06,
      "loss": 0.79317361,
      "num_input_tokens_seen": 39162835,
      "step": 1841,
      "time_per_iteration": 2.673769950866699
    },
    {
      "auxiliary_loss_clip": 0.01152607,
      "auxiliary_loss_mlp": 0.01036568,
      "balance_loss_clip": 1.05890179,
      "balance_loss_mlp": 1.02497458,
      "epoch": 0.22148740455720556,
      "flos": 17675699483520.0,
      "grad_norm": 2.694904066026041,
      "language_loss": 0.76247817,
      "learning_rate": 3.6276855228137304e-06,
      "loss": 0.78436995,
      "num_input_tokens_seen": 39181040,
      "step": 1842,
      "time_per_iteration": 2.707284927368164
    },
    {
      "auxiliary_loss_clip": 0.01217645,
      "auxiliary_loss_mlp": 0.00765144,
      "balance_loss_clip": 1.064026,
      "balance_loss_mlp": 1.0009166,
      "epoch": 0.22160764744784464,
      "flos": 21726710323200.0,
      "grad_norm": 2.2314190607423607,
      "language_loss": 0.81759357,
      "learning_rate": 3.6272327505038874e-06,
      "loss": 0.83742148,
      "num_input_tokens_seen": 39197505,
      "step": 1843,
      "time_per_iteration": 2.5897231101989746
    },
    {
      "auxiliary_loss_clip": 0.01165169,
      "auxiliary_loss_mlp": 0.01035488,
      "balance_loss_clip": 1.05719507,
      "balance_loss_mlp": 1.026124,
      "epoch": 0.22172789033848372,
      "flos": 23764712186880.0,
      "grad_norm": 2.447148086686586,
      "language_loss": 0.78499901,
      "learning_rate": 3.626779731344131e-06,
      "loss": 0.80700558,
      "num_input_tokens_seen": 39217295,
      "step": 1844,
      "time_per_iteration": 2.7728829383850098
    },
    {
      "auxiliary_loss_clip": 0.01216139,
      "auxiliary_loss_mlp": 0.01036084,
      "balance_loss_clip": 1.0642761,
      "balance_loss_mlp": 1.02646387,
      "epoch": 0.22184813322912283,
      "flos": 16982300361600.0,
      "grad_norm": 2.587831653155976,
      "language_loss": 0.85215747,
      "learning_rate": 3.6263264654031814e-06,
      "loss": 0.87467968,
      "num_input_tokens_seen": 39234195,
      "step": 1845,
      "time_per_iteration": 2.5795907974243164
    },
    {
      "auxiliary_loss_clip": 0.0107862,
      "auxiliary_loss_mlp": 0.01003736,
      "balance_loss_clip": 1.02934444,
      "balance_loss_mlp": 1.00135159,
      "epoch": 0.22196837611976192,
      "flos": 61823740314240.0,
      "grad_norm": 0.7006750785663416,
      "language_loss": 0.59167969,
      "learning_rate": 3.6258729527498008e-06,
      "loss": 0.61250329,
      "num_input_tokens_seen": 39295040,
      "step": 1846,
      "time_per_iteration": 3.2299644947052
    },
    {
      "auxiliary_loss_clip": 0.01194544,
      "auxiliary_loss_mlp": 0.01027727,
      "balance_loss_clip": 1.06169391,
      "balance_loss_mlp": 1.01777351,
      "epoch": 0.222088619010401,
      "flos": 25558019625600.0,
      "grad_norm": 5.008858775445267,
      "language_loss": 0.65485823,
      "learning_rate": 3.6254191934527854e-06,
      "loss": 0.67708093,
      "num_input_tokens_seen": 39314395,
      "step": 1847,
      "time_per_iteration": 2.78926157951355
    },
    {
      "auxiliary_loss_clip": 0.01173902,
      "auxiliary_loss_mlp": 0.01035288,
      "balance_loss_clip": 1.06272066,
      "balance_loss_mlp": 1.02503622,
      "epoch": 0.2222088619010401,
      "flos": 19318612677120.0,
      "grad_norm": 2.7523368927283065,
      "language_loss": 0.65042222,
      "learning_rate": 3.6249651875809715e-06,
      "loss": 0.67251408,
      "num_input_tokens_seen": 39334275,
      "step": 1848,
      "time_per_iteration": 2.6862504482269287
    },
    {
      "auxiliary_loss_clip": 0.011845,
      "auxiliary_loss_mlp": 0.01034707,
      "balance_loss_clip": 1.06268108,
      "balance_loss_mlp": 1.02468181,
      "epoch": 0.2223291047916792,
      "flos": 19099342103040.0,
      "grad_norm": 2.2794279333336545,
      "language_loss": 0.88872385,
      "learning_rate": 3.62451093520323e-06,
      "loss": 0.91091597,
      "num_input_tokens_seen": 39352180,
      "step": 1849,
      "time_per_iteration": 2.774221181869507
    },
    {
      "auxiliary_loss_clip": 0.0115026,
      "auxiliary_loss_mlp": 0.01030364,
      "balance_loss_clip": 1.05094564,
      "balance_loss_mlp": 1.02051711,
      "epoch": 0.22244934768231828,
      "flos": 20850418126080.0,
      "grad_norm": 7.686921950621746,
      "language_loss": 0.90544045,
      "learning_rate": 3.6240564363884714e-06,
      "loss": 0.92724669,
      "num_input_tokens_seen": 39372125,
      "step": 1850,
      "time_per_iteration": 2.8026583194732666
    },
    {
      "auxiliary_loss_clip": 0.01203429,
      "auxiliary_loss_mlp": 0.01034959,
      "balance_loss_clip": 1.05842233,
      "balance_loss_mlp": 1.02496958,
      "epoch": 0.2225695905729574,
      "flos": 15632921111040.0,
      "grad_norm": 1.9323874957189782,
      "language_loss": 0.70435619,
      "learning_rate": 3.623601691205643e-06,
      "loss": 0.72674012,
      "num_input_tokens_seen": 39391200,
      "step": 1851,
      "time_per_iteration": 2.735692024230957
    },
    {
      "auxiliary_loss_clip": 0.01196673,
      "auxiliary_loss_mlp": 0.01034722,
      "balance_loss_clip": 1.05701089,
      "balance_loss_mlp": 1.02477419,
      "epoch": 0.22268983346359647,
      "flos": 25373582265600.0,
      "grad_norm": 2.0804402693740895,
      "language_loss": 0.81471896,
      "learning_rate": 3.623146699723729e-06,
      "loss": 0.83703291,
      "num_input_tokens_seen": 39410660,
      "step": 1852,
      "time_per_iteration": 2.65704607963562
    },
    {
      "auxiliary_loss_clip": 0.01186343,
      "auxiliary_loss_mlp": 0.010314,
      "balance_loss_clip": 1.06373024,
      "balance_loss_mlp": 1.02103436,
      "epoch": 0.22281007635423555,
      "flos": 13261452359040.0,
      "grad_norm": 3.489718648786214,
      "language_loss": 0.77485317,
      "learning_rate": 3.6226914620117507e-06,
      "loss": 0.79703057,
      "num_input_tokens_seen": 39429280,
      "step": 1853,
      "time_per_iteration": 2.7184698581695557
    },
    {
      "auxiliary_loss_clip": 0.01169804,
      "auxiliary_loss_mlp": 0.01031101,
      "balance_loss_clip": 1.05184221,
      "balance_loss_mlp": 1.02156997,
      "epoch": 0.22293031924487464,
      "flos": 15340536403200.0,
      "grad_norm": 2.147676607591994,
      "language_loss": 0.80770695,
      "learning_rate": 3.622235978138768e-06,
      "loss": 0.82971597,
      "num_input_tokens_seen": 39446905,
      "step": 1854,
      "time_per_iteration": 2.706463098526001
    },
    {
      "auxiliary_loss_clip": 0.01200723,
      "auxiliary_loss_mlp": 0.01031829,
      "balance_loss_clip": 1.06280386,
      "balance_loss_mlp": 1.02174425,
      "epoch": 0.22305056213551375,
      "flos": 22564649773440.0,
      "grad_norm": 2.334516205402604,
      "language_loss": 0.8117429,
      "learning_rate": 3.621780248173877e-06,
      "loss": 0.83406842,
      "num_input_tokens_seen": 39465105,
      "step": 1855,
      "time_per_iteration": 2.677907705307007
    },
    {
      "auxiliary_loss_clip": 0.01104823,
      "auxiliary_loss_mlp": 0.01003545,
      "balance_loss_clip": 1.02771127,
      "balance_loss_mlp": 1.00123239,
      "epoch": 0.22317080502615283,
      "flos": 64880419887360.0,
      "grad_norm": 0.9163670325223952,
      "language_loss": 0.61016548,
      "learning_rate": 3.6213242721862125e-06,
      "loss": 0.63124919,
      "num_input_tokens_seen": 39523560,
      "step": 1856,
      "time_per_iteration": 3.2198057174682617
    },
    {
      "auxiliary_loss_clip": 0.01171627,
      "auxiliary_loss_mlp": 0.01028821,
      "balance_loss_clip": 1.05338693,
      "balance_loss_mlp": 1.01907015,
      "epoch": 0.2232910479167919,
      "flos": 25775997310080.0,
      "grad_norm": 1.8419840607744538,
      "language_loss": 0.75310349,
      "learning_rate": 3.620868050244945e-06,
      "loss": 0.77510798,
      "num_input_tokens_seen": 39544040,
      "step": 1857,
      "time_per_iteration": 2.747532606124878
    },
    {
      "auxiliary_loss_clip": 0.01176845,
      "auxiliary_loss_mlp": 0.01035845,
      "balance_loss_clip": 1.0538528,
      "balance_loss_mlp": 1.02461529,
      "epoch": 0.22341129080743102,
      "flos": 23251799928960.0,
      "grad_norm": 2.4527914060757507,
      "language_loss": 0.77715278,
      "learning_rate": 3.6204115824192817e-06,
      "loss": 0.79927969,
      "num_input_tokens_seen": 39561515,
      "step": 1858,
      "time_per_iteration": 2.7165987491607666
    },
    {
      "auxiliary_loss_clip": 0.01172828,
      "auxiliary_loss_mlp": 0.01035266,
      "balance_loss_clip": 1.05156064,
      "balance_loss_mlp": 1.02524602,
      "epoch": 0.2235315336980701,
      "flos": 21214552250880.0,
      "grad_norm": 2.488365459165997,
      "language_loss": 0.76699555,
      "learning_rate": 3.619954868778471e-06,
      "loss": 0.78907645,
      "num_input_tokens_seen": 39578210,
      "step": 1859,
      "time_per_iteration": 2.6350719928741455
    },
    {
      "auxiliary_loss_clip": 0.0118452,
      "auxiliary_loss_mlp": 0.01029586,
      "balance_loss_clip": 1.056862,
      "balance_loss_mlp": 1.02050209,
      "epoch": 0.2236517765887092,
      "flos": 19901945548800.0,
      "grad_norm": 2.8355746386640215,
      "language_loss": 0.83080876,
      "learning_rate": 3.6194979093917944e-06,
      "loss": 0.8529498,
      "num_input_tokens_seen": 39597625,
      "step": 1860,
      "time_per_iteration": 2.7272799015045166
    },
    {
      "auxiliary_loss_clip": 0.01177612,
      "auxiliary_loss_mlp": 0.01040456,
      "balance_loss_clip": 1.05411828,
      "balance_loss_mlp": 1.03009701,
      "epoch": 0.22377201947934827,
      "flos": 23214847812480.0,
      "grad_norm": 2.389515494373004,
      "language_loss": 0.87036407,
      "learning_rate": 3.6190407043285724e-06,
      "loss": 0.89254475,
      "num_input_tokens_seen": 39615360,
      "step": 1861,
      "time_per_iteration": 3.5463955402374268
    },
    {
      "auxiliary_loss_clip": 0.01216468,
      "auxiliary_loss_mlp": 0.01035878,
      "balance_loss_clip": 1.06103301,
      "balance_loss_mlp": 1.02616835,
      "epoch": 0.22389226236998738,
      "flos": 26794244056320.0,
      "grad_norm": 2.2519289821310915,
      "language_loss": 0.75847542,
      "learning_rate": 3.618583253658163e-06,
      "loss": 0.78099883,
      "num_input_tokens_seen": 39635460,
      "step": 1862,
      "time_per_iteration": 2.7567789554595947
    },
    {
      "auxiliary_loss_clip": 0.01162558,
      "auxiliary_loss_mlp": 0.00765307,
      "balance_loss_clip": 1.06085968,
      "balance_loss_mlp": 1.00090528,
      "epoch": 0.22401250526062647,
      "flos": 24170359455360.0,
      "grad_norm": 2.460246262400244,
      "language_loss": 0.86613905,
      "learning_rate": 3.618125557449961e-06,
      "loss": 0.88541776,
      "num_input_tokens_seen": 39653515,
      "step": 1863,
      "time_per_iteration": 2.7878665924072266
    },
    {
      "auxiliary_loss_clip": 0.01193377,
      "auxiliary_loss_mlp": 0.01034204,
      "balance_loss_clip": 1.05683053,
      "balance_loss_mlp": 1.02443445,
      "epoch": 0.22413274815126555,
      "flos": 16759761649920.0,
      "grad_norm": 2.1245305673400137,
      "language_loss": 0.831541,
      "learning_rate": 3.6176676157733983e-06,
      "loss": 0.85381687,
      "num_input_tokens_seen": 39668525,
      "step": 1864,
      "time_per_iteration": 3.54984974861145
    },
    {
      "auxiliary_loss_clip": 0.01159462,
      "auxiliary_loss_mlp": 0.01033801,
      "balance_loss_clip": 1.05221152,
      "balance_loss_mlp": 1.02334666,
      "epoch": 0.22425299104190466,
      "flos": 21360205900800.0,
      "grad_norm": 2.315493971244824,
      "language_loss": 0.75841653,
      "learning_rate": 3.6172094286979443e-06,
      "loss": 0.78034914,
      "num_input_tokens_seen": 39685895,
      "step": 1865,
      "time_per_iteration": 3.613912343978882
    },
    {
      "auxiliary_loss_clip": 0.0118125,
      "auxiliary_loss_mlp": 0.01031841,
      "balance_loss_clip": 1.05418837,
      "balance_loss_mlp": 1.02232838,
      "epoch": 0.22437323393254374,
      "flos": 32165547108480.0,
      "grad_norm": 1.512771859955036,
      "language_loss": 0.81430918,
      "learning_rate": 3.6167509962931064e-06,
      "loss": 0.83644009,
      "num_input_tokens_seen": 39711595,
      "step": 1866,
      "time_per_iteration": 3.733426809310913
    },
    {
      "auxiliary_loss_clip": 0.01162787,
      "auxiliary_loss_mlp": 0.01041233,
      "balance_loss_clip": 1.05748868,
      "balance_loss_mlp": 1.03066528,
      "epoch": 0.22449347682318282,
      "flos": 18002809664640.0,
      "grad_norm": 2.4153824926256,
      "language_loss": 0.76907849,
      "learning_rate": 3.6162923186284276e-06,
      "loss": 0.79111874,
      "num_input_tokens_seen": 39727555,
      "step": 1867,
      "time_per_iteration": 2.7301242351531982
    },
    {
      "auxiliary_loss_clip": 0.01182094,
      "auxiliary_loss_mlp": 0.01038268,
      "balance_loss_clip": 1.05478024,
      "balance_loss_mlp": 1.02814758,
      "epoch": 0.2246137197138219,
      "flos": 18697286194560.0,
      "grad_norm": 3.3831917687446946,
      "language_loss": 0.86203659,
      "learning_rate": 3.6158333957734888e-06,
      "loss": 0.88424015,
      "num_input_tokens_seen": 39746145,
      "step": 1868,
      "time_per_iteration": 2.6613547801971436
    },
    {
      "auxiliary_loss_clip": 0.01174235,
      "auxiliary_loss_mlp": 0.0103871,
      "balance_loss_clip": 1.05351102,
      "balance_loss_mlp": 1.02916789,
      "epoch": 0.22473396260446102,
      "flos": 15590653781760.0,
      "grad_norm": 3.6077309753674895,
      "language_loss": 0.83134723,
      "learning_rate": 3.6153742277979088e-06,
      "loss": 0.85347664,
      "num_input_tokens_seen": 39763575,
      "step": 1869,
      "time_per_iteration": 2.6783289909362793
    },
    {
      "auxiliary_loss_clip": 0.01180976,
      "auxiliary_loss_mlp": 0.01036406,
      "balance_loss_clip": 1.05407047,
      "balance_loss_mlp": 1.02682745,
      "epoch": 0.2248542054951001,
      "flos": 14465501182080.0,
      "grad_norm": 2.5027507586476974,
      "language_loss": 0.77990782,
      "learning_rate": 3.6149148147713434e-06,
      "loss": 0.8020817,
      "num_input_tokens_seen": 39781810,
      "step": 1870,
      "time_per_iteration": 2.7274792194366455
    },
    {
      "auxiliary_loss_clip": 0.01207009,
      "auxiliary_loss_mlp": 0.01032616,
      "balance_loss_clip": 1.06420708,
      "balance_loss_mlp": 1.02324629,
      "epoch": 0.22497444838573918,
      "flos": 19243882431360.0,
      "grad_norm": 2.64242613976117,
      "language_loss": 0.86630714,
      "learning_rate": 3.614455156763484e-06,
      "loss": 0.88870347,
      "num_input_tokens_seen": 39800115,
      "step": 1871,
      "time_per_iteration": 2.6262431144714355
    },
    {
      "auxiliary_loss_clip": 0.0114948,
      "auxiliary_loss_mlp": 0.01040982,
      "balance_loss_clip": 1.04890788,
      "balance_loss_mlp": 1.03049207,
      "epoch": 0.2250946912763783,
      "flos": 16910299549440.0,
      "grad_norm": 3.4972664763473094,
      "language_loss": 0.71705008,
      "learning_rate": 3.613995253844061e-06,
      "loss": 0.73895472,
      "num_input_tokens_seen": 39817795,
      "step": 1872,
      "time_per_iteration": 2.781992197036743
    },
    {
      "auxiliary_loss_clip": 0.01195595,
      "auxiliary_loss_mlp": 0.01035194,
      "balance_loss_clip": 1.0580616,
      "balance_loss_mlp": 1.02476907,
      "epoch": 0.22521493416701738,
      "flos": 24681368292480.0,
      "grad_norm": 3.515640940144834,
      "language_loss": 0.80787235,
      "learning_rate": 3.6135351060828414e-06,
      "loss": 0.83018023,
      "num_input_tokens_seen": 39838270,
      "step": 1873,
      "time_per_iteration": 2.705841064453125
    },
    {
      "auxiliary_loss_clip": 0.01221324,
      "auxiliary_loss_mlp": 0.0104308,
      "balance_loss_clip": 1.06415665,
      "balance_loss_mlp": 1.03233957,
      "epoch": 0.22533517705765646,
      "flos": 17821963664640.0,
      "grad_norm": 3.1593450817507494,
      "language_loss": 0.69453454,
      "learning_rate": 3.6130747135496285e-06,
      "loss": 0.71717858,
      "num_input_tokens_seen": 39857270,
      "step": 1874,
      "time_per_iteration": 2.5820014476776123
    },
    {
      "auxiliary_loss_clip": 0.01210112,
      "auxiliary_loss_mlp": 0.01036649,
      "balance_loss_clip": 1.0584662,
      "balance_loss_mlp": 1.02617633,
      "epoch": 0.22545541994829554,
      "flos": 33691390899840.0,
      "grad_norm": 2.003505718888647,
      "language_loss": 0.65842372,
      "learning_rate": 3.6126140763142646e-06,
      "loss": 0.68089128,
      "num_input_tokens_seen": 39882300,
      "step": 1875,
      "time_per_iteration": 2.746408224105835
    },
    {
      "auxiliary_loss_clip": 0.0121467,
      "auxiliary_loss_mlp": 0.0103354,
      "balance_loss_clip": 1.06067753,
      "balance_loss_mlp": 1.02403331,
      "epoch": 0.22557566283893465,
      "flos": 19171594310400.0,
      "grad_norm": 2.5537282761514963,
      "language_loss": 0.85908711,
      "learning_rate": 3.6121531944466275e-06,
      "loss": 0.88156915,
      "num_input_tokens_seen": 39899625,
      "step": 1876,
      "time_per_iteration": 2.6696348190307617
    },
    {
      "auxiliary_loss_clip": 0.01197416,
      "auxiliary_loss_mlp": 0.01036412,
      "balance_loss_clip": 1.05891728,
      "balance_loss_mlp": 1.02608836,
      "epoch": 0.22569590572957374,
      "flos": 20773281669120.0,
      "grad_norm": 2.347460720481523,
      "language_loss": 0.78125638,
      "learning_rate": 3.611692068016633e-06,
      "loss": 0.80359465,
      "num_input_tokens_seen": 39915955,
      "step": 1877,
      "time_per_iteration": 2.658524513244629
    },
    {
      "auxiliary_loss_clip": 0.01163147,
      "auxiliary_loss_mlp": 0.01039386,
      "balance_loss_clip": 1.05112553,
      "balance_loss_mlp": 1.02875221,
      "epoch": 0.22581614862021282,
      "flos": 18442715529600.0,
      "grad_norm": 2.319259921439485,
      "language_loss": 0.75214529,
      "learning_rate": 3.611230697094233e-06,
      "loss": 0.77417064,
      "num_input_tokens_seen": 39932655,
      "step": 1878,
      "time_per_iteration": 2.679168701171875
    },
    {
      "auxiliary_loss_clip": 0.01186276,
      "auxiliary_loss_mlp": 0.01041113,
      "balance_loss_clip": 1.05641305,
      "balance_loss_mlp": 1.03212488,
      "epoch": 0.22593639151085193,
      "flos": 20048389297920.0,
      "grad_norm": 2.490542031087404,
      "language_loss": 0.87155211,
      "learning_rate": 3.6107690817494173e-06,
      "loss": 0.89382601,
      "num_input_tokens_seen": 39952875,
      "step": 1879,
      "time_per_iteration": 2.715020179748535
    },
    {
      "auxiliary_loss_clip": 0.01144344,
      "auxiliary_loss_mlp": 0.01037718,
      "balance_loss_clip": 1.04955077,
      "balance_loss_mlp": 1.02779937,
      "epoch": 0.226056634401491,
      "flos": 13115116350720.0,
      "grad_norm": 3.8495490775116017,
      "language_loss": 0.70705295,
      "learning_rate": 3.6103072220522117e-06,
      "loss": 0.72887361,
      "num_input_tokens_seen": 39968405,
      "step": 1880,
      "time_per_iteration": 2.6595053672790527
    },
    {
      "auxiliary_loss_clip": 0.01171926,
      "auxiliary_loss_mlp": 0.01033147,
      "balance_loss_clip": 1.05405843,
      "balance_loss_mlp": 1.0239203,
      "epoch": 0.2261768772921301,
      "flos": 18988378012800.0,
      "grad_norm": 1.8700366559525883,
      "language_loss": 0.91759664,
      "learning_rate": 3.609845118072682e-06,
      "loss": 0.93964738,
      "num_input_tokens_seen": 39987075,
      "step": 1881,
      "time_per_iteration": 2.7835686206817627
    },
    {
      "auxiliary_loss_clip": 0.01202775,
      "auxiliary_loss_mlp": 0.00765717,
      "balance_loss_clip": 1.05743432,
      "balance_loss_mlp": 1.00106931,
      "epoch": 0.2262971201827692,
      "flos": 19974054101760.0,
      "grad_norm": 1.9431831628162566,
      "language_loss": 0.80088961,
      "learning_rate": 3.6093827698809276e-06,
      "loss": 0.82057452,
      "num_input_tokens_seen": 40006175,
      "step": 1882,
      "time_per_iteration": 2.6244235038757324
    },
    {
      "auxiliary_loss_clip": 0.01190868,
      "auxiliary_loss_mlp": 0.01027632,
      "balance_loss_clip": 1.05210817,
      "balance_loss_mlp": 1.0184108,
      "epoch": 0.2264173630734083,
      "flos": 16654543735680.0,
      "grad_norm": 3.7540043296123797,
      "language_loss": 0.84287113,
      "learning_rate": 3.6089201775470864e-06,
      "loss": 0.86505616,
      "num_input_tokens_seen": 40021630,
      "step": 1883,
      "time_per_iteration": 2.6386256217956543
    },
    {
      "auxiliary_loss_clip": 0.01153287,
      "auxiliary_loss_mlp": 0.01033291,
      "balance_loss_clip": 1.04914474,
      "balance_loss_mlp": 1.02370048,
      "epoch": 0.22653760596404737,
      "flos": 24389809597440.0,
      "grad_norm": 1.6267726635396331,
      "language_loss": 0.77647734,
      "learning_rate": 3.6084573411413334e-06,
      "loss": 0.79834318,
      "num_input_tokens_seen": 40041025,
      "step": 1884,
      "time_per_iteration": 2.7672119140625
    },
    {
      "auxiliary_loss_clip": 0.0116633,
      "auxiliary_loss_mlp": 0.01042169,
      "balance_loss_clip": 1.05354762,
      "balance_loss_mlp": 1.03142858,
      "epoch": 0.22665784885468646,
      "flos": 18332541538560.0,
      "grad_norm": 2.252267114944765,
      "language_loss": 0.81393075,
      "learning_rate": 3.607994260733881e-06,
      "loss": 0.8360157,
      "num_input_tokens_seen": 40060265,
      "step": 1885,
      "time_per_iteration": 2.702138900756836
    },
    {
      "auxiliary_loss_clip": 0.01185967,
      "auxiliary_loss_mlp": 0.01030831,
      "balance_loss_clip": 1.05306232,
      "balance_loss_mlp": 1.02193809,
      "epoch": 0.22677809174532557,
      "flos": 24058102475520.0,
      "grad_norm": 2.7480258433000384,
      "language_loss": 0.74656773,
      "learning_rate": 3.6075309363949776e-06,
      "loss": 0.76873571,
      "num_input_tokens_seen": 40079435,
      "step": 1886,
      "time_per_iteration": 2.7162277698516846
    },
    {
      "auxiliary_loss_clip": 0.01210632,
      "auxiliary_loss_mlp": 0.01036099,
      "balance_loss_clip": 1.05871522,
      "balance_loss_mlp": 1.0262171,
      "epoch": 0.22689833463596465,
      "flos": 20374242503040.0,
      "grad_norm": 3.9211199745360585,
      "language_loss": 0.81431985,
      "learning_rate": 3.6070673681949094e-06,
      "loss": 0.83678722,
      "num_input_tokens_seen": 40097800,
      "step": 1887,
      "time_per_iteration": 3.5330729484558105
    },
    {
      "auxiliary_loss_clip": 0.01181156,
      "auxiliary_loss_mlp": 0.00764507,
      "balance_loss_clip": 1.05450344,
      "balance_loss_mlp": 1.00097179,
      "epoch": 0.22701857752660373,
      "flos": 30120398438400.0,
      "grad_norm": 1.8265983648189121,
      "language_loss": 0.81671679,
      "learning_rate": 3.606603556203999e-06,
      "loss": 0.83617342,
      "num_input_tokens_seen": 40122745,
      "step": 1888,
      "time_per_iteration": 2.828099250793457
    },
    {
      "auxiliary_loss_clip": 0.01196873,
      "auxiliary_loss_mlp": 0.01035888,
      "balance_loss_clip": 1.0531106,
      "balance_loss_mlp": 1.02622008,
      "epoch": 0.22713882041724284,
      "flos": 22492182084480.0,
      "grad_norm": 2.219981082596709,
      "language_loss": 0.83695358,
      "learning_rate": 3.6061395004926066e-06,
      "loss": 0.85928118,
      "num_input_tokens_seen": 40141680,
      "step": 1889,
      "time_per_iteration": 2.659557819366455
    },
    {
      "auxiliary_loss_clip": 0.01211834,
      "auxiliary_loss_mlp": 0.01034849,
      "balance_loss_clip": 1.05811119,
      "balance_loss_mlp": 1.02430511,
      "epoch": 0.22725906330788193,
      "flos": 20521548178560.0,
      "grad_norm": 7.594772763661466,
      "language_loss": 0.84833384,
      "learning_rate": 3.605675201131129e-06,
      "loss": 0.87080061,
      "num_input_tokens_seen": 40160140,
      "step": 1890,
      "time_per_iteration": 3.572787046432495
    },
    {
      "auxiliary_loss_clip": 0.0120277,
      "auxiliary_loss_mlp": 0.01032483,
      "balance_loss_clip": 1.0593431,
      "balance_loss_mlp": 1.02245748,
      "epoch": 0.227379306198521,
      "flos": 18989922297600.0,
      "grad_norm": 3.1918408291837994,
      "language_loss": 0.79583478,
      "learning_rate": 3.60521065819e-06,
      "loss": 0.81818724,
      "num_input_tokens_seen": 40177450,
      "step": 1891,
      "time_per_iteration": 3.5321075916290283
    },
    {
      "auxiliary_loss_clip": 0.01183301,
      "auxiliary_loss_mlp": 0.01035765,
      "balance_loss_clip": 1.05224764,
      "balance_loss_mlp": 1.02665722,
      "epoch": 0.2274995490891601,
      "flos": 21798351999360.0,
      "grad_norm": 2.0380865161189696,
      "language_loss": 0.88032722,
      "learning_rate": 3.60474587173969e-06,
      "loss": 0.90251791,
      "num_input_tokens_seen": 40195935,
      "step": 1892,
      "time_per_iteration": 3.6196961402893066
    },
    {
      "auxiliary_loss_clip": 0.01195752,
      "auxiliary_loss_mlp": 0.01034534,
      "balance_loss_clip": 1.06022549,
      "balance_loss_mlp": 1.02542651,
      "epoch": 0.2276197919797992,
      "flos": 19058654972160.0,
      "grad_norm": 2.3209669309471512,
      "language_loss": 0.84464121,
      "learning_rate": 3.6042808418507084e-06,
      "loss": 0.86694407,
      "num_input_tokens_seen": 40213620,
      "step": 1893,
      "time_per_iteration": 2.640303134918213
    },
    {
      "auxiliary_loss_clip": 0.01199554,
      "auxiliary_loss_mlp": 0.01040366,
      "balance_loss_clip": 1.05936956,
      "balance_loss_mlp": 1.03030443,
      "epoch": 0.22774003487043828,
      "flos": 18806777827200.0,
      "grad_norm": 2.3848659221837933,
      "language_loss": 0.7719717,
      "learning_rate": 3.6038155685935976e-06,
      "loss": 0.79437089,
      "num_input_tokens_seen": 40230190,
      "step": 1894,
      "time_per_iteration": 2.658017873764038
    },
    {
      "auxiliary_loss_clip": 0.01196357,
      "auxiliary_loss_mlp": 0.01034151,
      "balance_loss_clip": 1.05805969,
      "balance_loss_mlp": 1.02463245,
      "epoch": 0.22786027776107737,
      "flos": 23002544476800.0,
      "grad_norm": 7.368234507231393,
      "language_loss": 0.70834446,
      "learning_rate": 3.6033500520389404e-06,
      "loss": 0.73064947,
      "num_input_tokens_seen": 40246860,
      "step": 1895,
      "time_per_iteration": 2.7004048824310303
    },
    {
      "auxiliary_loss_clip": 0.01075728,
      "auxiliary_loss_mlp": 0.01004341,
      "balance_loss_clip": 1.03040862,
      "balance_loss_mlp": 1.00199306,
      "epoch": 0.22798052065171648,
      "flos": 66706872600960.0,
      "grad_norm": 0.8115595839022945,
      "language_loss": 0.64770842,
      "learning_rate": 3.6028842922573553e-06,
      "loss": 0.66850913,
      "num_input_tokens_seen": 40311005,
      "step": 1896,
      "time_per_iteration": 3.3901093006134033
    },
    {
      "auxiliary_loss_clip": 0.01088772,
      "auxiliary_loss_mlp": 0.00755887,
      "balance_loss_clip": 1.03008151,
      "balance_loss_mlp": 1.00178456,
      "epoch": 0.22810076354235556,
      "flos": 62080896758400.0,
      "grad_norm": 0.8587956336581836,
      "language_loss": 0.62911236,
      "learning_rate": 3.602418289319497e-06,
      "loss": 0.64755893,
      "num_input_tokens_seen": 40369560,
      "step": 1897,
      "time_per_iteration": 3.2240233421325684
    },
    {
      "auxiliary_loss_clip": 0.01150895,
      "auxiliary_loss_mlp": 0.01036985,
      "balance_loss_clip": 1.05127764,
      "balance_loss_mlp": 1.02713811,
      "epoch": 0.22822100643299464,
      "flos": 23876358635520.0,
      "grad_norm": 2.6377734043383363,
      "language_loss": 0.73038548,
      "learning_rate": 3.601952043296059e-06,
      "loss": 0.75226426,
      "num_input_tokens_seen": 40389555,
      "step": 1898,
      "time_per_iteration": 2.7809786796569824
    },
    {
      "auxiliary_loss_clip": 0.01184732,
      "auxiliary_loss_mlp": 0.01036367,
      "balance_loss_clip": 1.05259967,
      "balance_loss_mlp": 1.0273428,
      "epoch": 0.22834124932363373,
      "flos": 20991331180800.0,
      "grad_norm": 2.10855413442385,
      "language_loss": 0.80807167,
      "learning_rate": 3.6014855542577696e-06,
      "loss": 0.83028263,
      "num_input_tokens_seen": 40406765,
      "step": 1899,
      "time_per_iteration": 2.675363779067993
    },
    {
      "auxiliary_loss_clip": 0.01179808,
      "auxiliary_loss_mlp": 0.01035399,
      "balance_loss_clip": 1.05440485,
      "balance_loss_mlp": 1.02573121,
      "epoch": 0.22846149221427284,
      "flos": 24901572620160.0,
      "grad_norm": 1.9905290881186342,
      "language_loss": 0.84516394,
      "learning_rate": 3.6010188222753943e-06,
      "loss": 0.86731595,
      "num_input_tokens_seen": 40427535,
      "step": 1900,
      "time_per_iteration": 2.724919557571411
    },
    {
      "auxiliary_loss_clip": 0.01092268,
      "auxiliary_loss_mlp": 0.01003625,
      "balance_loss_clip": 1.02971458,
      "balance_loss_mlp": 1.00152671,
      "epoch": 0.22858173510491192,
      "flos": 56132294319360.0,
      "grad_norm": 0.8971585760056745,
      "language_loss": 0.64069796,
      "learning_rate": 3.6005518474197372e-06,
      "loss": 0.66165692,
      "num_input_tokens_seen": 40479580,
      "step": 1901,
      "time_per_iteration": 3.145097494125366
    },
    {
      "auxiliary_loss_clip": 0.01198708,
      "auxiliary_loss_mlp": 0.01037253,
      "balance_loss_clip": 1.06089973,
      "balance_loss_mlp": 1.02719784,
      "epoch": 0.228701977995551,
      "flos": 24170826332160.0,
      "grad_norm": 2.376937990197744,
      "language_loss": 0.77814448,
      "learning_rate": 3.6000846297616373e-06,
      "loss": 0.80050421,
      "num_input_tokens_seen": 40497880,
      "step": 1902,
      "time_per_iteration": 2.685399293899536
    },
    {
      "auxiliary_loss_clip": 0.01220041,
      "auxiliary_loss_mlp": 0.01034085,
      "balance_loss_clip": 1.06521249,
      "balance_loss_mlp": 1.0235765,
      "epoch": 0.22882222088619011,
      "flos": 21387892308480.0,
      "grad_norm": 2.9158543007723132,
      "language_loss": 0.72659683,
      "learning_rate": 3.5996171693719717e-06,
      "loss": 0.74913812,
      "num_input_tokens_seen": 40513975,
      "step": 1903,
      "time_per_iteration": 2.606912851333618
    },
    {
      "auxiliary_loss_clip": 0.01107056,
      "auxiliary_loss_mlp": 0.01002813,
      "balance_loss_clip": 1.02971005,
      "balance_loss_mlp": 1.00072694,
      "epoch": 0.2289424637768292,
      "flos": 64589615377920.0,
      "grad_norm": 0.8402830213619247,
      "language_loss": 0.64757788,
      "learning_rate": 3.5991494663216528e-06,
      "loss": 0.6686765,
      "num_input_tokens_seen": 40576960,
      "step": 1904,
      "time_per_iteration": 3.2863552570343018
    },
    {
      "auxiliary_loss_clip": 0.01213358,
      "auxiliary_loss_mlp": 0.01026189,
      "balance_loss_clip": 1.06058931,
      "balance_loss_mlp": 1.0162828,
      "epoch": 0.22906270666746828,
      "flos": 22163419877760.0,
      "grad_norm": 1.9723511631818114,
      "language_loss": 0.87554455,
      "learning_rate": 3.5986815206816314e-06,
      "loss": 0.89793998,
      "num_input_tokens_seen": 40595780,
      "step": 1905,
      "time_per_iteration": 2.5956368446350098
    },
    {
      "auxiliary_loss_clip": 0.01210286,
      "auxiliary_loss_mlp": 0.01031377,
      "balance_loss_clip": 1.05881751,
      "balance_loss_mlp": 1.02174437,
      "epoch": 0.2291829495581074,
      "flos": 25772334122880.0,
      "grad_norm": 2.5509605689945407,
      "language_loss": 0.7457096,
      "learning_rate": 3.598213332522895e-06,
      "loss": 0.76812625,
      "num_input_tokens_seen": 40615810,
      "step": 1906,
      "time_per_iteration": 2.687082529067993
    },
    {
      "auxiliary_loss_clip": 0.01197247,
      "auxiliary_loss_mlp": 0.0103449,
      "balance_loss_clip": 1.05723798,
      "balance_loss_mlp": 1.02444696,
      "epoch": 0.22930319244874647,
      "flos": 31172760126720.0,
      "grad_norm": 1.8577766388843853,
      "language_loss": 0.77467829,
      "learning_rate": 3.597744901916466e-06,
      "loss": 0.79699564,
      "num_input_tokens_seen": 40637095,
      "step": 1907,
      "time_per_iteration": 2.685657024383545
    },
    {
      "auxiliary_loss_clip": 0.01217108,
      "auxiliary_loss_mlp": 0.01036086,
      "balance_loss_clip": 1.06028461,
      "balance_loss_mlp": 1.02623916,
      "epoch": 0.22942343533938556,
      "flos": 23254098399360.0,
      "grad_norm": 2.185938560597546,
      "language_loss": 0.76440191,
      "learning_rate": 3.5972762289334058e-06,
      "loss": 0.7869339,
      "num_input_tokens_seen": 40656725,
      "step": 1908,
      "time_per_iteration": 2.62160587310791
    },
    {
      "auxiliary_loss_clip": 0.01135027,
      "auxiliary_loss_mlp": 0.01040068,
      "balance_loss_clip": 1.05355132,
      "balance_loss_mlp": 1.02912402,
      "epoch": 0.22954367823002464,
      "flos": 14610903436800.0,
      "grad_norm": 10.292343208175309,
      "language_loss": 0.85200208,
      "learning_rate": 3.5968073136448116e-06,
      "loss": 0.87375307,
      "num_input_tokens_seen": 40674745,
      "step": 1909,
      "time_per_iteration": 2.665807008743286
    },
    {
      "auxiliary_loss_clip": 0.0120058,
      "auxiliary_loss_mlp": 0.01028513,
      "balance_loss_clip": 1.05749691,
      "balance_loss_mlp": 1.01921439,
      "epoch": 0.22966392112066375,
      "flos": 16763604405120.0,
      "grad_norm": 5.567898357289049,
      "language_loss": 0.91310298,
      "learning_rate": 3.596338156121818e-06,
      "loss": 0.93539393,
      "num_input_tokens_seen": 40693630,
      "step": 1910,
      "time_per_iteration": 2.5899336338043213
    },
    {
      "auxiliary_loss_clip": 0.01088361,
      "auxiliary_loss_mlp": 0.01008881,
      "balance_loss_clip": 1.0258944,
      "balance_loss_mlp": 1.00684214,
      "epoch": 0.22978416401130283,
      "flos": 67474247783040.0,
      "grad_norm": 0.827635615357902,
      "language_loss": 0.59336019,
      "learning_rate": 3.595868756435595e-06,
      "loss": 0.61433262,
      "num_input_tokens_seen": 40761310,
      "step": 1911,
      "time_per_iteration": 3.2890682220458984
    },
    {
      "auxiliary_loss_clip": 0.01174272,
      "auxiliary_loss_mlp": 0.01036892,
      "balance_loss_clip": 1.05849361,
      "balance_loss_mlp": 1.02583528,
      "epoch": 0.22990440690194192,
      "flos": 19865137086720.0,
      "grad_norm": 2.3446043564954606,
      "language_loss": 0.80218375,
      "learning_rate": 3.5953991146573504e-06,
      "loss": 0.8242954,
      "num_input_tokens_seen": 40779955,
      "step": 1912,
      "time_per_iteration": 2.6832809448242188
    },
    {
      "auxiliary_loss_clip": 0.01200041,
      "auxiliary_loss_mlp": 0.0103906,
      "balance_loss_clip": 1.0551542,
      "balance_loss_mlp": 1.02824759,
      "epoch": 0.23002464979258103,
      "flos": 13289246507520.0,
      "grad_norm": 3.810678026521688,
      "language_loss": 0.83133256,
      "learning_rate": 3.5949292308583294e-06,
      "loss": 0.85372359,
      "num_input_tokens_seen": 40793200,
      "step": 1913,
      "time_per_iteration": 3.449061632156372
    },
    {
      "auxiliary_loss_clip": 0.01213864,
      "auxiliary_loss_mlp": 0.01040536,
      "balance_loss_clip": 1.06106162,
      "balance_loss_mlp": 1.02989066,
      "epoch": 0.2301448926832201,
      "flos": 22163779013760.0,
      "grad_norm": 2.312903276174975,
      "language_loss": 0.80481696,
      "learning_rate": 3.594459105109811e-06,
      "loss": 0.82736099,
      "num_input_tokens_seen": 40812380,
      "step": 1914,
      "time_per_iteration": 2.5734827518463135
    },
    {
      "auxiliary_loss_clip": 0.01202703,
      "auxiliary_loss_mlp": 0.01038215,
      "balance_loss_clip": 1.05928516,
      "balance_loss_mlp": 1.02820802,
      "epoch": 0.2302651355738592,
      "flos": 20704477167360.0,
      "grad_norm": 2.0535223142972967,
      "language_loss": 0.81305301,
      "learning_rate": 3.593988737483115e-06,
      "loss": 0.83546221,
      "num_input_tokens_seen": 40832320,
      "step": 1915,
      "time_per_iteration": 2.6523566246032715
    },
    {
      "auxiliary_loss_clip": 0.01184269,
      "auxiliary_loss_mlp": 0.01035808,
      "balance_loss_clip": 1.05791116,
      "balance_loss_mlp": 1.02538323,
      "epoch": 0.23038537846449827,
      "flos": 18588943797120.0,
      "grad_norm": 2.612209647421837,
      "language_loss": 0.78474045,
      "learning_rate": 3.5935181280495947e-06,
      "loss": 0.80694115,
      "num_input_tokens_seen": 40850900,
      "step": 1916,
      "time_per_iteration": 3.6101200580596924
    },
    {
      "auxiliary_loss_clip": 0.01081584,
      "auxiliary_loss_mlp": 0.01003799,
      "balance_loss_clip": 1.0212599,
      "balance_loss_mlp": 1.00158203,
      "epoch": 0.23050562135513739,
      "flos": 64224260190720.0,
      "grad_norm": 0.8002380842083977,
      "language_loss": 0.54299456,
      "learning_rate": 3.5930472768806412e-06,
      "loss": 0.56384844,
      "num_input_tokens_seen": 40909570,
      "step": 1917,
      "time_per_iteration": 4.175180196762085
    },
    {
      "auxiliary_loss_clip": 0.01213102,
      "auxiliary_loss_mlp": 0.01032487,
      "balance_loss_clip": 1.06085348,
      "balance_loss_mlp": 1.02248502,
      "epoch": 0.23062586424577647,
      "flos": 17313396952320.0,
      "grad_norm": 2.9830732562478968,
      "language_loss": 0.77248573,
      "learning_rate": 3.5925761840476826e-06,
      "loss": 0.7949416,
      "num_input_tokens_seen": 40928180,
      "step": 1918,
      "time_per_iteration": 3.5137860774993896
    },
    {
      "auxiliary_loss_clip": 0.01180747,
      "auxiliary_loss_mlp": 0.01032016,
      "balance_loss_clip": 1.05970621,
      "balance_loss_mlp": 1.02236021,
      "epoch": 0.23074610713641555,
      "flos": 27855979194240.0,
      "grad_norm": 2.3059921643812813,
      "language_loss": 0.8144871,
      "learning_rate": 3.592104849622183e-06,
      "loss": 0.83661479,
      "num_input_tokens_seen": 40950435,
      "step": 1919,
      "time_per_iteration": 2.6971824169158936
    },
    {
      "auxiliary_loss_clip": 0.01145746,
      "auxiliary_loss_mlp": 0.01039485,
      "balance_loss_clip": 1.05359972,
      "balance_loss_mlp": 1.02997208,
      "epoch": 0.23086635002705466,
      "flos": 28841798937600.0,
      "grad_norm": 1.648637048108304,
      "language_loss": 0.72957635,
      "learning_rate": 3.591633273675644e-06,
      "loss": 0.7514286,
      "num_input_tokens_seen": 40972670,
      "step": 1920,
      "time_per_iteration": 2.8208978176116943
    },
    {
      "auxiliary_loss_clip": 0.0108051,
      "auxiliary_loss_mlp": 0.01009847,
      "balance_loss_clip": 1.03736222,
      "balance_loss_mlp": 1.00721216,
      "epoch": 0.23098659291769374,
      "flos": 62923681566720.0,
      "grad_norm": 0.9089346813454777,
      "language_loss": 0.58154774,
      "learning_rate": 3.591161456279602e-06,
      "loss": 0.60245132,
      "num_input_tokens_seen": 41018215,
      "step": 1921,
      "time_per_iteration": 3.171978235244751
    },
    {
      "auxiliary_loss_clip": 0.01191631,
      "auxiliary_loss_mlp": 0.01036939,
      "balance_loss_clip": 1.05791402,
      "balance_loss_mlp": 1.02693141,
      "epoch": 0.23110683580833283,
      "flos": 23476816679040.0,
      "grad_norm": 4.172921017366292,
      "language_loss": 0.80441523,
      "learning_rate": 3.590689397505633e-06,
      "loss": 0.82670093,
      "num_input_tokens_seen": 41039125,
      "step": 1922,
      "time_per_iteration": 2.7086033821105957
    },
    {
      "auxiliary_loss_clip": 0.01209856,
      "auxiliary_loss_mlp": 0.0103794,
      "balance_loss_clip": 1.05955362,
      "balance_loss_mlp": 1.02856374,
      "epoch": 0.2312270786989719,
      "flos": 27271066124160.0,
      "grad_norm": 1.9984050822606354,
      "language_loss": 0.86853862,
      "learning_rate": 3.590217097425347e-06,
      "loss": 0.89101654,
      "num_input_tokens_seen": 41059025,
      "step": 1923,
      "time_per_iteration": 2.6914286613464355
    },
    {
      "auxiliary_loss_clip": 0.01217749,
      "auxiliary_loss_mlp": 0.01038885,
      "balance_loss_clip": 1.06441355,
      "balance_loss_mlp": 1.02849615,
      "epoch": 0.23134732158961102,
      "flos": 13261344618240.0,
      "grad_norm": 3.180738552801169,
      "language_loss": 0.70619231,
      "learning_rate": 3.589744556110391e-06,
      "loss": 0.72875869,
      "num_input_tokens_seen": 41077015,
      "step": 1924,
      "time_per_iteration": 2.579301595687866
    },
    {
      "auxiliary_loss_clip": 0.01177917,
      "auxiliary_loss_mlp": 0.01036419,
      "balance_loss_clip": 1.05336607,
      "balance_loss_mlp": 1.02614331,
      "epoch": 0.2314675644802501,
      "flos": 36977648250240.0,
      "grad_norm": 2.0494474299839585,
      "language_loss": 0.84518307,
      "learning_rate": 3.58927177363245e-06,
      "loss": 0.8673265,
      "num_input_tokens_seen": 41099840,
      "step": 1925,
      "time_per_iteration": 2.826348304748535
    },
    {
      "auxiliary_loss_clip": 0.01161051,
      "auxiliary_loss_mlp": 0.01034788,
      "balance_loss_clip": 1.05109394,
      "balance_loss_mlp": 1.02326059,
      "epoch": 0.2315878073708892,
      "flos": 23842207779840.0,
      "grad_norm": 4.2063729783792825,
      "language_loss": 0.72526681,
      "learning_rate": 3.5887987500632447e-06,
      "loss": 0.74722517,
      "num_input_tokens_seen": 41117845,
      "step": 1926,
      "time_per_iteration": 2.6821634769439697
    },
    {
      "auxiliary_loss_clip": 0.01174421,
      "auxiliary_loss_mlp": 0.01028904,
      "balance_loss_clip": 1.05630469,
      "balance_loss_mlp": 1.01932585,
      "epoch": 0.2317080502615283,
      "flos": 23039424766080.0,
      "grad_norm": 2.9251378631771043,
      "language_loss": 0.8425622,
      "learning_rate": 3.5883254854745325e-06,
      "loss": 0.86459547,
      "num_input_tokens_seen": 41136235,
      "step": 1927,
      "time_per_iteration": 2.7368109226226807
    },
    {
      "auxiliary_loss_clip": 0.01203568,
      "auxiliary_loss_mlp": 0.01032833,
      "balance_loss_clip": 1.05825496,
      "balance_loss_mlp": 1.02227163,
      "epoch": 0.23182829315216738,
      "flos": 11254656435840.0,
      "grad_norm": 2.8369511866931068,
      "language_loss": 0.75266194,
      "learning_rate": 3.587851979938107e-06,
      "loss": 0.77502596,
      "num_input_tokens_seen": 41153125,
      "step": 1928,
      "time_per_iteration": 2.6279940605163574
    },
    {
      "auxiliary_loss_clip": 0.01199699,
      "auxiliary_loss_mlp": 0.01044479,
      "balance_loss_clip": 1.05915332,
      "balance_loss_mlp": 1.0340662,
      "epoch": 0.23194853604280646,
      "flos": 19828939155840.0,
      "grad_norm": 2.5173860282966154,
      "language_loss": 0.77202064,
      "learning_rate": 3.5873782335257985e-06,
      "loss": 0.79446244,
      "num_input_tokens_seen": 41171290,
      "step": 1929,
      "time_per_iteration": 2.651021957397461
    },
    {
      "auxiliary_loss_clip": 0.01169085,
      "auxiliary_loss_mlp": 0.01043019,
      "balance_loss_clip": 1.057814,
      "balance_loss_mlp": 1.03212285,
      "epoch": 0.23206877893344555,
      "flos": 15305020830720.0,
      "grad_norm": 2.2277469218688153,
      "language_loss": 0.78823781,
      "learning_rate": 3.5869042463094744e-06,
      "loss": 0.81035888,
      "num_input_tokens_seen": 41189005,
      "step": 1930,
      "time_per_iteration": 2.6858103275299072
    },
    {
      "auxiliary_loss_clip": 0.01136646,
      "auxiliary_loss_mlp": 0.01034124,
      "balance_loss_clip": 1.04988813,
      "balance_loss_mlp": 1.02356768,
      "epoch": 0.23218902182408466,
      "flos": 22711488572160.0,
      "grad_norm": 1.869988668749133,
      "language_loss": 0.7731728,
      "learning_rate": 3.586430018361038e-06,
      "loss": 0.79488045,
      "num_input_tokens_seen": 41208775,
      "step": 1931,
      "time_per_iteration": 2.7825987339019775
    },
    {
      "auxiliary_loss_clip": 0.01169141,
      "auxiliary_loss_mlp": 0.01040569,
      "balance_loss_clip": 1.05244637,
      "balance_loss_mlp": 1.03009629,
      "epoch": 0.23230926471472374,
      "flos": 22710734386560.0,
      "grad_norm": 1.9608534513653555,
      "language_loss": 0.75923729,
      "learning_rate": 3.5859555497524283e-06,
      "loss": 0.7813344,
      "num_input_tokens_seen": 41226010,
      "step": 1932,
      "time_per_iteration": 2.673743963241577
    },
    {
      "auxiliary_loss_clip": 0.01199449,
      "auxiliary_loss_mlp": 0.0103734,
      "balance_loss_clip": 1.06053948,
      "balance_loss_mlp": 1.02780342,
      "epoch": 0.23242950760536282,
      "flos": 20375499479040.0,
      "grad_norm": 2.302545750135597,
      "language_loss": 0.9195286,
      "learning_rate": 3.5854808405556237e-06,
      "loss": 0.9418965,
      "num_input_tokens_seen": 41245245,
      "step": 1933,
      "time_per_iteration": 2.720630645751953
    },
    {
      "auxiliary_loss_clip": 0.01169323,
      "auxiliary_loss_mlp": 0.01033053,
      "balance_loss_clip": 1.05411923,
      "balance_loss_mlp": 1.02361107,
      "epoch": 0.23254975049600193,
      "flos": 16908324301440.0,
      "grad_norm": 2.6518617984897443,
      "language_loss": 0.7640788,
      "learning_rate": 3.5850058908426355e-06,
      "loss": 0.78610253,
      "num_input_tokens_seen": 41263795,
      "step": 1934,
      "time_per_iteration": 2.7200772762298584
    },
    {
      "auxiliary_loss_clip": 0.01183608,
      "auxiliary_loss_mlp": 0.01036463,
      "balance_loss_clip": 1.053267,
      "balance_loss_mlp": 1.02701616,
      "epoch": 0.23266999338664102,
      "flos": 23294821443840.0,
      "grad_norm": 2.010392403267564,
      "language_loss": 0.85563564,
      "learning_rate": 3.584530700685514e-06,
      "loss": 0.87783635,
      "num_input_tokens_seen": 41284055,
      "step": 1935,
      "time_per_iteration": 2.721214771270752
    },
    {
      "auxiliary_loss_clip": 0.01181151,
      "auxiliary_loss_mlp": 0.0103207,
      "balance_loss_clip": 1.0590781,
      "balance_loss_mlp": 1.022259,
      "epoch": 0.2327902362772801,
      "flos": 19569987031680.0,
      "grad_norm": 2.215274319574889,
      "language_loss": 0.8867116,
      "learning_rate": 3.5840552701563448e-06,
      "loss": 0.90884387,
      "num_input_tokens_seen": 41300255,
      "step": 1936,
      "time_per_iteration": 2.6790263652801514
    },
    {
      "auxiliary_loss_clip": 0.01210016,
      "auxiliary_loss_mlp": 0.01038871,
      "balance_loss_clip": 1.05832052,
      "balance_loss_mlp": 1.02956665,
      "epoch": 0.2329104791679192,
      "flos": 16727514215040.0,
      "grad_norm": 2.561820210003978,
      "language_loss": 0.82029152,
      "learning_rate": 3.5835795993272513e-06,
      "loss": 0.84278047,
      "num_input_tokens_seen": 41318540,
      "step": 1937,
      "time_per_iteration": 2.5738909244537354
    },
    {
      "auxiliary_loss_clip": 0.01105222,
      "auxiliary_loss_mlp": 0.01036501,
      "balance_loss_clip": 1.04756975,
      "balance_loss_mlp": 1.02619481,
      "epoch": 0.2330307220585583,
      "flos": 22163743100160.0,
      "grad_norm": 1.9887716808686196,
      "language_loss": 0.71402603,
      "learning_rate": 3.583103688270391e-06,
      "loss": 0.73544323,
      "num_input_tokens_seen": 41338320,
      "step": 1938,
      "time_per_iteration": 3.004837989807129
    },
    {
      "auxiliary_loss_clip": 0.01168089,
      "auxiliary_loss_mlp": 0.01032118,
      "balance_loss_clip": 1.05370998,
      "balance_loss_mlp": 1.0220325,
      "epoch": 0.23315096494919738,
      "flos": 19317319787520.0,
      "grad_norm": 3.1969866990678324,
      "language_loss": 0.89302576,
      "learning_rate": 3.58262753705796e-06,
      "loss": 0.9150278,
      "num_input_tokens_seen": 41353210,
      "step": 1939,
      "time_per_iteration": 4.1691484451293945
    },
    {
      "auxiliary_loss_clip": 0.01083228,
      "auxiliary_loss_mlp": 0.01008097,
      "balance_loss_clip": 1.02596927,
      "balance_loss_mlp": 1.00587928,
      "epoch": 0.23327120783983646,
      "flos": 53031048946560.0,
      "grad_norm": 0.7869695707959454,
      "language_loss": 0.55556172,
      "learning_rate": 3.5821511457621902e-06,
      "loss": 0.57647502,
      "num_input_tokens_seen": 41410510,
      "step": 1940,
      "time_per_iteration": 3.253009557723999
    },
    {
      "auxiliary_loss_clip": 0.01176109,
      "auxiliary_loss_mlp": 0.01047007,
      "balance_loss_clip": 1.0559504,
      "balance_loss_mlp": 1.03583133,
      "epoch": 0.23339145073047557,
      "flos": 17126984344320.0,
      "grad_norm": 16.358084656869632,
      "language_loss": 0.81418169,
      "learning_rate": 3.5816745144553497e-06,
      "loss": 0.83641285,
      "num_input_tokens_seen": 41425830,
      "step": 1941,
      "time_per_iteration": 2.668553352355957
    },
    {
      "auxiliary_loss_clip": 0.01147844,
      "auxiliary_loss_mlp": 0.01030666,
      "balance_loss_clip": 1.05245125,
      "balance_loss_mlp": 1.02021778,
      "epoch": 0.23351169362111465,
      "flos": 13078918419840.0,
      "grad_norm": 1.8874232734363083,
      "language_loss": 0.75640577,
      "learning_rate": 3.5811976432097424e-06,
      "loss": 0.77819085,
      "num_input_tokens_seen": 41443500,
      "step": 1942,
      "time_per_iteration": 3.7219457626342773
    },
    {
      "auxiliary_loss_clip": 0.01197136,
      "auxiliary_loss_mlp": 0.00765487,
      "balance_loss_clip": 1.0590204,
      "balance_loss_mlp": 1.0008384,
      "epoch": 0.23363193651175373,
      "flos": 15851257931520.0,
      "grad_norm": 2.652551843517523,
      "language_loss": 0.84654796,
      "learning_rate": 3.58072053209771e-06,
      "loss": 0.86617416,
      "num_input_tokens_seen": 41460055,
      "step": 1943,
      "time_per_iteration": 3.5429184436798096
    },
    {
      "auxiliary_loss_clip": 0.01175832,
      "auxiliary_loss_mlp": 0.01041339,
      "balance_loss_clip": 1.05335808,
      "balance_loss_mlp": 1.03051448,
      "epoch": 0.23375217940239285,
      "flos": 21025769345280.0,
      "grad_norm": 2.0992521587035395,
      "language_loss": 0.7889539,
      "learning_rate": 3.5802431811916296e-06,
      "loss": 0.81112564,
      "num_input_tokens_seen": 41476665,
      "step": 1944,
      "time_per_iteration": 3.6838607788085938
    },
    {
      "auxiliary_loss_clip": 0.01177648,
      "auxiliary_loss_mlp": 0.0103747,
      "balance_loss_clip": 1.05498183,
      "balance_loss_mlp": 1.02807057,
      "epoch": 0.23387242229303193,
      "flos": 20594698225920.0,
      "grad_norm": 2.935185755685507,
      "language_loss": 0.80402327,
      "learning_rate": 3.579765590563916e-06,
      "loss": 0.82617438,
      "num_input_tokens_seen": 41496065,
      "step": 1945,
      "time_per_iteration": 2.706803321838379
    },
    {
      "auxiliary_loss_clip": 0.0118592,
      "auxiliary_loss_mlp": 0.01038104,
      "balance_loss_clip": 1.0556953,
      "balance_loss_mlp": 1.02832901,
      "epoch": 0.233992665183671,
      "flos": 24279491952000.0,
      "grad_norm": 3.686965369359839,
      "language_loss": 0.82125109,
      "learning_rate": 3.579287760287017e-06,
      "loss": 0.84349132,
      "num_input_tokens_seen": 41516815,
      "step": 1946,
      "time_per_iteration": 2.7241063117980957
    },
    {
      "auxiliary_loss_clip": 0.01193413,
      "auxiliary_loss_mlp": 0.01033345,
      "balance_loss_clip": 1.0570004,
      "balance_loss_mlp": 1.023844,
      "epoch": 0.2341129080743101,
      "flos": 30154621121280.0,
      "grad_norm": 2.3239861432827067,
      "language_loss": 0.72890013,
      "learning_rate": 3.578809690433421e-06,
      "loss": 0.75116777,
      "num_input_tokens_seen": 41538525,
      "step": 1947,
      "time_per_iteration": 2.6950912475585938
    },
    {
      "auxiliary_loss_clip": 0.01218651,
      "auxiliary_loss_mlp": 0.01042904,
      "balance_loss_clip": 1.06217957,
      "balance_loss_mlp": 1.03251505,
      "epoch": 0.2342331509649492,
      "flos": 22784135829120.0,
      "grad_norm": 2.3303986744999197,
      "language_loss": 0.81401575,
      "learning_rate": 3.578331381075651e-06,
      "loss": 0.8366313,
      "num_input_tokens_seen": 41559025,
      "step": 1948,
      "time_per_iteration": 2.5736050605773926
    },
    {
      "auxiliary_loss_clip": 0.01198886,
      "auxiliary_loss_mlp": 0.01043065,
      "balance_loss_clip": 1.05799699,
      "balance_loss_mlp": 1.03338504,
      "epoch": 0.2343533938555883,
      "flos": 23623152687360.0,
      "grad_norm": 3.179421382306876,
      "language_loss": 0.69912469,
      "learning_rate": 3.5778528322862646e-06,
      "loss": 0.72154427,
      "num_input_tokens_seen": 41577845,
      "step": 1949,
      "time_per_iteration": 2.662689447402954
    },
    {
      "auxiliary_loss_clip": 0.01200461,
      "auxiliary_loss_mlp": 0.01032384,
      "balance_loss_clip": 1.05766606,
      "balance_loss_mlp": 1.02241182,
      "epoch": 0.23447363674622737,
      "flos": 24570332375040.0,
      "grad_norm": 1.6018091473868232,
      "language_loss": 0.86570346,
      "learning_rate": 3.5773740441378585e-06,
      "loss": 0.88803196,
      "num_input_tokens_seen": 41598600,
      "step": 1950,
      "time_per_iteration": 2.632403612136841
    },
    {
      "auxiliary_loss_clip": 0.01197038,
      "auxiliary_loss_mlp": 0.01030992,
      "balance_loss_clip": 1.05846024,
      "balance_loss_mlp": 1.02168155,
      "epoch": 0.23459387963686648,
      "flos": 53140322119680.0,
      "grad_norm": 1.9233195385520032,
      "language_loss": 0.73819208,
      "learning_rate": 3.5768950167030633e-06,
      "loss": 0.76047242,
      "num_input_tokens_seen": 41623300,
      "step": 1951,
      "time_per_iteration": 2.9693715572357178
    },
    {
      "auxiliary_loss_clip": 0.01173176,
      "auxiliary_loss_mlp": 0.01040782,
      "balance_loss_clip": 1.05385244,
      "balance_loss_mlp": 1.03060162,
      "epoch": 0.23471412252750556,
      "flos": 23951412103680.0,
      "grad_norm": 1.8057207786555303,
      "language_loss": 0.78311634,
      "learning_rate": 3.576415750054548e-06,
      "loss": 0.80525595,
      "num_input_tokens_seen": 41643420,
      "step": 1952,
      "time_per_iteration": 2.7148361206054688
    },
    {
      "auxiliary_loss_clip": 0.0117159,
      "auxiliary_loss_mlp": 0.01035473,
      "balance_loss_clip": 1.05407071,
      "balance_loss_mlp": 1.02624643,
      "epoch": 0.23483436541814465,
      "flos": 15706573948800.0,
      "grad_norm": 1.9006016815197615,
      "language_loss": 0.85836512,
      "learning_rate": 3.5759362442650172e-06,
      "loss": 0.88043571,
      "num_input_tokens_seen": 41660170,
      "step": 1953,
      "time_per_iteration": 2.6864397525787354
    },
    {
      "auxiliary_loss_clip": 0.01201631,
      "auxiliary_loss_mlp": 0.01035244,
      "balance_loss_clip": 1.06293678,
      "balance_loss_mlp": 1.02468765,
      "epoch": 0.23495460830878373,
      "flos": 24936262179840.0,
      "grad_norm": 2.451700401787181,
      "language_loss": 0.85213423,
      "learning_rate": 3.5754564994072113e-06,
      "loss": 0.87450302,
      "num_input_tokens_seen": 41679010,
      "step": 1954,
      "time_per_iteration": 2.631483554840088
    },
    {
      "auxiliary_loss_clip": 0.01179444,
      "auxiliary_loss_mlp": 0.01040768,
      "balance_loss_clip": 1.05448854,
      "balance_loss_mlp": 1.03027129,
      "epoch": 0.23507485119942284,
      "flos": 30482665056000.0,
      "grad_norm": 2.281646406729147,
      "language_loss": 0.60124522,
      "learning_rate": 3.5749765155539067e-06,
      "loss": 0.62344736,
      "num_input_tokens_seen": 41699495,
      "step": 1955,
      "time_per_iteration": 2.7441000938415527
    },
    {
      "auxiliary_loss_clip": 0.0116821,
      "auxiliary_loss_mlp": 0.01037683,
      "balance_loss_clip": 1.05503154,
      "balance_loss_mlp": 1.02620959,
      "epoch": 0.23519509409006192,
      "flos": 18329129746560.0,
      "grad_norm": 2.2563439850321845,
      "language_loss": 0.92266214,
      "learning_rate": 3.574496292777917e-06,
      "loss": 0.9447211,
      "num_input_tokens_seen": 41717705,
      "step": 1956,
      "time_per_iteration": 2.713909149169922
    },
    {
      "auxiliary_loss_clip": 0.01190576,
      "auxiliary_loss_mlp": 0.01036396,
      "balance_loss_clip": 1.05750108,
      "balance_loss_mlp": 1.02549386,
      "epoch": 0.235315336980701,
      "flos": 29643217234560.0,
      "grad_norm": 4.416154661613202,
      "language_loss": 0.71510023,
      "learning_rate": 3.574015831152092e-06,
      "loss": 0.73736995,
      "num_input_tokens_seen": 41738120,
      "step": 1957,
      "time_per_iteration": 2.7646992206573486
    },
    {
      "auxiliary_loss_clip": 0.01171297,
      "auxiliary_loss_mlp": 0.01037641,
      "balance_loss_clip": 1.05550623,
      "balance_loss_mlp": 1.02797902,
      "epoch": 0.23543557987134012,
      "flos": 18551704371840.0,
      "grad_norm": 2.1657030861780986,
      "language_loss": 0.83356643,
      "learning_rate": 3.573535130749316e-06,
      "loss": 0.85565585,
      "num_input_tokens_seen": 41756070,
      "step": 1958,
      "time_per_iteration": 2.7113566398620605
    },
    {
      "auxiliary_loss_clip": 0.01173824,
      "auxiliary_loss_mlp": 0.01040674,
      "balance_loss_clip": 1.05674732,
      "balance_loss_mlp": 1.03058875,
      "epoch": 0.2355558227619792,
      "flos": 24679033908480.0,
      "grad_norm": 2.32715424022086,
      "language_loss": 0.73966348,
      "learning_rate": 3.5730541916425127e-06,
      "loss": 0.76180845,
      "num_input_tokens_seen": 41777550,
      "step": 1959,
      "time_per_iteration": 2.760509490966797
    },
    {
      "auxiliary_loss_clip": 0.011676,
      "auxiliary_loss_mlp": 0.01036635,
      "balance_loss_clip": 1.05380201,
      "balance_loss_mlp": 1.0263834,
      "epoch": 0.23567606565261828,
      "flos": 21944795748480.0,
      "grad_norm": 4.497737178257931,
      "language_loss": 0.86270058,
      "learning_rate": 3.572573013904639e-06,
      "loss": 0.88474298,
      "num_input_tokens_seen": 41797460,
      "step": 1960,
      "time_per_iteration": 2.7796976566314697
    },
    {
      "auxiliary_loss_clip": 0.01209561,
      "auxiliary_loss_mlp": 0.01031886,
      "balance_loss_clip": 1.05805063,
      "balance_loss_mlp": 1.02212906,
      "epoch": 0.2357963085432574,
      "flos": 13589352639360.0,
      "grad_norm": 2.144069170185652,
      "language_loss": 0.91961336,
      "learning_rate": 3.572091597608689e-06,
      "loss": 0.94202781,
      "num_input_tokens_seen": 41815585,
      "step": 1961,
      "time_per_iteration": 2.7263734340667725
    },
    {
      "auxiliary_loss_clip": 0.01188851,
      "auxiliary_loss_mlp": 0.0103846,
      "balance_loss_clip": 1.05745411,
      "balance_loss_mlp": 1.02827907,
      "epoch": 0.23591655143389648,
      "flos": 22088689632000.0,
      "grad_norm": 2.2484658261004653,
      "language_loss": 0.73312354,
      "learning_rate": 3.571609942827694e-06,
      "loss": 0.7553966,
      "num_input_tokens_seen": 41834700,
      "step": 1962,
      "time_per_iteration": 2.7929601669311523
    },
    {
      "auxiliary_loss_clip": 0.01179431,
      "auxiliary_loss_mlp": 0.01038521,
      "balance_loss_clip": 1.05456209,
      "balance_loss_mlp": 1.0288527,
      "epoch": 0.23603679432453556,
      "flos": 17017349057280.0,
      "grad_norm": 1.7712787463125517,
      "language_loss": 0.88594162,
      "learning_rate": 3.57112804963472e-06,
      "loss": 0.90812111,
      "num_input_tokens_seen": 41852915,
      "step": 1963,
      "time_per_iteration": 2.728073835372925
    },
    {
      "auxiliary_loss_clip": 0.01161727,
      "auxiliary_loss_mlp": 0.01039478,
      "balance_loss_clip": 1.05680025,
      "balance_loss_mlp": 1.03001904,
      "epoch": 0.23615703721517464,
      "flos": 19171307001600.0,
      "grad_norm": 1.7904824707714606,
      "language_loss": 0.76605976,
      "learning_rate": 3.57064591810287e-06,
      "loss": 0.78807181,
      "num_input_tokens_seen": 41870415,
      "step": 1964,
      "time_per_iteration": 2.818753719329834
    },
    {
      "auxiliary_loss_clip": 0.01212821,
      "auxiliary_loss_mlp": 0.00764456,
      "balance_loss_clip": 1.0614841,
      "balance_loss_mlp": 1.00066268,
      "epoch": 0.23627728010581375,
      "flos": 19098803399040.0,
      "grad_norm": 2.4922525355660943,
      "language_loss": 0.80490601,
      "learning_rate": 3.570163548305284e-06,
      "loss": 0.82467878,
      "num_input_tokens_seen": 41889345,
      "step": 1965,
      "time_per_iteration": 3.607412576675415
    },
    {
      "auxiliary_loss_clip": 0.01183061,
      "auxiliary_loss_mlp": 0.01038946,
      "balance_loss_clip": 1.05592847,
      "balance_loss_mlp": 1.02848554,
      "epoch": 0.23639752299645284,
      "flos": 14282213057280.0,
      "grad_norm": 3.483136336137784,
      "language_loss": 0.70155245,
      "learning_rate": 3.569680940315135e-06,
      "loss": 0.72377253,
      "num_input_tokens_seen": 41905745,
      "step": 1966,
      "time_per_iteration": 2.725879669189453
    },
    {
      "auxiliary_loss_clip": 0.01173689,
      "auxiliary_loss_mlp": 0.01035078,
      "balance_loss_clip": 1.05448198,
      "balance_loss_mlp": 1.02400923,
      "epoch": 0.23651776588709192,
      "flos": 22893411980160.0,
      "grad_norm": 2.008246096625574,
      "language_loss": 0.81737256,
      "learning_rate": 3.5691980942056356e-06,
      "loss": 0.83946025,
      "num_input_tokens_seen": 41925115,
      "step": 1967,
      "time_per_iteration": 3.782923936843872
    },
    {
      "auxiliary_loss_clip": 0.01198319,
      "auxiliary_loss_mlp": 0.01034648,
      "balance_loss_clip": 1.05576921,
      "balance_loss_mlp": 1.02404404,
      "epoch": 0.23663800877773103,
      "flos": 18624531196800.0,
      "grad_norm": 2.1170206880090614,
      "language_loss": 0.79525512,
      "learning_rate": 3.5687150100500332e-06,
      "loss": 0.81758481,
      "num_input_tokens_seen": 41944815,
      "step": 1968,
      "time_per_iteration": 2.651975631713867
    },
    {
      "auxiliary_loss_clip": 0.01198145,
      "auxiliary_loss_mlp": 0.01040294,
      "balance_loss_clip": 1.05793345,
      "balance_loss_mlp": 1.0303638,
      "epoch": 0.2367582516683701,
      "flos": 25555828896000.0,
      "grad_norm": 1.733356478797489,
      "language_loss": 0.7453838,
      "learning_rate": 3.568231687921611e-06,
      "loss": 0.76776826,
      "num_input_tokens_seen": 41964990,
      "step": 1969,
      "time_per_iteration": 4.4237449169158936
    },
    {
      "auxiliary_loss_clip": 0.01207899,
      "auxiliary_loss_mlp": 0.01039275,
      "balance_loss_clip": 1.05738711,
      "balance_loss_mlp": 1.02999496,
      "epoch": 0.2368784945590092,
      "flos": 23295072839040.0,
      "grad_norm": 1.5623991225259253,
      "language_loss": 0.80483359,
      "learning_rate": 3.5677481278936883e-06,
      "loss": 0.82730532,
      "num_input_tokens_seen": 41984570,
      "step": 1970,
      "time_per_iteration": 2.6627728939056396
    },
    {
      "auxiliary_loss_clip": 0.01095315,
      "auxiliary_loss_mlp": 0.01006133,
      "balance_loss_clip": 1.03529024,
      "balance_loss_mlp": 1.00414217,
      "epoch": 0.23699873744964828,
      "flos": 69859291875840.0,
      "grad_norm": 0.8268789166211841,
      "language_loss": 0.57798064,
      "learning_rate": 3.5672643300396214e-06,
      "loss": 0.59899509,
      "num_input_tokens_seen": 42053715,
      "step": 1971,
      "time_per_iteration": 3.4624624252319336
    },
    {
      "auxiliary_loss_clip": 0.01167828,
      "auxiliary_loss_mlp": 0.01032423,
      "balance_loss_clip": 1.05527377,
      "balance_loss_mlp": 1.02261806,
      "epoch": 0.2371189803402874,
      "flos": 21835052720640.0,
      "grad_norm": 2.987003397972215,
      "language_loss": 0.67722964,
      "learning_rate": 3.566780294432802e-06,
      "loss": 0.6992321,
      "num_input_tokens_seen": 42070890,
      "step": 1972,
      "time_per_iteration": 2.728698968887329
    },
    {
      "auxiliary_loss_clip": 0.01211538,
      "auxiliary_loss_mlp": 0.01040336,
      "balance_loss_clip": 1.05868077,
      "balance_loss_mlp": 1.03015542,
      "epoch": 0.23723922323092647,
      "flos": 21908490076800.0,
      "grad_norm": 4.99851176282828,
      "language_loss": 0.75097215,
      "learning_rate": 3.566296021146657e-06,
      "loss": 0.77349079,
      "num_input_tokens_seen": 42090270,
      "step": 1973,
      "time_per_iteration": 2.6775996685028076
    },
    {
      "auxiliary_loss_clip": 0.01213711,
      "auxiliary_loss_mlp": 0.01033602,
      "balance_loss_clip": 1.06004465,
      "balance_loss_mlp": 1.02391672,
      "epoch": 0.23735946612156555,
      "flos": 32708803380480.0,
      "grad_norm": 2.0770141046460258,
      "language_loss": 0.73362607,
      "learning_rate": 3.565811510254652e-06,
      "loss": 0.75609922,
      "num_input_tokens_seen": 42111150,
      "step": 1974,
      "time_per_iteration": 2.7020983695983887
    },
    {
      "auxiliary_loss_clip": 0.01117449,
      "auxiliary_loss_mlp": 0.01002799,
      "balance_loss_clip": 1.04451942,
      "balance_loss_mlp": 1.0004859,
      "epoch": 0.23747970901220466,
      "flos": 70546944821760.0,
      "grad_norm": 0.8340068779108762,
      "language_loss": 0.58233798,
      "learning_rate": 3.5653267618302845e-06,
      "loss": 0.60354054,
      "num_input_tokens_seen": 42178730,
      "step": 1975,
      "time_per_iteration": 3.3298583030700684
    },
    {
      "auxiliary_loss_clip": 0.01210754,
      "auxiliary_loss_mlp": 0.01033942,
      "balance_loss_clip": 1.05854416,
      "balance_loss_mlp": 1.02411914,
      "epoch": 0.23759995190284375,
      "flos": 20849807594880.0,
      "grad_norm": 1.846372160079544,
      "language_loss": 0.85704225,
      "learning_rate": 3.564841775947093e-06,
      "loss": 0.87948918,
      "num_input_tokens_seen": 42199620,
      "step": 1976,
      "time_per_iteration": 2.6351826190948486
    },
    {
      "auxiliary_loss_clip": 0.0116454,
      "auxiliary_loss_mlp": 0.01037242,
      "balance_loss_clip": 1.05238295,
      "balance_loss_mlp": 1.02718019,
      "epoch": 0.23772019479348283,
      "flos": 32921645420160.0,
      "grad_norm": 2.7526769269689577,
      "language_loss": 0.76480389,
      "learning_rate": 3.5643565526786475e-06,
      "loss": 0.78682166,
      "num_input_tokens_seen": 42219560,
      "step": 1977,
      "time_per_iteration": 2.7980806827545166
    },
    {
      "auxiliary_loss_clip": 0.01214946,
      "auxiliary_loss_mlp": 0.01030868,
      "balance_loss_clip": 1.06131196,
      "balance_loss_mlp": 1.02121854,
      "epoch": 0.2378404376841219,
      "flos": 32342765834880.0,
      "grad_norm": 12.886705168617794,
      "language_loss": 0.77325833,
      "learning_rate": 3.5638710920985574e-06,
      "loss": 0.79571646,
      "num_input_tokens_seen": 42241020,
      "step": 1978,
      "time_per_iteration": 2.687873601913452
    },
    {
      "auxiliary_loss_clip": 0.01203027,
      "auxiliary_loss_mlp": 0.00765548,
      "balance_loss_clip": 1.0567832,
      "balance_loss_mlp": 1.00069046,
      "epoch": 0.23796068057476102,
      "flos": 22997624313600.0,
      "grad_norm": 2.840450634526938,
      "language_loss": 0.82379967,
      "learning_rate": 3.5633853942804655e-06,
      "loss": 0.84348541,
      "num_input_tokens_seen": 42259345,
      "step": 1979,
      "time_per_iteration": 2.655470371246338
    },
    {
      "auxiliary_loss_clip": 0.01164534,
      "auxiliary_loss_mlp": 0.01039653,
      "balance_loss_clip": 1.05068958,
      "balance_loss_mlp": 1.02867389,
      "epoch": 0.2380809234654001,
      "flos": 13480938414720.0,
      "grad_norm": 3.6781863135157122,
      "language_loss": 0.7705214,
      "learning_rate": 3.5628994592980527e-06,
      "loss": 0.7925632,
      "num_input_tokens_seen": 42277250,
      "step": 1980,
      "time_per_iteration": 2.6891067028045654
    },
    {
      "auxiliary_loss_clip": 0.01212986,
      "auxiliary_loss_mlp": 0.01035701,
      "balance_loss_clip": 1.05905545,
      "balance_loss_mlp": 1.02575898,
      "epoch": 0.2382011663560392,
      "flos": 16871803148160.0,
      "grad_norm": 1.7277145467076493,
      "language_loss": 0.70288521,
      "learning_rate": 3.562413287225034e-06,
      "loss": 0.72537202,
      "num_input_tokens_seen": 42295360,
      "step": 1981,
      "time_per_iteration": 2.556370735168457
    },
    {
      "auxiliary_loss_clip": 0.01193464,
      "auxiliary_loss_mlp": 0.01044769,
      "balance_loss_clip": 1.0596025,
      "balance_loss_mlp": 1.03451085,
      "epoch": 0.2383214092466783,
      "flos": 18441135331200.0,
      "grad_norm": 2.83827075708479,
      "language_loss": 0.89315999,
      "learning_rate": 3.5619268781351623e-06,
      "loss": 0.91554224,
      "num_input_tokens_seen": 42313430,
      "step": 1982,
      "time_per_iteration": 2.704265832901001
    },
    {
      "auxiliary_loss_clip": 0.01179016,
      "auxiliary_loss_mlp": 0.01037739,
      "balance_loss_clip": 1.05936408,
      "balance_loss_mlp": 1.02866709,
      "epoch": 0.23844165213731738,
      "flos": 19755717281280.0,
      "grad_norm": 2.009027121618499,
      "language_loss": 0.77016032,
      "learning_rate": 3.5614402321022256e-06,
      "loss": 0.79232788,
      "num_input_tokens_seen": 42331260,
      "step": 1983,
      "time_per_iteration": 2.645494222640991
    },
    {
      "auxiliary_loss_clip": 0.01142343,
      "auxiliary_loss_mlp": 0.01041973,
      "balance_loss_clip": 1.05187809,
      "balance_loss_mlp": 1.03243041,
      "epoch": 0.23856189502795647,
      "flos": 23367360960000.0,
      "grad_norm": 1.7686829448512968,
      "language_loss": 0.87083721,
      "learning_rate": 3.5609533492000463e-06,
      "loss": 0.89268041,
      "num_input_tokens_seen": 42350150,
      "step": 1984,
      "time_per_iteration": 2.8393704891204834
    },
    {
      "auxiliary_loss_clip": 0.01180976,
      "auxiliary_loss_mlp": 0.01036005,
      "balance_loss_clip": 1.05966246,
      "balance_loss_mlp": 1.02577066,
      "epoch": 0.23868213791859555,
      "flos": 23475056912640.0,
      "grad_norm": 2.3741227029999137,
      "language_loss": 0.78769839,
      "learning_rate": 3.560466229502485e-06,
      "loss": 0.80986816,
      "num_input_tokens_seen": 42369495,
      "step": 1985,
      "time_per_iteration": 2.8316457271575928
    },
    {
      "auxiliary_loss_clip": 0.01179492,
      "auxiliary_loss_mlp": 0.00765,
      "balance_loss_clip": 1.05901265,
      "balance_loss_mlp": 1.00072527,
      "epoch": 0.23880238080923466,
      "flos": 16617340224000.0,
      "grad_norm": 2.060615563519021,
      "language_loss": 0.90056872,
      "learning_rate": 3.5599788730834384e-06,
      "loss": 0.92001373,
      "num_input_tokens_seen": 42387455,
      "step": 1986,
      "time_per_iteration": 2.68450927734375
    },
    {
      "auxiliary_loss_clip": 0.01202622,
      "auxiliary_loss_mlp": 0.01036115,
      "balance_loss_clip": 1.0595938,
      "balance_loss_mlp": 1.02629769,
      "epoch": 0.23892262369987374,
      "flos": 17348409734400.0,
      "grad_norm": 6.983394046605042,
      "language_loss": 0.7906993,
      "learning_rate": 3.559491280016836e-06,
      "loss": 0.81308669,
      "num_input_tokens_seen": 42405400,
      "step": 1987,
      "time_per_iteration": 2.639719247817993
    },
    {
      "auxiliary_loss_clip": 0.01182768,
      "auxiliary_loss_mlp": 0.01036348,
      "balance_loss_clip": 1.05700672,
      "balance_loss_mlp": 1.02637661,
      "epoch": 0.23904286659051283,
      "flos": 22309899540480.0,
      "grad_norm": 1.7701738136566376,
      "language_loss": 0.71124113,
      "learning_rate": 3.5590034503766465e-06,
      "loss": 0.73343229,
      "num_input_tokens_seen": 42425065,
      "step": 1988,
      "time_per_iteration": 2.745944023132324
    },
    {
      "auxiliary_loss_clip": 0.01215618,
      "auxiliary_loss_mlp": 0.01042243,
      "balance_loss_clip": 1.06205511,
      "balance_loss_mlp": 1.03225315,
      "epoch": 0.23916310948115194,
      "flos": 21178246579200.0,
      "grad_norm": 2.2573178453710097,
      "language_loss": 0.81116229,
      "learning_rate": 3.558515384236874e-06,
      "loss": 0.83374095,
      "num_input_tokens_seen": 42442495,
      "step": 1989,
      "time_per_iteration": 2.6128361225128174
    },
    {
      "auxiliary_loss_clip": 0.01158033,
      "auxiliary_loss_mlp": 0.00765471,
      "balance_loss_clip": 1.05472374,
      "balance_loss_mlp": 1.00064957,
      "epoch": 0.23928335237179102,
      "flos": 14137349506560.0,
      "grad_norm": 1.9087884241875184,
      "language_loss": 0.84081233,
      "learning_rate": 3.558027081671556e-06,
      "loss": 0.86004734,
      "num_input_tokens_seen": 42459480,
      "step": 1990,
      "time_per_iteration": 2.7520062923431396
    },
    {
      "auxiliary_loss_clip": 0.012036,
      "auxiliary_loss_mlp": 0.0103712,
      "balance_loss_clip": 1.05837548,
      "balance_loss_mlp": 1.02696395,
      "epoch": 0.2394035952624301,
      "flos": 23769596436480.0,
      "grad_norm": 3.171122643729757,
      "language_loss": 0.68608105,
      "learning_rate": 3.557538542754769e-06,
      "loss": 0.70848823,
      "num_input_tokens_seen": 42479175,
      "step": 1991,
      "time_per_iteration": 3.6641550064086914
    },
    {
      "auxiliary_loss_clip": 0.01215343,
      "auxiliary_loss_mlp": 0.01042179,
      "balance_loss_clip": 1.06241548,
      "balance_loss_mlp": 1.03102136,
      "epoch": 0.2395238381530692,
      "flos": 24206198250240.0,
      "grad_norm": 2.200211060809397,
      "language_loss": 0.67143953,
      "learning_rate": 3.557049767560623e-06,
      "loss": 0.69401479,
      "num_input_tokens_seen": 42498090,
      "step": 1992,
      "time_per_iteration": 2.6709232330322266
    },
    {
      "auxiliary_loss_clip": 0.01159363,
      "auxiliary_loss_mlp": 0.01032797,
      "balance_loss_clip": 1.05816722,
      "balance_loss_mlp": 1.02317691,
      "epoch": 0.2396440810437083,
      "flos": 25295763450240.0,
      "grad_norm": 2.1047936524406317,
      "language_loss": 0.86007529,
      "learning_rate": 3.5565607561632655e-06,
      "loss": 0.88199693,
      "num_input_tokens_seen": 42516930,
      "step": 1993,
      "time_per_iteration": 2.8814609050750732
    },
    {
      "auxiliary_loss_clip": 0.01179853,
      "auxiliary_loss_mlp": 0.0103404,
      "balance_loss_clip": 1.05660939,
      "balance_loss_mlp": 1.02337682,
      "epoch": 0.23976432393434738,
      "flos": 28543093436160.0,
      "grad_norm": 2.992230044159644,
      "language_loss": 0.79044586,
      "learning_rate": 3.5560715086368787e-06,
      "loss": 0.81258476,
      "num_input_tokens_seen": 42534800,
      "step": 1994,
      "time_per_iteration": 3.6999244689941406
    },
    {
      "auxiliary_loss_clip": 0.01178053,
      "auxiliary_loss_mlp": 0.01038751,
      "balance_loss_clip": 1.05819428,
      "balance_loss_mlp": 1.02849901,
      "epoch": 0.23988456682498646,
      "flos": 19494358945920.0,
      "grad_norm": 2.7324409232418683,
      "language_loss": 0.8253864,
      "learning_rate": 3.5555820250556816e-06,
      "loss": 0.84755445,
      "num_input_tokens_seen": 42552000,
      "step": 1995,
      "time_per_iteration": 3.614504337310791
    },
    {
      "auxiliary_loss_clip": 0.01190818,
      "auxiliary_loss_mlp": 0.01036306,
      "balance_loss_clip": 1.06128597,
      "balance_loss_mlp": 1.02568424,
      "epoch": 0.24000480971562557,
      "flos": 20266331068800.0,
      "grad_norm": 4.030747429049923,
      "language_loss": 0.69356394,
      "learning_rate": 3.5550923054939278e-06,
      "loss": 0.71583521,
      "num_input_tokens_seen": 42571455,
      "step": 1996,
      "time_per_iteration": 3.694998025894165
    },
    {
      "auxiliary_loss_clip": 0.01149292,
      "auxiliary_loss_mlp": 0.01037034,
      "balance_loss_clip": 1.05255103,
      "balance_loss_mlp": 1.02713358,
      "epoch": 0.24012505260626466,
      "flos": 25443176866560.0,
      "grad_norm": 2.251061970684212,
      "language_loss": 0.74747276,
      "learning_rate": 3.5546023500259083e-06,
      "loss": 0.76933599,
      "num_input_tokens_seen": 42592550,
      "step": 1997,
      "time_per_iteration": 2.8813302516937256
    },
    {
      "auxiliary_loss_clip": 0.01160887,
      "auxiliary_loss_mlp": 0.01033658,
      "balance_loss_clip": 1.05597234,
      "balance_loss_mlp": 1.02413344,
      "epoch": 0.24024529549690374,
      "flos": 15553342529280.0,
      "grad_norm": 2.2278629049654937,
      "language_loss": 0.81059968,
      "learning_rate": 3.5541121587259477e-06,
      "loss": 0.83254516,
      "num_input_tokens_seen": 42610385,
      "step": 1998,
      "time_per_iteration": 2.7959396839141846
    },
    {
      "auxiliary_loss_clip": 0.01105963,
      "auxiliary_loss_mlp": 0.01004751,
      "balance_loss_clip": 1.0379827,
      "balance_loss_mlp": 1.00273609,
      "epoch": 0.24036553838754285,
      "flos": 57122351867520.0,
      "grad_norm": 0.8483933557061812,
      "language_loss": 0.57813954,
      "learning_rate": 3.553621731668408e-06,
      "loss": 0.59924668,
      "num_input_tokens_seen": 42673595,
      "step": 1999,
      "time_per_iteration": 3.2449729442596436
    },
    {
      "auxiliary_loss_clip": 0.0119422,
      "auxiliary_loss_mlp": 0.01039136,
      "balance_loss_clip": 1.05755472,
      "balance_loss_mlp": 1.02872288,
      "epoch": 0.24048578127818193,
      "flos": 24969946158720.0,
      "grad_norm": 1.7383209746081076,
      "language_loss": 0.83310312,
      "learning_rate": 3.553131068927688e-06,
      "loss": 0.85543668,
      "num_input_tokens_seen": 42692000,
      "step": 2000,
      "time_per_iteration": 2.7133445739746094
    },
    {
      "auxiliary_loss_clip": 0.01170899,
      "auxiliary_loss_mlp": 0.0103501,
      "balance_loss_clip": 1.05794787,
      "balance_loss_mlp": 1.02474046,
      "epoch": 0.24060602416882101,
      "flos": 23330947547520.0,
      "grad_norm": 1.716602722049479,
      "language_loss": 0.80298728,
      "learning_rate": 3.552640170578219e-06,
      "loss": 0.82504636,
      "num_input_tokens_seen": 42712250,
      "step": 2001,
      "time_per_iteration": 2.8069820404052734
    },
    {
      "auxiliary_loss_clip": 0.0118589,
      "auxiliary_loss_mlp": 0.01037171,
      "balance_loss_clip": 1.05877948,
      "balance_loss_mlp": 1.02719927,
      "epoch": 0.2407262670594601,
      "flos": 14173260128640.0,
      "grad_norm": 2.3431572687297706,
      "language_loss": 0.78004354,
      "learning_rate": 3.5521490366944703e-06,
      "loss": 0.80227411,
      "num_input_tokens_seen": 42729900,
      "step": 2002,
      "time_per_iteration": 2.6220438480377197
    },
    {
      "auxiliary_loss_clip": 0.01168054,
      "auxiliary_loss_mlp": 0.01039688,
      "balance_loss_clip": 1.05651522,
      "balance_loss_mlp": 1.02975774,
      "epoch": 0.2408465099500992,
      "flos": 13663113217920.0,
      "grad_norm": 2.9677083367355737,
      "language_loss": 0.80454993,
      "learning_rate": 3.5516576673509474e-06,
      "loss": 0.82662731,
      "num_input_tokens_seen": 42747900,
      "step": 2003,
      "time_per_iteration": 2.7236740589141846
    },
    {
      "auxiliary_loss_clip": 0.01214283,
      "auxiliary_loss_mlp": 0.01037657,
      "balance_loss_clip": 1.06021154,
      "balance_loss_mlp": 1.02735734,
      "epoch": 0.2409667528407383,
      "flos": 31248029076480.0,
      "grad_norm": 1.9551561629891352,
      "language_loss": 0.85969341,
      "learning_rate": 3.5511660626221896e-06,
      "loss": 0.88221282,
      "num_input_tokens_seen": 42768540,
      "step": 2004,
      "time_per_iteration": 2.6839094161987305
    },
    {
      "auxiliary_loss_clip": 0.01183742,
      "auxiliary_loss_mlp": 0.00765447,
      "balance_loss_clip": 1.05851483,
      "balance_loss_mlp": 1.00070953,
      "epoch": 0.24108699573137737,
      "flos": 22199941031040.0,
      "grad_norm": 3.8103128675627143,
      "language_loss": 0.89175749,
      "learning_rate": 3.5506742225827744e-06,
      "loss": 0.9112494,
      "num_input_tokens_seen": 42785395,
      "step": 2005,
      "time_per_iteration": 2.658069372177124
    },
    {
      "auxiliary_loss_clip": 0.01165989,
      "auxiliary_loss_mlp": 0.01038791,
      "balance_loss_clip": 1.05478621,
      "balance_loss_mlp": 1.02816963,
      "epoch": 0.24120723862201648,
      "flos": 26103035664000.0,
      "grad_norm": 2.319572626270833,
      "language_loss": 0.90465385,
      "learning_rate": 3.5501821473073116e-06,
      "loss": 0.92670166,
      "num_input_tokens_seen": 42801980,
      "step": 2006,
      "time_per_iteration": 2.7185373306274414
    },
    {
      "auxiliary_loss_clip": 0.01166932,
      "auxiliary_loss_mlp": 0.01037401,
      "balance_loss_clip": 1.05856574,
      "balance_loss_mlp": 1.028234,
      "epoch": 0.24132748151265557,
      "flos": 18624926246400.0,
      "grad_norm": 2.5675914254246432,
      "language_loss": 0.86949396,
      "learning_rate": 3.54968983687045e-06,
      "loss": 0.89153731,
      "num_input_tokens_seen": 42818850,
      "step": 2007,
      "time_per_iteration": 2.6639606952667236
    },
    {
      "auxiliary_loss_clip": 0.01186992,
      "auxiliary_loss_mlp": 0.01034617,
      "balance_loss_clip": 1.05817759,
      "balance_loss_mlp": 1.02428782,
      "epoch": 0.24144772440329465,
      "flos": 15267673664640.0,
      "grad_norm": 5.599681791832032,
      "language_loss": 0.89493906,
      "learning_rate": 3.549197291346872e-06,
      "loss": 0.91715515,
      "num_input_tokens_seen": 42835375,
      "step": 2008,
      "time_per_iteration": 2.6706700325012207
    },
    {
      "auxiliary_loss_clip": 0.01200275,
      "auxiliary_loss_mlp": 0.0103439,
      "balance_loss_clip": 1.0591352,
      "balance_loss_mlp": 1.02411413,
      "epoch": 0.24156796729393373,
      "flos": 24024274842240.0,
      "grad_norm": 2.4559462413325366,
      "language_loss": 0.79653823,
      "learning_rate": 3.548704510811297e-06,
      "loss": 0.81888485,
      "num_input_tokens_seen": 42854570,
      "step": 2009,
      "time_per_iteration": 2.6534206867218018
    },
    {
      "auxiliary_loss_clip": 0.01157506,
      "auxiliary_loss_mlp": 0.01030234,
      "balance_loss_clip": 1.05218899,
      "balance_loss_mlp": 1.0199635,
      "epoch": 0.24168821018457284,
      "flos": 26286790665600.0,
      "grad_norm": 3.9723532804591235,
      "language_loss": 0.74644762,
      "learning_rate": 3.5482114953384787e-06,
      "loss": 0.76832497,
      "num_input_tokens_seen": 42873800,
      "step": 2010,
      "time_per_iteration": 2.7932276725769043
    },
    {
      "auxiliary_loss_clip": 0.0120553,
      "auxiliary_loss_mlp": 0.01039524,
      "balance_loss_clip": 1.06070685,
      "balance_loss_mlp": 1.02986801,
      "epoch": 0.24180845307521193,
      "flos": 18223193560320.0,
      "grad_norm": 2.115904867805552,
      "language_loss": 0.84445441,
      "learning_rate": 3.5477182450032077e-06,
      "loss": 0.86690491,
      "num_input_tokens_seen": 42892400,
      "step": 2011,
      "time_per_iteration": 2.6598806381225586
    },
    {
      "auxiliary_loss_clip": 0.01198819,
      "auxiliary_loss_mlp": 0.01039634,
      "balance_loss_clip": 1.06034636,
      "balance_loss_mlp": 1.02917361,
      "epoch": 0.241928695965851,
      "flos": 20449260057600.0,
      "grad_norm": 2.2011405519957794,
      "language_loss": 0.83051968,
      "learning_rate": 3.5472247598803097e-06,
      "loss": 0.8529042,
      "num_input_tokens_seen": 42911745,
      "step": 2012,
      "time_per_iteration": 2.73429012298584
    },
    {
      "auxiliary_loss_clip": 0.01212486,
      "auxiliary_loss_mlp": 0.01034037,
      "balance_loss_clip": 1.05932772,
      "balance_loss_mlp": 1.0235827,
      "epoch": 0.24204893885649012,
      "flos": 25556475340800.0,
      "grad_norm": 14.414151707812543,
      "language_loss": 0.85294187,
      "learning_rate": 3.546731040044645e-06,
      "loss": 0.87540716,
      "num_input_tokens_seen": 42926915,
      "step": 2013,
      "time_per_iteration": 2.6350269317626953
    },
    {
      "auxiliary_loss_clip": 0.01214582,
      "auxiliary_loss_mlp": 0.01040463,
      "balance_loss_clip": 1.06210208,
      "balance_loss_mlp": 1.03022909,
      "epoch": 0.2421691817471292,
      "flos": 30660207004800.0,
      "grad_norm": 1.9019973223995599,
      "language_loss": 0.75359774,
      "learning_rate": 3.546237085571112e-06,
      "loss": 0.77614814,
      "num_input_tokens_seen": 42945350,
      "step": 2014,
      "time_per_iteration": 2.7070388793945312
    },
    {
      "auxiliary_loss_clip": 0.01200731,
      "auxiliary_loss_mlp": 0.01040306,
      "balance_loss_clip": 1.05989289,
      "balance_loss_mlp": 1.03019106,
      "epoch": 0.24228942463776829,
      "flos": 21945011230080.0,
      "grad_norm": 2.1927855688270417,
      "language_loss": 0.72535086,
      "learning_rate": 3.5457428965346425e-06,
      "loss": 0.74776125,
      "num_input_tokens_seen": 42964290,
      "step": 2015,
      "time_per_iteration": 2.707904815673828
    },
    {
      "auxiliary_loss_clip": 0.01139492,
      "auxiliary_loss_mlp": 0.01038266,
      "balance_loss_clip": 1.05338597,
      "balance_loss_mlp": 1.02840114,
      "epoch": 0.2424096675284074,
      "flos": 33984493879680.0,
      "grad_norm": 1.738795974554511,
      "language_loss": 0.74727052,
      "learning_rate": 3.545248473010205e-06,
      "loss": 0.76904809,
      "num_input_tokens_seen": 42987095,
      "step": 2016,
      "time_per_iteration": 3.775074005126953
    },
    {
      "auxiliary_loss_clip": 0.01217829,
      "auxiliary_loss_mlp": 0.00765687,
      "balance_loss_clip": 1.06156278,
      "balance_loss_mlp": 1.00059283,
      "epoch": 0.24252991041904648,
      "flos": 21653416621440.0,
      "grad_norm": 1.8967531969010005,
      "language_loss": 0.87898171,
      "learning_rate": 3.544753815072802e-06,
      "loss": 0.89881694,
      "num_input_tokens_seen": 43005750,
      "step": 2017,
      "time_per_iteration": 2.653320550918579
    },
    {
      "auxiliary_loss_clip": 0.01114328,
      "auxiliary_loss_mlp": 0.0103577,
      "balance_loss_clip": 1.04510069,
      "balance_loss_mlp": 1.02597117,
      "epoch": 0.24265015330968556,
      "flos": 21870065502720.0,
      "grad_norm": 2.1156823808454788,
      "language_loss": 0.88167858,
      "learning_rate": 3.544258922797474e-06,
      "loss": 0.90317953,
      "num_input_tokens_seen": 43023870,
      "step": 2018,
      "time_per_iteration": 2.762349843978882
    },
    {
      "auxiliary_loss_clip": 0.01213472,
      "auxiliary_loss_mlp": 0.01035509,
      "balance_loss_clip": 1.06151104,
      "balance_loss_mlp": 1.02542341,
      "epoch": 0.24277039620032465,
      "flos": 25628260671360.0,
      "grad_norm": 2.1663152406064627,
      "language_loss": 0.78361022,
      "learning_rate": 3.543763796259295e-06,
      "loss": 0.80610001,
      "num_input_tokens_seen": 43043825,
      "step": 2019,
      "time_per_iteration": 2.674020290374756
    },
    {
      "auxiliary_loss_clip": 0.01198249,
      "auxiliary_loss_mlp": 0.01035427,
      "balance_loss_clip": 1.05914366,
      "balance_loss_mlp": 1.02547884,
      "epoch": 0.24289063909096376,
      "flos": 26286575184000.0,
      "grad_norm": 1.8976977203169956,
      "language_loss": 0.91214162,
      "learning_rate": 3.5432684355333754e-06,
      "loss": 0.93447834,
      "num_input_tokens_seen": 43062480,
      "step": 2020,
      "time_per_iteration": 3.630615711212158
    },
    {
      "auxiliary_loss_clip": 0.01197658,
      "auxiliary_loss_mlp": 0.01032436,
      "balance_loss_clip": 1.05699801,
      "balance_loss_mlp": 1.02187455,
      "epoch": 0.24301088198160284,
      "flos": 25075056332160.0,
      "grad_norm": 2.459321933808246,
      "language_loss": 0.76646554,
      "learning_rate": 3.5427728406948613e-06,
      "loss": 0.78876644,
      "num_input_tokens_seen": 43081595,
      "step": 2021,
      "time_per_iteration": 3.604193925857544
    },
    {
      "auxiliary_loss_clip": 0.01089574,
      "auxiliary_loss_mlp": 0.01002889,
      "balance_loss_clip": 1.02868152,
      "balance_loss_mlp": 1.00080287,
      "epoch": 0.24313112487224192,
      "flos": 69900948673920.0,
      "grad_norm": 0.7958063599227025,
      "language_loss": 0.57874006,
      "learning_rate": 3.542277011818934e-06,
      "loss": 0.59966475,
      "num_input_tokens_seen": 43145430,
      "step": 2022,
      "time_per_iteration": 4.293999195098877
    },
    {
      "auxiliary_loss_clip": 0.01183391,
      "auxiliary_loss_mlp": 0.01031875,
      "balance_loss_clip": 1.05768824,
      "balance_loss_mlp": 1.02304184,
      "epoch": 0.24325136776288103,
      "flos": 40662334235520.0,
      "grad_norm": 2.38061674499193,
      "language_loss": 0.74062324,
      "learning_rate": 3.5417809489808104e-06,
      "loss": 0.76277596,
      "num_input_tokens_seen": 43167040,
      "step": 2023,
      "time_per_iteration": 2.8267881870269775
    },
    {
      "auxiliary_loss_clip": 0.01202115,
      "auxiliary_loss_mlp": 0.01037558,
      "balance_loss_clip": 1.06207561,
      "balance_loss_mlp": 1.02760983,
      "epoch": 0.24337161065352012,
      "flos": 25046400257280.0,
      "grad_norm": 1.8290291536382646,
      "language_loss": 0.72434151,
      "learning_rate": 3.5412846522557422e-06,
      "loss": 0.74673826,
      "num_input_tokens_seen": 43187930,
      "step": 2024,
      "time_per_iteration": 2.7770321369171143
    },
    {
      "auxiliary_loss_clip": 0.01215433,
      "auxiliary_loss_mlp": 0.01038503,
      "balance_loss_clip": 1.06268084,
      "balance_loss_mlp": 1.02860844,
      "epoch": 0.2434918535441592,
      "flos": 18661160090880.0,
      "grad_norm": 3.5759426026039076,
      "language_loss": 0.74037898,
      "learning_rate": 3.540788121719018e-06,
      "loss": 0.76291829,
      "num_input_tokens_seen": 43206350,
      "step": 2025,
      "time_per_iteration": 2.6739232540130615
    },
    {
      "auxiliary_loss_clip": 0.01162018,
      "auxiliary_loss_mlp": 0.01029825,
      "balance_loss_clip": 1.05833817,
      "balance_loss_mlp": 1.02040708,
      "epoch": 0.24361209643479828,
      "flos": 23915142345600.0,
      "grad_norm": 1.9525908821329123,
      "language_loss": 0.81684649,
      "learning_rate": 3.5402913574459604e-06,
      "loss": 0.83876497,
      "num_input_tokens_seen": 43226255,
      "step": 2026,
      "time_per_iteration": 2.8382441997528076
    },
    {
      "auxiliary_loss_clip": 0.01134507,
      "auxiliary_loss_mlp": 0.01032234,
      "balance_loss_clip": 1.04900861,
      "balance_loss_mlp": 1.02280998,
      "epoch": 0.2437323393254374,
      "flos": 28657505232000.0,
      "grad_norm": 2.1791462552988636,
      "language_loss": 0.85920799,
      "learning_rate": 3.5397943595119297e-06,
      "loss": 0.88087535,
      "num_input_tokens_seen": 43247675,
      "step": 2027,
      "time_per_iteration": 3.0320804119110107
    },
    {
      "auxiliary_loss_clip": 0.01181685,
      "auxiliary_loss_mlp": 0.01036022,
      "balance_loss_clip": 1.06036174,
      "balance_loss_mlp": 1.02629483,
      "epoch": 0.24385258221607647,
      "flos": 23550325862400.0,
      "grad_norm": 4.475027494552107,
      "language_loss": 0.77100861,
      "learning_rate": 3.5392971279923177e-06,
      "loss": 0.79318571,
      "num_input_tokens_seen": 43265895,
      "step": 2028,
      "time_per_iteration": 2.90582537651062
    },
    {
      "auxiliary_loss_clip": 0.01161985,
      "auxiliary_loss_mlp": 0.01038543,
      "balance_loss_clip": 1.05227256,
      "balance_loss_mlp": 1.0284282,
      "epoch": 0.24397282510671556,
      "flos": 25336091445120.0,
      "grad_norm": 2.735913459405839,
      "language_loss": 0.83022892,
      "learning_rate": 3.5387996629625557e-06,
      "loss": 0.85223418,
      "num_input_tokens_seen": 43283485,
      "step": 2029,
      "time_per_iteration": 2.7067718505859375
    },
    {
      "auxiliary_loss_clip": 0.01115913,
      "auxiliary_loss_mlp": 0.01003068,
      "balance_loss_clip": 1.03254104,
      "balance_loss_mlp": 1.0011605,
      "epoch": 0.24409306799735467,
      "flos": 65187421430400.0,
      "grad_norm": 0.8116011816495262,
      "language_loss": 0.54968417,
      "learning_rate": 3.5383019644981083e-06,
      "loss": 0.57087398,
      "num_input_tokens_seen": 43347180,
      "step": 2030,
      "time_per_iteration": 3.1867733001708984
    },
    {
      "auxiliary_loss_clip": 0.0118372,
      "auxiliary_loss_mlp": 0.01037124,
      "balance_loss_clip": 1.05943441,
      "balance_loss_mlp": 1.02673507,
      "epoch": 0.24421331088799375,
      "flos": 19537093152000.0,
      "grad_norm": 2.570066423171972,
      "language_loss": 0.73133808,
      "learning_rate": 3.5378040326744763e-06,
      "loss": 0.75354648,
      "num_input_tokens_seen": 43366665,
      "step": 2031,
      "time_per_iteration": 2.6384527683258057
    },
    {
      "auxiliary_loss_clip": 0.01168581,
      "auxiliary_loss_mlp": 0.01036645,
      "balance_loss_clip": 1.06007528,
      "balance_loss_mlp": 1.02700162,
      "epoch": 0.24433355377863283,
      "flos": 21068575378560.0,
      "grad_norm": 2.2096005779334797,
      "language_loss": 0.857867,
      "learning_rate": 3.5373058675671946e-06,
      "loss": 0.87991923,
      "num_input_tokens_seen": 43384670,
      "step": 2032,
      "time_per_iteration": 2.7296409606933594
    },
    {
      "auxiliary_loss_clip": 0.01144473,
      "auxiliary_loss_mlp": 0.0103767,
      "balance_loss_clip": 1.05142188,
      "balance_loss_mlp": 1.02750754,
      "epoch": 0.24445379666927192,
      "flos": 22637189289600.0,
      "grad_norm": 2.514126869431076,
      "language_loss": 0.72859848,
      "learning_rate": 3.536807469251836e-06,
      "loss": 0.75041997,
      "num_input_tokens_seen": 43403825,
      "step": 2033,
      "time_per_iteration": 2.7600364685058594
    },
    {
      "auxiliary_loss_clip": 0.01172068,
      "auxiliary_loss_mlp": 0.01037039,
      "balance_loss_clip": 1.05254388,
      "balance_loss_mlp": 1.02722239,
      "epoch": 0.24457403955991103,
      "flos": 21251612108160.0,
      "grad_norm": 3.079338675454707,
      "language_loss": 0.82810831,
      "learning_rate": 3.5363088378040055e-06,
      "loss": 0.85019946,
      "num_input_tokens_seen": 43422715,
      "step": 2034,
      "time_per_iteration": 2.8573055267333984
    },
    {
      "auxiliary_loss_clip": 0.01110879,
      "auxiliary_loss_mlp": 0.00755255,
      "balance_loss_clip": 1.02828383,
      "balance_loss_mlp": 1.00055492,
      "epoch": 0.2446942824505501,
      "flos": 66997820764800.0,
      "grad_norm": 0.7591812012724692,
      "language_loss": 0.64284837,
      "learning_rate": 3.5358099732993463e-06,
      "loss": 0.66150975,
      "num_input_tokens_seen": 43481825,
      "step": 2035,
      "time_per_iteration": 3.114967107772827
    },
    {
      "auxiliary_loss_clip": 0.01191251,
      "auxiliary_loss_mlp": 0.01038105,
      "balance_loss_clip": 1.05793977,
      "balance_loss_mlp": 1.02794862,
      "epoch": 0.2448145253411892,
      "flos": 20411122792320.0,
      "grad_norm": 2.125605031312977,
      "language_loss": 0.89667505,
      "learning_rate": 3.535310875813535e-06,
      "loss": 0.91896862,
      "num_input_tokens_seen": 43500220,
      "step": 2036,
      "time_per_iteration": 2.6819303035736084
    },
    {
      "auxiliary_loss_clip": 0.01196903,
      "auxiliary_loss_mlp": 0.01039999,
      "balance_loss_clip": 1.05845189,
      "balance_loss_mlp": 1.0304029,
      "epoch": 0.2449347682318283,
      "flos": 28804739080320.0,
      "grad_norm": 1.8293719527655232,
      "language_loss": 0.81798041,
      "learning_rate": 3.5348115454222843e-06,
      "loss": 0.84034944,
      "num_input_tokens_seen": 43522805,
      "step": 2037,
      "time_per_iteration": 2.683457851409912
    },
    {
      "auxiliary_loss_clip": 0.01176982,
      "auxiliary_loss_mlp": 0.01040947,
      "balance_loss_clip": 1.05285549,
      "balance_loss_mlp": 1.03039694,
      "epoch": 0.2450550111224674,
      "flos": 22528990546560.0,
      "grad_norm": 2.1568862054604265,
      "language_loss": 0.86730862,
      "learning_rate": 3.5343119822013425e-06,
      "loss": 0.88948798,
      "num_input_tokens_seen": 43541915,
      "step": 2038,
      "time_per_iteration": 2.6563141345977783
    },
    {
      "auxiliary_loss_clip": 0.01207191,
      "auxiliary_loss_mlp": 0.01041554,
      "balance_loss_clip": 1.06160426,
      "balance_loss_mlp": 1.03167772,
      "epoch": 0.24517525401310647,
      "flos": 21759137326080.0,
      "grad_norm": 1.9221797844441748,
      "language_loss": 0.77858287,
      "learning_rate": 3.533812186226493e-06,
      "loss": 0.80107033,
      "num_input_tokens_seen": 43562625,
      "step": 2039,
      "time_per_iteration": 2.7181053161621094
    },
    {
      "auxiliary_loss_clip": 0.01208335,
      "auxiliary_loss_mlp": 0.01033017,
      "balance_loss_clip": 1.05885792,
      "balance_loss_mlp": 1.02349186,
      "epoch": 0.24529549690374555,
      "flos": 25043311687680.0,
      "grad_norm": 1.6905698385586518,
      "language_loss": 0.75919688,
      "learning_rate": 3.5333121575735545e-06,
      "loss": 0.78161037,
      "num_input_tokens_seen": 43582265,
      "step": 2040,
      "time_per_iteration": 2.6123206615448
    },
    {
      "auxiliary_loss_clip": 0.01181923,
      "auxiliary_loss_mlp": 0.01032315,
      "balance_loss_clip": 1.05770659,
      "balance_loss_mlp": 1.02264762,
      "epoch": 0.24541573979438466,
      "flos": 32123638915200.0,
      "grad_norm": 2.651263188535779,
      "language_loss": 0.75390327,
      "learning_rate": 3.532811896318381e-06,
      "loss": 0.77604562,
      "num_input_tokens_seen": 43604335,
      "step": 2041,
      "time_per_iteration": 2.7912018299102783
    },
    {
      "auxiliary_loss_clip": 0.01173154,
      "auxiliary_loss_mlp": 0.01033091,
      "balance_loss_clip": 1.05653989,
      "balance_loss_mlp": 1.02288628,
      "epoch": 0.24553598268502375,
      "flos": 31357556622720.0,
      "grad_norm": 13.075809416925239,
      "language_loss": 0.82063532,
      "learning_rate": 3.5323114025368615e-06,
      "loss": 0.84269774,
      "num_input_tokens_seen": 43619400,
      "step": 2042,
      "time_per_iteration": 3.7641818523406982
    },
    {
      "auxiliary_loss_clip": 0.01190509,
      "auxiliary_loss_mlp": 0.01040051,
      "balance_loss_clip": 1.05386281,
      "balance_loss_mlp": 1.03016281,
      "epoch": 0.24565622557566283,
      "flos": 14027462824320.0,
      "grad_norm": 6.431210233954517,
      "language_loss": 0.82230502,
      "learning_rate": 3.53181067630492e-06,
      "loss": 0.84461063,
      "num_input_tokens_seen": 43636870,
      "step": 2043,
      "time_per_iteration": 2.6107115745544434
    },
    {
      "auxiliary_loss_clip": 0.01171942,
      "auxiliary_loss_mlp": 0.01041521,
      "balance_loss_clip": 1.05349135,
      "balance_loss_mlp": 1.03201985,
      "epoch": 0.24577646846630194,
      "flos": 16581465515520.0,
      "grad_norm": 1.8556758533871944,
      "language_loss": 0.76223731,
      "learning_rate": 3.5313097176985175e-06,
      "loss": 0.78437191,
      "num_input_tokens_seen": 43655180,
      "step": 2044,
      "time_per_iteration": 2.7360823154449463
    },
    {
      "auxiliary_loss_clip": 0.01197874,
      "auxiliary_loss_mlp": 0.01038578,
      "balance_loss_clip": 1.05821276,
      "balance_loss_mlp": 1.02924967,
      "epoch": 0.24589671135694102,
      "flos": 18807424272000.0,
      "grad_norm": 1.8388424858560708,
      "language_loss": 0.81585276,
      "learning_rate": 3.5308085267936482e-06,
      "loss": 0.83821726,
      "num_input_tokens_seen": 43672895,
      "step": 2045,
      "time_per_iteration": 2.6189157962799072
    },
    {
      "auxiliary_loss_clip": 0.01138885,
      "auxiliary_loss_mlp": 0.00765004,
      "balance_loss_clip": 1.05370128,
      "balance_loss_mlp": 1.0005424,
      "epoch": 0.2460169542475801,
      "flos": 19938538529280.0,
      "grad_norm": 1.8155050058829276,
      "language_loss": 0.89688802,
      "learning_rate": 3.530307103666342e-06,
      "loss": 0.91592687,
      "num_input_tokens_seen": 43691975,
      "step": 2046,
      "time_per_iteration": 3.8107922077178955
    },
    {
      "auxiliary_loss_clip": 0.01174868,
      "auxiliary_loss_mlp": 0.01036657,
      "balance_loss_clip": 1.057127,
      "balance_loss_mlp": 1.02672052,
      "epoch": 0.24613719713821922,
      "flos": 24171221381760.0,
      "grad_norm": 4.107715126636233,
      "language_loss": 0.80340886,
      "learning_rate": 3.5298054483926658e-06,
      "loss": 0.82552409,
      "num_input_tokens_seen": 43712670,
      "step": 2047,
      "time_per_iteration": 3.0098817348480225
    },
    {
      "auxiliary_loss_clip": 0.01205234,
      "auxiliary_loss_mlp": 0.01038292,
      "balance_loss_clip": 1.06002212,
      "balance_loss_mlp": 1.02753973,
      "epoch": 0.2462574400288583,
      "flos": 30221055325440.0,
      "grad_norm": 2.5493713921016283,
      "language_loss": 0.82666504,
      "learning_rate": 3.5293035610487187e-06,
      "loss": 0.84910029,
      "num_input_tokens_seen": 43732035,
      "step": 2048,
      "time_per_iteration": 4.182825326919556
    },
    {
      "auxiliary_loss_clip": 0.01083281,
      "auxiliary_loss_mlp": 0.01008407,
      "balance_loss_clip": 1.02531004,
      "balance_loss_mlp": 1.00627315,
      "epoch": 0.24637768291949738,
      "flos": 68943030819840.0,
      "grad_norm": 0.7239360895710083,
      "language_loss": 0.61987591,
      "learning_rate": 3.5288014417106374e-06,
      "loss": 0.64079273,
      "num_input_tokens_seen": 43798055,
      "step": 2049,
      "time_per_iteration": 3.337618827819824
    },
    {
      "auxiliary_loss_clip": 0.01164081,
      "auxiliary_loss_mlp": 0.01032834,
      "balance_loss_clip": 1.05405986,
      "balance_loss_mlp": 1.02296972,
      "epoch": 0.24649792581013646,
      "flos": 34383999922560.0,
      "grad_norm": 2.3595246376297543,
      "language_loss": 0.75594622,
      "learning_rate": 3.528299090454593e-06,
      "loss": 0.77791536,
      "num_input_tokens_seen": 43818590,
      "step": 2050,
      "time_per_iteration": 2.8811540603637695
    },
    {
      "auxiliary_loss_clip": 0.01202195,
      "auxiliary_loss_mlp": 0.0103596,
      "balance_loss_clip": 1.06064606,
      "balance_loss_mlp": 1.02532089,
      "epoch": 0.24661816870077558,
      "flos": 19680448331520.0,
      "grad_norm": 2.390247633135295,
      "language_loss": 0.82734758,
      "learning_rate": 3.527796507356792e-06,
      "loss": 0.84972918,
      "num_input_tokens_seen": 43832480,
      "step": 2051,
      "time_per_iteration": 2.7242166996002197
    },
    {
      "auxiliary_loss_clip": 0.01201358,
      "auxiliary_loss_mlp": 0.01037948,
      "balance_loss_clip": 1.05755496,
      "balance_loss_mlp": 1.02820277,
      "epoch": 0.24673841159141466,
      "flos": 20002279213440.0,
      "grad_norm": 2.7180965374643273,
      "language_loss": 0.89651322,
      "learning_rate": 3.527293692493475e-06,
      "loss": 0.91890627,
      "num_input_tokens_seen": 43848345,
      "step": 2052,
      "time_per_iteration": 2.739546060562134
    },
    {
      "auxiliary_loss_clip": 0.01202183,
      "auxiliary_loss_mlp": 0.01040646,
      "balance_loss_clip": 1.0584991,
      "balance_loss_mlp": 1.02997673,
      "epoch": 0.24685865448205374,
      "flos": 21646593037440.0,
      "grad_norm": 3.0559759164521534,
      "language_loss": 0.73077273,
      "learning_rate": 3.52679064594092e-06,
      "loss": 0.75320101,
      "num_input_tokens_seen": 43865685,
      "step": 2053,
      "time_per_iteration": 2.742281675338745
    },
    {
      "auxiliary_loss_clip": 0.01133657,
      "auxiliary_loss_mlp": 0.01030971,
      "balance_loss_clip": 1.04106259,
      "balance_loss_mlp": 1.02148771,
      "epoch": 0.24697889737269285,
      "flos": 17960470508160.0,
      "grad_norm": 3.394304796272737,
      "language_loss": 0.74966234,
      "learning_rate": 3.5262873677754375e-06,
      "loss": 0.77130866,
      "num_input_tokens_seen": 43883690,
      "step": 2054,
      "time_per_iteration": 2.890406370162964
    },
    {
      "auxiliary_loss_clip": 0.01210177,
      "auxiliary_loss_mlp": 0.01035012,
      "balance_loss_clip": 1.0610702,
      "balance_loss_mlp": 1.02555227,
      "epoch": 0.24709914026333193,
      "flos": 27344611221120.0,
      "grad_norm": 1.7321416245176058,
      "language_loss": 0.80612814,
      "learning_rate": 3.5257838580733745e-06,
      "loss": 0.82858002,
      "num_input_tokens_seen": 43903295,
      "step": 2055,
      "time_per_iteration": 2.8042657375335693
    },
    {
      "auxiliary_loss_clip": 0.01200856,
      "auxiliary_loss_mlp": 0.01033341,
      "balance_loss_clip": 1.05878901,
      "balance_loss_mlp": 1.02304721,
      "epoch": 0.24721938315397102,
      "flos": 19275519335040.0,
      "grad_norm": 2.1448791513439796,
      "language_loss": 0.87206191,
      "learning_rate": 3.5252801169111138e-06,
      "loss": 0.89440387,
      "num_input_tokens_seen": 43920960,
      "step": 2056,
      "time_per_iteration": 2.7577953338623047
    },
    {
      "auxiliary_loss_clip": 0.01181285,
      "auxiliary_loss_mlp": 0.01039343,
      "balance_loss_clip": 1.05972767,
      "balance_loss_mlp": 1.02974069,
      "epoch": 0.2473396260446101,
      "flos": 23185796688000.0,
      "grad_norm": 1.8257254478481604,
      "language_loss": 0.79878592,
      "learning_rate": 3.524776144365072e-06,
      "loss": 0.82099223,
      "num_input_tokens_seen": 43939415,
      "step": 2057,
      "time_per_iteration": 2.9318113327026367
    },
    {
      "auxiliary_loss_clip": 0.01179008,
      "auxiliary_loss_mlp": 0.01045851,
      "balance_loss_clip": 1.06057215,
      "balance_loss_mlp": 1.03620148,
      "epoch": 0.2474598689352492,
      "flos": 21142443697920.0,
      "grad_norm": 1.7296872760580633,
      "language_loss": 0.79152167,
      "learning_rate": 3.5242719405117016e-06,
      "loss": 0.81377023,
      "num_input_tokens_seen": 43959220,
      "step": 2058,
      "time_per_iteration": 2.8375179767608643
    },
    {
      "auxiliary_loss_clip": 0.01185682,
      "auxiliary_loss_mlp": 0.0076495,
      "balance_loss_clip": 1.05809426,
      "balance_loss_mlp": 1.00054789,
      "epoch": 0.2475801118258883,
      "flos": 21648352803840.0,
      "grad_norm": 2.5084681589595856,
      "language_loss": 0.74840117,
      "learning_rate": 3.5237675054274893e-06,
      "loss": 0.76790744,
      "num_input_tokens_seen": 43978420,
      "step": 2059,
      "time_per_iteration": 2.824300765991211
    },
    {
      "auxiliary_loss_clip": 0.01195481,
      "auxiliary_loss_mlp": 0.01032347,
      "balance_loss_clip": 1.05616498,
      "balance_loss_mlp": 1.02267289,
      "epoch": 0.24770035471652738,
      "flos": 22674500542080.0,
      "grad_norm": 1.8879911199354242,
      "language_loss": 0.80312002,
      "learning_rate": 3.5232628391889584e-06,
      "loss": 0.82539827,
      "num_input_tokens_seen": 43996710,
      "step": 2060,
      "time_per_iteration": 2.7522695064544678
    },
    {
      "auxiliary_loss_clip": 0.01150928,
      "auxiliary_loss_mlp": 0.01032884,
      "balance_loss_clip": 1.05439234,
      "balance_loss_mlp": 1.02371728,
      "epoch": 0.2478205976071665,
      "flos": 22163814927360.0,
      "grad_norm": 3.7823550577878327,
      "language_loss": 0.64433658,
      "learning_rate": 3.522757941872666e-06,
      "loss": 0.66617477,
      "num_input_tokens_seen": 44014865,
      "step": 2061,
      "time_per_iteration": 2.836489677429199
    },
    {
      "auxiliary_loss_clip": 0.01216368,
      "auxiliary_loss_mlp": 0.00763843,
      "balance_loss_clip": 1.06544161,
      "balance_loss_mlp": 1.0004611,
      "epoch": 0.24794084049780557,
      "flos": 24973106555520.0,
      "grad_norm": 1.9213396395174953,
      "language_loss": 0.82350236,
      "learning_rate": 3.5222528135552042e-06,
      "loss": 0.8433044,
      "num_input_tokens_seen": 44036325,
      "step": 2062,
      "time_per_iteration": 2.7551403045654297
    },
    {
      "auxiliary_loss_clip": 0.01196508,
      "auxiliary_loss_mlp": 0.01033459,
      "balance_loss_clip": 1.06041312,
      "balance_loss_mlp": 1.02456605,
      "epoch": 0.24806108338844465,
      "flos": 18296379521280.0,
      "grad_norm": 2.6364570522645643,
      "language_loss": 0.80666083,
      "learning_rate": 3.521747454313201e-06,
      "loss": 0.82896054,
      "num_input_tokens_seen": 44055005,
      "step": 2063,
      "time_per_iteration": 2.7217555046081543
    },
    {
      "auxiliary_loss_clip": 0.0115845,
      "auxiliary_loss_mlp": 0.01031237,
      "balance_loss_clip": 1.0514884,
      "balance_loss_mlp": 1.02066302,
      "epoch": 0.24818132627908374,
      "flos": 19282163351040.0,
      "grad_norm": 2.8138088439362523,
      "language_loss": 0.67231488,
      "learning_rate": 3.521241864223319e-06,
      "loss": 0.69421178,
      "num_input_tokens_seen": 44073965,
      "step": 2064,
      "time_per_iteration": 2.811619281768799
    },
    {
      "auxiliary_loss_clip": 0.01089748,
      "auxiliary_loss_mlp": 0.0100276,
      "balance_loss_clip": 1.02581513,
      "balance_loss_mlp": 1.00045907,
      "epoch": 0.24830156916972285,
      "flos": 70285837881600.0,
      "grad_norm": 0.7950553177015903,
      "language_loss": 0.61991215,
      "learning_rate": 3.5207360433622552e-06,
      "loss": 0.64083719,
      "num_input_tokens_seen": 44135965,
      "step": 2065,
      "time_per_iteration": 3.3014681339263916
    },
    {
      "auxiliary_loss_clip": 0.01176094,
      "auxiliary_loss_mlp": 0.0103343,
      "balance_loss_clip": 1.05637121,
      "balance_loss_mlp": 1.02448893,
      "epoch": 0.24842181206036193,
      "flos": 40409128287360.0,
      "grad_norm": 1.9767388972482969,
      "language_loss": 0.74661988,
      "learning_rate": 3.5202299918067437e-06,
      "loss": 0.76871514,
      "num_input_tokens_seen": 44159560,
      "step": 2066,
      "time_per_iteration": 2.952139139175415
    },
    {
      "auxiliary_loss_clip": 0.01191661,
      "auxiliary_loss_mlp": 0.01029962,
      "balance_loss_clip": 1.05683839,
      "balance_loss_mlp": 1.02039611,
      "epoch": 0.248542054951001,
      "flos": 20082432412800.0,
      "grad_norm": 2.4115896185735,
      "language_loss": 0.69690764,
      "learning_rate": 3.519723709633551e-06,
      "loss": 0.71912384,
      "num_input_tokens_seen": 44178320,
      "step": 2067,
      "time_per_iteration": 2.6985466480255127
    },
    {
      "auxiliary_loss_clip": 0.0117005,
      "auxiliary_loss_mlp": 0.0103987,
      "balance_loss_clip": 1.05371046,
      "balance_loss_mlp": 1.03062487,
      "epoch": 0.24866229784164012,
      "flos": 23513948363520.0,
      "grad_norm": 2.5456413462936425,
      "language_loss": 0.83950841,
      "learning_rate": 3.519217196919479e-06,
      "loss": 0.86160761,
      "num_input_tokens_seen": 44197305,
      "step": 2068,
      "time_per_iteration": 3.9210190773010254
    },
    {
      "auxiliary_loss_clip": 0.01188038,
      "auxiliary_loss_mlp": 0.01028826,
      "balance_loss_clip": 1.06088519,
      "balance_loss_mlp": 1.01893163,
      "epoch": 0.2487825407322792,
      "flos": 19865101173120.0,
      "grad_norm": 1.9082019589335015,
      "language_loss": 0.73112613,
      "learning_rate": 3.518710453741367e-06,
      "loss": 0.75329477,
      "num_input_tokens_seen": 44216505,
      "step": 2069,
      "time_per_iteration": 2.921764612197876
    },
    {
      "auxiliary_loss_clip": 0.01171122,
      "auxiliary_loss_mlp": 0.00764668,
      "balance_loss_clip": 1.05412054,
      "balance_loss_mlp": 1.00037456,
      "epoch": 0.2489027836229183,
      "flos": 22017622573440.0,
      "grad_norm": 2.2877345526597606,
      "language_loss": 0.68410498,
      "learning_rate": 3.518203480176086e-06,
      "loss": 0.70346284,
      "num_input_tokens_seen": 44235435,
      "step": 2070,
      "time_per_iteration": 2.764524459838867
    },
    {
      "auxiliary_loss_clip": 0.01121801,
      "auxiliary_loss_mlp": 0.01034896,
      "balance_loss_clip": 1.04631329,
      "balance_loss_mlp": 1.02523971,
      "epoch": 0.2490230265135574,
      "flos": 23294354567040.0,
      "grad_norm": 2.451052085552238,
      "language_loss": 0.8049444,
      "learning_rate": 3.517696276300545e-06,
      "loss": 0.82651132,
      "num_input_tokens_seen": 44256975,
      "step": 2071,
      "time_per_iteration": 3.1783180236816406
    },
    {
      "auxiliary_loss_clip": 0.01201793,
      "auxiliary_loss_mlp": 0.0103751,
      "balance_loss_clip": 1.06605566,
      "balance_loss_mlp": 1.02757394,
      "epoch": 0.24914326940419648,
      "flos": 19826784339840.0,
      "grad_norm": 3.2207487926735845,
      "language_loss": 0.69791341,
      "learning_rate": 3.517188842191685e-06,
      "loss": 0.72030646,
      "num_input_tokens_seen": 44275125,
      "step": 2072,
      "time_per_iteration": 3.7400591373443604
    },
    {
      "auxiliary_loss_clip": 0.01191927,
      "auxiliary_loss_mlp": 0.01039257,
      "balance_loss_clip": 1.05583453,
      "balance_loss_mlp": 1.0300777,
      "epoch": 0.24926351229483557,
      "flos": 20229271211520.0,
      "grad_norm": 1.6837915052542278,
      "language_loss": 0.73876727,
      "learning_rate": 3.5166811779264837e-06,
      "loss": 0.76107913,
      "num_input_tokens_seen": 44295445,
      "step": 2073,
      "time_per_iteration": 2.767338275909424
    },
    {
      "auxiliary_loss_clip": 0.01209892,
      "auxiliary_loss_mlp": 0.01036649,
      "balance_loss_clip": 1.05863369,
      "balance_loss_mlp": 1.02677846,
      "epoch": 0.24938375518547465,
      "flos": 23294570048640.0,
      "grad_norm": 2.1669018670170836,
      "language_loss": 0.77729058,
      "learning_rate": 3.5161732835819545e-06,
      "loss": 0.79975605,
      "num_input_tokens_seen": 44314755,
      "step": 2074,
      "time_per_iteration": 3.701648712158203
    },
    {
      "auxiliary_loss_clip": 0.0120973,
      "auxiliary_loss_mlp": 0.01045866,
      "balance_loss_clip": 1.05990684,
      "balance_loss_mlp": 1.03623414,
      "epoch": 0.24950399807611376,
      "flos": 17311673099520.0,
      "grad_norm": 2.373086213798395,
      "language_loss": 0.83161569,
      "learning_rate": 3.515665159235143e-06,
      "loss": 0.85417163,
      "num_input_tokens_seen": 44333640,
      "step": 2075,
      "time_per_iteration": 2.6778178215026855
    },
    {
      "auxiliary_loss_clip": 0.01177445,
      "auxiliary_loss_mlp": 0.01030846,
      "balance_loss_clip": 1.05240023,
      "balance_loss_mlp": 1.02214956,
      "epoch": 0.24962424096675284,
      "flos": 19024863252480.0,
      "grad_norm": 1.720002874889821,
      "language_loss": 0.75437963,
      "learning_rate": 3.5151568049631318e-06,
      "loss": 0.77646255,
      "num_input_tokens_seen": 44352355,
      "step": 2076,
      "time_per_iteration": 2.7661502361297607
    },
    {
      "auxiliary_loss_clip": 0.01212187,
      "auxiliary_loss_mlp": 0.010338,
      "balance_loss_clip": 1.06087756,
      "balance_loss_mlp": 1.02441263,
      "epoch": 0.24974448385739192,
      "flos": 33398790710400.0,
      "grad_norm": 2.33962585203716,
      "language_loss": 0.80274928,
      "learning_rate": 3.5146482208430385e-06,
      "loss": 0.82520908,
      "num_input_tokens_seen": 44374185,
      "step": 2077,
      "time_per_iteration": 2.7240095138549805
    },
    {
      "auxiliary_loss_clip": 0.01119686,
      "auxiliary_loss_mlp": 0.01036704,
      "balance_loss_clip": 1.04357481,
      "balance_loss_mlp": 1.02598703,
      "epoch": 0.24986472674803104,
      "flos": 30007279532160.0,
      "grad_norm": 1.9876625207351892,
      "language_loss": 0.67939067,
      "learning_rate": 3.514139406952014e-06,
      "loss": 0.7009545,
      "num_input_tokens_seen": 44396210,
      "step": 2078,
      "time_per_iteration": 2.969722270965576
    },
    {
      "auxiliary_loss_clip": 0.01196874,
      "auxiliary_loss_mlp": 0.01036294,
      "balance_loss_clip": 1.05887973,
      "balance_loss_mlp": 1.0267272,
      "epoch": 0.24998496963867012,
      "flos": 26613074833920.0,
      "grad_norm": 1.9053723760923638,
      "language_loss": 0.83095759,
      "learning_rate": 3.5136303633672454e-06,
      "loss": 0.85328925,
      "num_input_tokens_seen": 44416340,
      "step": 2079,
      "time_per_iteration": 2.78934383392334
    },
    {
      "auxiliary_loss_clip": 0.01175117,
      "auxiliary_loss_mlp": 0.00764958,
      "balance_loss_clip": 1.05699205,
      "balance_loss_mlp": 1.00041342,
      "epoch": 0.25010521252930923,
      "flos": 23553989049600.0,
      "grad_norm": 1.916182810327473,
      "language_loss": 0.74538177,
      "learning_rate": 3.5131210901659544e-06,
      "loss": 0.76478255,
      "num_input_tokens_seen": 44438095,
      "step": 2080,
      "time_per_iteration": 2.9293713569641113
    },
    {
      "auxiliary_loss_clip": 0.01160653,
      "auxiliary_loss_mlp": 0.01032507,
      "balance_loss_clip": 1.05276704,
      "balance_loss_mlp": 1.02312541,
      "epoch": 0.2502254554199483,
      "flos": 23441193365760.0,
      "grad_norm": 2.7974632088318345,
      "language_loss": 0.82008088,
      "learning_rate": 3.5126115874253967e-06,
      "loss": 0.84201252,
      "num_input_tokens_seen": 44457650,
      "step": 2081,
      "time_per_iteration": 2.87642502784729
    },
    {
      "auxiliary_loss_clip": 0.01165725,
      "auxiliary_loss_mlp": 0.01038326,
      "balance_loss_clip": 1.05714345,
      "balance_loss_mlp": 1.02878892,
      "epoch": 0.2503456983105874,
      "flos": 28761681651840.0,
      "grad_norm": 2.3144013342467757,
      "language_loss": 0.81058764,
      "learning_rate": 3.5121018552228644e-06,
      "loss": 0.83262813,
      "num_input_tokens_seen": 44476155,
      "step": 2082,
      "time_per_iteration": 2.855018138885498
    },
    {
      "auxiliary_loss_clip": 0.01166169,
      "auxiliary_loss_mlp": 0.01032248,
      "balance_loss_clip": 1.05354679,
      "balance_loss_mlp": 1.02302718,
      "epoch": 0.2504659412012265,
      "flos": 18770256673920.0,
      "grad_norm": 2.1202660526467185,
      "language_loss": 0.7656942,
      "learning_rate": 3.5115918936356827e-06,
      "loss": 0.78767836,
      "num_input_tokens_seen": 44492910,
      "step": 2083,
      "time_per_iteration": 2.850407600402832
    },
    {
      "auxiliary_loss_clip": 0.01143838,
      "auxiliary_loss_mlp": 0.01030193,
      "balance_loss_clip": 1.04978132,
      "balance_loss_mlp": 1.02064383,
      "epoch": 0.25058618409186556,
      "flos": 16873383346560.0,
      "grad_norm": 2.3254730007314115,
      "language_loss": 0.78762031,
      "learning_rate": 3.5110817027412123e-06,
      "loss": 0.80936062,
      "num_input_tokens_seen": 44512000,
      "step": 2084,
      "time_per_iteration": 2.801792860031128
    },
    {
      "auxiliary_loss_clip": 0.01160109,
      "auxiliary_loss_mlp": 0.01038458,
      "balance_loss_clip": 1.05161953,
      "balance_loss_mlp": 1.0293144,
      "epoch": 0.25070642698250467,
      "flos": 24425540651520.0,
      "grad_norm": 1.992033321140226,
      "language_loss": 0.68939644,
      "learning_rate": 3.5105712826168493e-06,
      "loss": 0.71138215,
      "num_input_tokens_seen": 44531650,
      "step": 2085,
      "time_per_iteration": 2.808412790298462
    },
    {
      "auxiliary_loss_clip": 0.01193775,
      "auxiliary_loss_mlp": 0.00764503,
      "balance_loss_clip": 1.05737984,
      "balance_loss_mlp": 1.00033522,
      "epoch": 0.2508266698731437,
      "flos": 20260944028800.0,
      "grad_norm": 2.054083386523751,
      "language_loss": 0.7089113,
      "learning_rate": 3.5100606333400235e-06,
      "loss": 0.72849405,
      "num_input_tokens_seen": 44548785,
      "step": 2086,
      "time_per_iteration": 2.778442144393921
    },
    {
      "auxiliary_loss_clip": 0.01193523,
      "auxiliary_loss_mlp": 0.01036873,
      "balance_loss_clip": 1.06006527,
      "balance_loss_mlp": 1.02658474,
      "epoch": 0.25094691276378284,
      "flos": 19245318975360.0,
      "grad_norm": 2.455778940500041,
      "language_loss": 0.76703489,
      "learning_rate": 3.5095497549882006e-06,
      "loss": 0.78933883,
      "num_input_tokens_seen": 44567230,
      "step": 2087,
      "time_per_iteration": 2.6960651874542236
    },
    {
      "auxiliary_loss_clip": 0.01196978,
      "auxiliary_loss_mlp": 0.01030883,
      "balance_loss_clip": 1.0614233,
      "balance_loss_mlp": 1.02098846,
      "epoch": 0.25106715565442195,
      "flos": 26943237671040.0,
      "grad_norm": 2.6274202721294753,
      "language_loss": 0.72131842,
      "learning_rate": 3.50903864763888e-06,
      "loss": 0.74359703,
      "num_input_tokens_seen": 44588020,
      "step": 2088,
      "time_per_iteration": 2.8209304809570312
    },
    {
      "auxiliary_loss_clip": 0.01197862,
      "auxiliary_loss_mlp": 0.01034692,
      "balance_loss_clip": 1.05658317,
      "balance_loss_mlp": 1.0245589,
      "epoch": 0.251187398545061,
      "flos": 48359570572800.0,
      "grad_norm": 1.979996228397373,
      "language_loss": 0.76201183,
      "learning_rate": 3.5085273113695965e-06,
      "loss": 0.78433734,
      "num_input_tokens_seen": 44612590,
      "step": 2089,
      "time_per_iteration": 2.9548180103302
    },
    {
      "auxiliary_loss_clip": 0.01210386,
      "auxiliary_loss_mlp": 0.01035909,
      "balance_loss_clip": 1.06060505,
      "balance_loss_mlp": 1.02615726,
      "epoch": 0.2513076414357001,
      "flos": 27016100409600.0,
      "grad_norm": 2.4255743283953586,
      "language_loss": 0.78970957,
      "learning_rate": 3.508015746257919e-06,
      "loss": 0.81217253,
      "num_input_tokens_seen": 44631630,
      "step": 2090,
      "time_per_iteration": 2.7150540351867676
    },
    {
      "auxiliary_loss_clip": 0.0116894,
      "auxiliary_loss_mlp": 0.01037783,
      "balance_loss_clip": 1.05553675,
      "balance_loss_mlp": 1.02754307,
      "epoch": 0.2514278843263392,
      "flos": 19463619882240.0,
      "grad_norm": 2.0982594889736257,
      "language_loss": 0.83303434,
      "learning_rate": 3.5075039523814518e-06,
      "loss": 0.85510159,
      "num_input_tokens_seen": 44650820,
      "step": 2091,
      "time_per_iteration": 2.772280216217041
    },
    {
      "auxiliary_loss_clip": 0.01198772,
      "auxiliary_loss_mlp": 0.01037313,
      "balance_loss_clip": 1.05619061,
      "balance_loss_mlp": 1.02739477,
      "epoch": 0.2515481272169783,
      "flos": 16866092885760.0,
      "grad_norm": 2.193088283659089,
      "language_loss": 0.82297373,
      "learning_rate": 3.506991929817834e-06,
      "loss": 0.84533453,
      "num_input_tokens_seen": 44667540,
      "step": 2092,
      "time_per_iteration": 2.7446253299713135
    },
    {
      "auxiliary_loss_clip": 0.01210816,
      "auxiliary_loss_mlp": 0.01042636,
      "balance_loss_clip": 1.06287575,
      "balance_loss_mlp": 1.03322458,
      "epoch": 0.2516683701076174,
      "flos": 23732464752000.0,
      "grad_norm": 1.9132180173084559,
      "language_loss": 0.82404816,
      "learning_rate": 3.506479678644738e-06,
      "loss": 0.84658265,
      "num_input_tokens_seen": 44687935,
      "step": 2093,
      "time_per_iteration": 2.7368557453155518
    },
    {
      "auxiliary_loss_clip": 0.01141322,
      "auxiliary_loss_mlp": 0.01033733,
      "balance_loss_clip": 1.05021346,
      "balance_loss_mlp": 1.02388644,
      "epoch": 0.2517886129982565,
      "flos": 27635954434560.0,
      "grad_norm": 3.787529160621812,
      "language_loss": 0.74330527,
      "learning_rate": 3.505967198939873e-06,
      "loss": 0.76505584,
      "num_input_tokens_seen": 44704975,
      "step": 2094,
      "time_per_iteration": 3.78482985496521
    },
    {
      "auxiliary_loss_clip": 0.01175368,
      "auxiliary_loss_mlp": 0.01027657,
      "balance_loss_clip": 1.05153322,
      "balance_loss_mlp": 1.01847243,
      "epoch": 0.25190885588889556,
      "flos": 38104596529920.0,
      "grad_norm": 2.289844789887299,
      "language_loss": 0.7836237,
      "learning_rate": 3.5054544907809813e-06,
      "loss": 0.80565393,
      "num_input_tokens_seen": 44725475,
      "step": 2095,
      "time_per_iteration": 2.9173803329467773
    },
    {
      "auxiliary_loss_clip": 0.01179268,
      "auxiliary_loss_mlp": 0.00765011,
      "balance_loss_clip": 1.05823636,
      "balance_loss_mlp": 1.00048053,
      "epoch": 0.25202909877953467,
      "flos": 22269894768000.0,
      "grad_norm": 2.3562729946411687,
      "language_loss": 0.80759048,
      "learning_rate": 3.50494155424584e-06,
      "loss": 0.82703328,
      "num_input_tokens_seen": 44744380,
      "step": 2096,
      "time_per_iteration": 2.813070774078369
    },
    {
      "auxiliary_loss_clip": 0.01199673,
      "auxiliary_loss_mlp": 0.01036545,
      "balance_loss_clip": 1.05877781,
      "balance_loss_mlp": 1.02688909,
      "epoch": 0.2521493416701738,
      "flos": 21761759018880.0,
      "grad_norm": 1.538458745889626,
      "language_loss": 0.82841027,
      "learning_rate": 3.504428389412262e-06,
      "loss": 0.8507725,
      "num_input_tokens_seen": 44765190,
      "step": 2097,
      "time_per_iteration": 2.71197509765625
    },
    {
      "auxiliary_loss_clip": 0.01192604,
      "auxiliary_loss_mlp": 0.01034004,
      "balance_loss_clip": 1.05614793,
      "balance_loss_mlp": 1.02360332,
      "epoch": 0.25226958456081283,
      "flos": 27746738956800.0,
      "grad_norm": 2.206747127047161,
      "language_loss": 0.73337299,
      "learning_rate": 3.5039149963580927e-06,
      "loss": 0.75563908,
      "num_input_tokens_seen": 44785210,
      "step": 2098,
      "time_per_iteration": 3.728706121444702
    },
    {
      "auxiliary_loss_clip": 0.01175896,
      "auxiliary_loss_mlp": 0.01036115,
      "balance_loss_clip": 1.05799174,
      "balance_loss_mlp": 1.0262562,
      "epoch": 0.25238982745145194,
      "flos": 30732171903360.0,
      "grad_norm": 2.9626831965411182,
      "language_loss": 0.70408916,
      "learning_rate": 3.503401375161215e-06,
      "loss": 0.72620928,
      "num_input_tokens_seen": 44804955,
      "step": 2099,
      "time_per_iteration": 3.8294928073883057
    },
    {
      "auxiliary_loss_clip": 0.01206507,
      "auxiliary_loss_mlp": 0.01032362,
      "balance_loss_clip": 1.05835581,
      "balance_loss_mlp": 1.02217519,
      "epoch": 0.252510070342091,
      "flos": 20266331068800.0,
      "grad_norm": 1.558947084924156,
      "language_loss": 0.83935559,
      "learning_rate": 3.502887525899544e-06,
      "loss": 0.86174434,
      "num_input_tokens_seen": 44823935,
      "step": 2100,
      "time_per_iteration": 3.644076347351074
    },
    {
      "auxiliary_loss_clip": 0.01181774,
      "auxiliary_loss_mlp": 0.01043802,
      "balance_loss_clip": 1.05500484,
      "balance_loss_mlp": 1.03396153,
      "epoch": 0.2526303132327301,
      "flos": 22747399194240.0,
      "grad_norm": 3.010336497512438,
      "language_loss": 0.83238769,
      "learning_rate": 3.50237344865103e-06,
      "loss": 0.85464346,
      "num_input_tokens_seen": 44844935,
      "step": 2101,
      "time_per_iteration": 2.90242075920105
    },
    {
      "auxiliary_loss_clip": 0.0120977,
      "auxiliary_loss_mlp": 0.01034468,
      "balance_loss_clip": 1.05916834,
      "balance_loss_mlp": 1.02459788,
      "epoch": 0.2527505561233692,
      "flos": 30263466309120.0,
      "grad_norm": 2.166633424514748,
      "language_loss": 0.76218939,
      "learning_rate": 3.501859143493658e-06,
      "loss": 0.78463173,
      "num_input_tokens_seen": 44865565,
      "step": 2102,
      "time_per_iteration": 2.692326545715332
    },
    {
      "auxiliary_loss_clip": 0.01113325,
      "auxiliary_loss_mlp": 0.01002113,
      "balance_loss_clip": 1.03145957,
      "balance_loss_mlp": 0.99987143,
      "epoch": 0.2528707990140083,
      "flos": 58492917164160.0,
      "grad_norm": 0.9944362767761876,
      "language_loss": 0.60552663,
      "learning_rate": 3.5013446105054488e-06,
      "loss": 0.62668103,
      "num_input_tokens_seen": 44918485,
      "step": 2103,
      "time_per_iteration": 2.9874167442321777
    },
    {
      "auxiliary_loss_clip": 0.0115198,
      "auxiliary_loss_mlp": 0.01030251,
      "balance_loss_clip": 1.05325305,
      "balance_loss_mlp": 1.02055883,
      "epoch": 0.2529910419046474,
      "flos": 24645134448000.0,
      "grad_norm": 1.826116255193779,
      "language_loss": 0.75234997,
      "learning_rate": 3.5008298497644555e-06,
      "loss": 0.77417231,
      "num_input_tokens_seen": 44937530,
      "step": 2104,
      "time_per_iteration": 2.7872188091278076
    },
    {
      "auxiliary_loss_clip": 0.01171006,
      "auxiliary_loss_mlp": 0.01033557,
      "balance_loss_clip": 1.05623889,
      "balance_loss_mlp": 1.02370977,
      "epoch": 0.2531112847952865,
      "flos": 23842135952640.0,
      "grad_norm": 1.5824702435074918,
      "language_loss": 0.8812499,
      "learning_rate": 3.500314861348767e-06,
      "loss": 0.90329546,
      "num_input_tokens_seen": 44958165,
      "step": 2105,
      "time_per_iteration": 2.8249404430389404
    },
    {
      "auxiliary_loss_clip": 0.01156062,
      "auxiliary_loss_mlp": 0.01036635,
      "balance_loss_clip": 1.05506158,
      "balance_loss_mlp": 1.02699125,
      "epoch": 0.25323152768592555,
      "flos": 16143822207360.0,
      "grad_norm": 2.088055278211382,
      "language_loss": 0.77033854,
      "learning_rate": 3.499799645336507e-06,
      "loss": 0.79226553,
      "num_input_tokens_seen": 44975060,
      "step": 2106,
      "time_per_iteration": 2.9788529872894287
    },
    {
      "auxiliary_loss_clip": 0.01196619,
      "auxiliary_loss_mlp": 0.01035717,
      "balance_loss_clip": 1.05902719,
      "balance_loss_mlp": 1.02653813,
      "epoch": 0.25335177057656466,
      "flos": 28405161210240.0,
      "grad_norm": 1.9487282598925275,
      "language_loss": 0.87118274,
      "learning_rate": 3.4992842018058336e-06,
      "loss": 0.89350611,
      "num_input_tokens_seen": 44997960,
      "step": 2107,
      "time_per_iteration": 2.7902863025665283
    },
    {
      "auxiliary_loss_clip": 0.0117032,
      "auxiliary_loss_mlp": 0.01032964,
      "balance_loss_clip": 1.05404294,
      "balance_loss_mlp": 1.02351642,
      "epoch": 0.25347201346720377,
      "flos": 18799666934400.0,
      "grad_norm": 2.079012602543159,
      "language_loss": 0.88438392,
      "learning_rate": 3.4987685308349384e-06,
      "loss": 0.90641677,
      "num_input_tokens_seen": 45015690,
      "step": 2108,
      "time_per_iteration": 2.8269617557525635
    },
    {
      "auxiliary_loss_clip": 0.01162452,
      "auxiliary_loss_mlp": 0.01032585,
      "balance_loss_clip": 1.05016184,
      "balance_loss_mlp": 1.02244604,
      "epoch": 0.2535922563578428,
      "flos": 15815490963840.0,
      "grad_norm": 2.3815593466259215,
      "language_loss": 0.61252975,
      "learning_rate": 3.4982526325020497e-06,
      "loss": 0.63448012,
      "num_input_tokens_seen": 45032660,
      "step": 2109,
      "time_per_iteration": 2.7323598861694336
    },
    {
      "auxiliary_loss_clip": 0.01183323,
      "auxiliary_loss_mlp": 0.01039757,
      "balance_loss_clip": 1.05629325,
      "balance_loss_mlp": 1.03007686,
      "epoch": 0.25371249924848194,
      "flos": 16318922031360.0,
      "grad_norm": 2.3535751722183953,
      "language_loss": 0.81786191,
      "learning_rate": 3.4977365068854273e-06,
      "loss": 0.84009272,
      "num_input_tokens_seen": 45048280,
      "step": 2110,
      "time_per_iteration": 2.7319705486297607
    },
    {
      "auxiliary_loss_clip": 0.01170902,
      "auxiliary_loss_mlp": 0.01039338,
      "balance_loss_clip": 1.0531168,
      "balance_loss_mlp": 1.02937233,
      "epoch": 0.25383274213912105,
      "flos": 21761615364480.0,
      "grad_norm": 2.7380751622515964,
      "language_loss": 0.7373144,
      "learning_rate": 3.4972201540633676e-06,
      "loss": 0.75941676,
      "num_input_tokens_seen": 45067635,
      "step": 2111,
      "time_per_iteration": 2.772465944290161
    },
    {
      "auxiliary_loss_clip": 0.01171114,
      "auxiliary_loss_mlp": 0.01040407,
      "balance_loss_clip": 1.05370092,
      "balance_loss_mlp": 1.0305903,
      "epoch": 0.2539529850297601,
      "flos": 21396870708480.0,
      "grad_norm": 2.644865874237856,
      "language_loss": 0.85352665,
      "learning_rate": 3.4967035741142008e-06,
      "loss": 0.87564188,
      "num_input_tokens_seen": 45086455,
      "step": 2112,
      "time_per_iteration": 2.734591007232666
    },
    {
      "auxiliary_loss_clip": 0.0117725,
      "auxiliary_loss_mlp": 0.01038565,
      "balance_loss_clip": 1.06290865,
      "balance_loss_mlp": 1.02912927,
      "epoch": 0.2540732279203992,
      "flos": 25228467319680.0,
      "grad_norm": 1.945353072080155,
      "language_loss": 0.81861377,
      "learning_rate": 3.4961867671162917e-06,
      "loss": 0.84077191,
      "num_input_tokens_seen": 45106385,
      "step": 2113,
      "time_per_iteration": 2.7956430912017822
    },
    {
      "auxiliary_loss_clip": 0.01211733,
      "auxiliary_loss_mlp": 0.01035047,
      "balance_loss_clip": 1.05980217,
      "balance_loss_mlp": 1.02539742,
      "epoch": 0.2541934708110383,
      "flos": 19427386037760.0,
      "grad_norm": 24.17308036490114,
      "language_loss": 0.77476478,
      "learning_rate": 3.4956697331480402e-06,
      "loss": 0.79723257,
      "num_input_tokens_seen": 45124955,
      "step": 2114,
      "time_per_iteration": 2.6350395679473877
    },
    {
      "auxiliary_loss_clip": 0.01169144,
      "auxiliary_loss_mlp": 0.01033151,
      "balance_loss_clip": 1.05109441,
      "balance_loss_mlp": 1.02302432,
      "epoch": 0.2543137137016774,
      "flos": 23949436855680.0,
      "grad_norm": 1.7929217185537303,
      "language_loss": 0.80137265,
      "learning_rate": 3.495152472287879e-06,
      "loss": 0.82339561,
      "num_input_tokens_seen": 45145665,
      "step": 2115,
      "time_per_iteration": 2.8250572681427
    },
    {
      "auxiliary_loss_clip": 0.01170133,
      "auxiliary_loss_mlp": 0.01040135,
      "balance_loss_clip": 1.05603552,
      "balance_loss_mlp": 1.03024101,
      "epoch": 0.2544339565923165,
      "flos": 25593283802880.0,
      "grad_norm": 2.2045217641814268,
      "language_loss": 0.7365135,
      "learning_rate": 3.4946349846142766e-06,
      "loss": 0.75861615,
      "num_input_tokens_seen": 45164805,
      "step": 2116,
      "time_per_iteration": 2.7879958152770996
    },
    {
      "auxiliary_loss_clip": 0.01206955,
      "auxiliary_loss_mlp": 0.01031452,
      "balance_loss_clip": 1.05637228,
      "balance_loss_mlp": 1.02161705,
      "epoch": 0.25455419948295555,
      "flos": 21689470897920.0,
      "grad_norm": 2.6228960204533265,
      "language_loss": 0.75870514,
      "learning_rate": 3.4941172702057353e-06,
      "loss": 0.78108925,
      "num_input_tokens_seen": 45184865,
      "step": 2117,
      "time_per_iteration": 2.684295415878296
    },
    {
      "auxiliary_loss_clip": 0.01178897,
      "auxiliary_loss_mlp": 0.01028739,
      "balance_loss_clip": 1.05703068,
      "balance_loss_mlp": 1.01932168,
      "epoch": 0.25467444237359466,
      "flos": 26250341339520.0,
      "grad_norm": 2.095004872376581,
      "language_loss": 0.80710781,
      "learning_rate": 3.4935993291407924e-06,
      "loss": 0.82918417,
      "num_input_tokens_seen": 45203690,
      "step": 2118,
      "time_per_iteration": 2.715148687362671
    },
    {
      "auxiliary_loss_clip": 0.01173171,
      "auxiliary_loss_mlp": 0.01037342,
      "balance_loss_clip": 1.05348051,
      "balance_loss_mlp": 1.02786446,
      "epoch": 0.25479468526423377,
      "flos": 26979686997120.0,
      "grad_norm": 2.330762367787955,
      "language_loss": 0.7109521,
      "learning_rate": 3.4930811614980183e-06,
      "loss": 0.73305726,
      "num_input_tokens_seen": 45225385,
      "step": 2119,
      "time_per_iteration": 2.790679931640625
    },
    {
      "auxiliary_loss_clip": 0.01184125,
      "auxiliary_loss_mlp": 0.01039836,
      "balance_loss_clip": 1.05327165,
      "balance_loss_mlp": 1.03040099,
      "epoch": 0.2549149281548728,
      "flos": 23475811098240.0,
      "grad_norm": 1.8511334546024447,
      "language_loss": 0.79416692,
      "learning_rate": 3.4925627673560198e-06,
      "loss": 0.81640649,
      "num_input_tokens_seen": 45246045,
      "step": 2120,
      "time_per_iteration": 3.616147041320801
    },
    {
      "auxiliary_loss_clip": 0.01166544,
      "auxiliary_loss_mlp": 0.01032999,
      "balance_loss_clip": 1.05476046,
      "balance_loss_mlp": 1.02347386,
      "epoch": 0.25503517104551193,
      "flos": 25812302981760.0,
      "grad_norm": 2.0014682201325864,
      "language_loss": 0.88637519,
      "learning_rate": 3.4920441467934357e-06,
      "loss": 0.90837061,
      "num_input_tokens_seen": 45266560,
      "step": 2121,
      "time_per_iteration": 2.879087209701538
    },
    {
      "auxiliary_loss_clip": 0.01154851,
      "auxiliary_loss_mlp": 0.01032899,
      "balance_loss_clip": 1.05275333,
      "balance_loss_mlp": 1.02312422,
      "epoch": 0.25515541393615104,
      "flos": 26645106787200.0,
      "grad_norm": 2.662889723242326,
      "language_loss": 0.82796359,
      "learning_rate": 3.491525299888941e-06,
      "loss": 0.84984112,
      "num_input_tokens_seen": 45285405,
      "step": 2122,
      "time_per_iteration": 2.838909149169922
    },
    {
      "auxiliary_loss_clip": 0.01076056,
      "auxiliary_loss_mlp": 0.00754994,
      "balance_loss_clip": 1.02702904,
      "balance_loss_mlp": 0.99997914,
      "epoch": 0.2552756568267901,
      "flos": 65955945847680.0,
      "grad_norm": 0.8856006731170495,
      "language_loss": 0.62598377,
      "learning_rate": 3.491006226721244e-06,
      "loss": 0.64429426,
      "num_input_tokens_seen": 45349615,
      "step": 2123,
      "time_per_iteration": 3.4008078575134277
    },
    {
      "auxiliary_loss_clip": 0.01182109,
      "auxiliary_loss_mlp": 0.00765308,
      "balance_loss_clip": 1.05752134,
      "balance_loss_mlp": 1.00032997,
      "epoch": 0.2553958997174292,
      "flos": 17931096161280.0,
      "grad_norm": 2.0940899259548558,
      "language_loss": 0.7790463,
      "learning_rate": 3.4904869273690882e-06,
      "loss": 0.79852057,
      "num_input_tokens_seen": 45367505,
      "step": 2124,
      "time_per_iteration": 3.8781068325042725
    },
    {
      "auxiliary_loss_clip": 0.01196375,
      "auxiliary_loss_mlp": 0.01038066,
      "balance_loss_clip": 1.05640233,
      "balance_loss_mlp": 1.02888119,
      "epoch": 0.2555161426080683,
      "flos": 23367791923200.0,
      "grad_norm": 7.451055772982942,
      "language_loss": 0.89110541,
      "learning_rate": 3.489967401911251e-06,
      "loss": 0.91344976,
      "num_input_tokens_seen": 45386805,
      "step": 2125,
      "time_per_iteration": 2.731290102005005
    },
    {
      "auxiliary_loss_clip": 0.01213874,
      "auxiliary_loss_mlp": 0.01041714,
      "balance_loss_clip": 1.06037855,
      "balance_loss_mlp": 1.03123546,
      "epoch": 0.2556363854987074,
      "flos": 40625130723840.0,
      "grad_norm": 1.8677728351604943,
      "language_loss": 0.69655752,
      "learning_rate": 3.4894476504265428e-06,
      "loss": 0.71911347,
      "num_input_tokens_seen": 45411045,
      "step": 2126,
      "time_per_iteration": 4.784552335739136
    },
    {
      "auxiliary_loss_clip": 0.01091001,
      "auxiliary_loss_mlp": 0.0100738,
      "balance_loss_clip": 1.0253849,
      "balance_loss_mlp": 1.00504398,
      "epoch": 0.2557566283893465,
      "flos": 68019443389440.0,
      "grad_norm": 0.7361090218422459,
      "language_loss": 0.54396695,
      "learning_rate": 3.4889276729938104e-06,
      "loss": 0.56495076,
      "num_input_tokens_seen": 45469575,
      "step": 2127,
      "time_per_iteration": 3.181098461151123
    },
    {
      "auxiliary_loss_clip": 0.01178412,
      "auxiliary_loss_mlp": 0.01034596,
      "balance_loss_clip": 1.056705,
      "balance_loss_mlp": 1.02494597,
      "epoch": 0.2558768712799856,
      "flos": 22635645004800.0,
      "grad_norm": 2.61794494991567,
      "language_loss": 0.80379659,
      "learning_rate": 3.488407469691934e-06,
      "loss": 0.82592666,
      "num_input_tokens_seen": 45490270,
      "step": 2128,
      "time_per_iteration": 2.7905967235565186
    },
    {
      "auxiliary_loss_clip": 0.0117545,
      "auxiliary_loss_mlp": 0.01038346,
      "balance_loss_clip": 1.05155885,
      "balance_loss_mlp": 1.02751637,
      "epoch": 0.25599711417062465,
      "flos": 26396354125440.0,
      "grad_norm": 6.184548946773142,
      "language_loss": 0.80730665,
      "learning_rate": 3.487887040599828e-06,
      "loss": 0.82944459,
      "num_input_tokens_seen": 45510070,
      "step": 2129,
      "time_per_iteration": 2.7658798694610596
    },
    {
      "auxiliary_loss_clip": 0.01209172,
      "auxiliary_loss_mlp": 0.0103584,
      "balance_loss_clip": 1.05855966,
      "balance_loss_mlp": 1.0254693,
      "epoch": 0.25611735706126376,
      "flos": 22852042490880.0,
      "grad_norm": 3.0497883968533217,
      "language_loss": 0.76680368,
      "learning_rate": 3.4873663857964407e-06,
      "loss": 0.78925383,
      "num_input_tokens_seen": 45527285,
      "step": 2130,
      "time_per_iteration": 2.7820546627044678
    },
    {
      "auxiliary_loss_clip": 0.01150067,
      "auxiliary_loss_mlp": 0.01037488,
      "balance_loss_clip": 1.0517019,
      "balance_loss_mlp": 1.02768862,
      "epoch": 0.2562375999519028,
      "flos": 23367863750400.0,
      "grad_norm": 1.9981819365431217,
      "language_loss": 0.66441989,
      "learning_rate": 3.4868455053607556e-06,
      "loss": 0.68629539,
      "num_input_tokens_seen": 45546900,
      "step": 2131,
      "time_per_iteration": 2.795693874359131
    },
    {
      "auxiliary_loss_clip": 0.01198031,
      "auxiliary_loss_mlp": 0.01040856,
      "balance_loss_clip": 1.0555222,
      "balance_loss_mlp": 1.02991891,
      "epoch": 0.2563578428425419,
      "flos": 22856962654080.0,
      "grad_norm": 2.321354515367387,
      "language_loss": 0.71469575,
      "learning_rate": 3.486324399371789e-06,
      "loss": 0.73708463,
      "num_input_tokens_seen": 45566200,
      "step": 2132,
      "time_per_iteration": 2.706897258758545
    },
    {
      "auxiliary_loss_clip": 0.01157375,
      "auxiliary_loss_mlp": 0.0103432,
      "balance_loss_clip": 1.05261362,
      "balance_loss_mlp": 1.02515292,
      "epoch": 0.25647808573318104,
      "flos": 21653883498240.0,
      "grad_norm": 2.3529984929513006,
      "language_loss": 0.78497601,
      "learning_rate": 3.485803067908593e-06,
      "loss": 0.80689299,
      "num_input_tokens_seen": 45585710,
      "step": 2133,
      "time_per_iteration": 2.81225323677063
    },
    {
      "auxiliary_loss_clip": 0.01109165,
      "auxiliary_loss_mlp": 0.01035258,
      "balance_loss_clip": 1.04120779,
      "balance_loss_mlp": 1.02474403,
      "epoch": 0.2565983286238201,
      "flos": 33730569659520.0,
      "grad_norm": 1.7131906706540836,
      "language_loss": 0.79711282,
      "learning_rate": 3.485281511050253e-06,
      "loss": 0.81855714,
      "num_input_tokens_seen": 45607845,
      "step": 2134,
      "time_per_iteration": 2.927603244781494
    },
    {
      "auxiliary_loss_clip": 0.01194759,
      "auxiliary_loss_mlp": 0.01031059,
      "balance_loss_clip": 1.05515313,
      "balance_loss_mlp": 1.02156401,
      "epoch": 0.2567185715144592,
      "flos": 16216002587520.0,
      "grad_norm": 2.7731304703744524,
      "language_loss": 0.9006232,
      "learning_rate": 3.484759728875889e-06,
      "loss": 0.92288136,
      "num_input_tokens_seen": 45623210,
      "step": 2135,
      "time_per_iteration": 2.775146245956421
    },
    {
      "auxiliary_loss_clip": 0.01138723,
      "auxiliary_loss_mlp": 0.01041339,
      "balance_loss_clip": 1.0508976,
      "balance_loss_mlp": 1.03093755,
      "epoch": 0.2568388144050983,
      "flos": 17458475984640.0,
      "grad_norm": 1.8329001645156806,
      "language_loss": 0.80755979,
      "learning_rate": 3.4842377214646543e-06,
      "loss": 0.82936043,
      "num_input_tokens_seen": 45641505,
      "step": 2136,
      "time_per_iteration": 2.832365036010742
    },
    {
      "auxiliary_loss_clip": 0.01209816,
      "auxiliary_loss_mlp": 0.01032661,
      "balance_loss_clip": 1.06009054,
      "balance_loss_mlp": 1.02370846,
      "epoch": 0.25695905729573737,
      "flos": 20887442069760.0,
      "grad_norm": 1.9241005057382108,
      "language_loss": 0.66732454,
      "learning_rate": 3.483715488895737e-06,
      "loss": 0.6897493,
      "num_input_tokens_seen": 45661835,
      "step": 2137,
      "time_per_iteration": 2.663938045501709
    },
    {
      "auxiliary_loss_clip": 0.01145476,
      "auxiliary_loss_mlp": 0.01035919,
      "balance_loss_clip": 1.04854965,
      "balance_loss_mlp": 1.02618527,
      "epoch": 0.2570793001863765,
      "flos": 24717278914560.0,
      "grad_norm": 1.9509059256255346,
      "language_loss": 0.78783184,
      "learning_rate": 3.48319303124836e-06,
      "loss": 0.80964577,
      "num_input_tokens_seen": 45682215,
      "step": 2138,
      "time_per_iteration": 2.8674726486206055
    },
    {
      "auxiliary_loss_clip": 0.01177483,
      "auxiliary_loss_mlp": 0.01041974,
      "balance_loss_clip": 1.05764079,
      "balance_loss_mlp": 1.031883,
      "epoch": 0.2571995430770156,
      "flos": 26906896085760.0,
      "grad_norm": 2.5561622939841393,
      "language_loss": 0.66996789,
      "learning_rate": 3.4826703486017798e-06,
      "loss": 0.69216251,
      "num_input_tokens_seen": 45701840,
      "step": 2139,
      "time_per_iteration": 2.931387424468994
    },
    {
      "auxiliary_loss_clip": 0.01192368,
      "auxiliary_loss_mlp": 0.01035344,
      "balance_loss_clip": 1.05831504,
      "balance_loss_mlp": 1.0252769,
      "epoch": 0.25731978596765465,
      "flos": 19792561656960.0,
      "grad_norm": 1.8913248310369348,
      "language_loss": 0.76746511,
      "learning_rate": 3.4821474410352867e-06,
      "loss": 0.78974223,
      "num_input_tokens_seen": 45720500,
      "step": 2140,
      "time_per_iteration": 2.725515604019165
    },
    {
      "auxiliary_loss_clip": 0.01080166,
      "auxiliary_loss_mlp": 0.010066,
      "balance_loss_clip": 1.02787566,
      "balance_loss_mlp": 1.00424004,
      "epoch": 0.25744002885829376,
      "flos": 70564970471040.0,
      "grad_norm": 1.0690847285607001,
      "language_loss": 0.62592328,
      "learning_rate": 3.481624308628205e-06,
      "loss": 0.64679098,
      "num_input_tokens_seen": 45781870,
      "step": 2141,
      "time_per_iteration": 3.4509148597717285
    },
    {
      "auxiliary_loss_clip": 0.01179578,
      "auxiliary_loss_mlp": 0.01042891,
      "balance_loss_clip": 1.05512834,
      "balance_loss_mlp": 1.03240037,
      "epoch": 0.25756027174893287,
      "flos": 18038181582720.0,
      "grad_norm": 3.5636346313760625,
      "language_loss": 1.00075114,
      "learning_rate": 3.481100951459893e-06,
      "loss": 1.0229758,
      "num_input_tokens_seen": 45794890,
      "step": 2142,
      "time_per_iteration": 2.765876531600952
    },
    {
      "auxiliary_loss_clip": 0.01190534,
      "auxiliary_loss_mlp": 0.01032798,
      "balance_loss_clip": 1.05724955,
      "balance_loss_mlp": 1.02243304,
      "epoch": 0.2576805146395719,
      "flos": 22674069578880.0,
      "grad_norm": 1.7108477102943007,
      "language_loss": 0.78869939,
      "learning_rate": 3.4805773696097453e-06,
      "loss": 0.81093276,
      "num_input_tokens_seen": 45815780,
      "step": 2143,
      "time_per_iteration": 2.75534725189209
    },
    {
      "auxiliary_loss_clip": 0.01177468,
      "auxiliary_loss_mlp": 0.01038022,
      "balance_loss_clip": 1.05837083,
      "balance_loss_mlp": 1.02891445,
      "epoch": 0.25780075753021103,
      "flos": 16472225278080.0,
      "grad_norm": 2.092290207792063,
      "language_loss": 0.88063502,
      "learning_rate": 3.4800535631571874e-06,
      "loss": 0.90278983,
      "num_input_tokens_seen": 45831310,
      "step": 2144,
      "time_per_iteration": 2.8079986572265625
    },
    {
      "auxiliary_loss_clip": 0.01184843,
      "auxiliary_loss_mlp": 0.01037538,
      "balance_loss_clip": 1.05578279,
      "balance_loss_mlp": 1.02692258,
      "epoch": 0.25792100042085014,
      "flos": 22820297846400.0,
      "grad_norm": 2.4049017093568725,
      "language_loss": 0.76524204,
      "learning_rate": 3.4795295321816804e-06,
      "loss": 0.78746581,
      "num_input_tokens_seen": 45850135,
      "step": 2145,
      "time_per_iteration": 2.7470638751983643
    },
    {
      "auxiliary_loss_clip": 0.01166792,
      "auxiliary_loss_mlp": 0.01039792,
      "balance_loss_clip": 1.05238223,
      "balance_loss_mlp": 1.02982593,
      "epoch": 0.2580412433114892,
      "flos": 18697286194560.0,
      "grad_norm": 3.314422632594769,
      "language_loss": 0.91288388,
      "learning_rate": 3.47900527676272e-06,
      "loss": 0.9349497,
      "num_input_tokens_seen": 45868470,
      "step": 2146,
      "time_per_iteration": 3.6622936725616455
    },
    {
      "auxiliary_loss_clip": 0.01213274,
      "auxiliary_loss_mlp": 0.01044486,
      "balance_loss_clip": 1.06343484,
      "balance_loss_mlp": 1.03472877,
      "epoch": 0.2581614862021283,
      "flos": 14283146810880.0,
      "grad_norm": 2.274901630807928,
      "language_loss": 0.88746941,
      "learning_rate": 3.478480796979835e-06,
      "loss": 0.91004705,
      "num_input_tokens_seen": 45886355,
      "step": 2147,
      "time_per_iteration": 2.8576083183288574
    },
    {
      "auxiliary_loss_clip": 0.01175965,
      "auxiliary_loss_mlp": 0.01034403,
      "balance_loss_clip": 1.05579114,
      "balance_loss_mlp": 1.02409148,
      "epoch": 0.25828172909276736,
      "flos": 29498281856640.0,
      "grad_norm": 2.7903651755737644,
      "language_loss": 0.77905107,
      "learning_rate": 3.4779560929125894e-06,
      "loss": 0.80115473,
      "num_input_tokens_seen": 45907900,
      "step": 2148,
      "time_per_iteration": 2.7434988021850586
    },
    {
      "auxiliary_loss_clip": 0.01064798,
      "auxiliary_loss_mlp": 0.01006299,
      "balance_loss_clip": 1.02136183,
      "balance_loss_mlp": 1.00416517,
      "epoch": 0.2584019719834065,
      "flos": 67114387376640.0,
      "grad_norm": 0.6843763675756703,
      "language_loss": 0.56946301,
      "learning_rate": 3.4774311646405783e-06,
      "loss": 0.59017396,
      "num_input_tokens_seen": 45977805,
      "step": 2149,
      "time_per_iteration": 3.47347354888916
    },
    {
      "auxiliary_loss_clip": 0.01150933,
      "auxiliary_loss_mlp": 0.01031857,
      "balance_loss_clip": 1.04926896,
      "balance_loss_mlp": 1.0214144,
      "epoch": 0.2585222148740456,
      "flos": 22893555634560.0,
      "grad_norm": 5.004169448472716,
      "language_loss": 0.83967167,
      "learning_rate": 3.476906012243435e-06,
      "loss": 0.86149955,
      "num_input_tokens_seen": 45996715,
      "step": 2150,
      "time_per_iteration": 3.713287353515625
    },
    {
      "auxiliary_loss_clip": 0.01184103,
      "auxiliary_loss_mlp": 0.01038721,
      "balance_loss_clip": 1.05645382,
      "balance_loss_mlp": 1.02877927,
      "epoch": 0.25864245776468464,
      "flos": 28909202808960.0,
      "grad_norm": 1.7704039038727888,
      "language_loss": 0.81496239,
      "learning_rate": 3.476380635800824e-06,
      "loss": 0.83719063,
      "num_input_tokens_seen": 46017915,
      "step": 2151,
      "time_per_iteration": 3.8134374618530273
    },
    {
      "auxiliary_loss_clip": 0.01174656,
      "auxiliary_loss_mlp": 0.01034654,
      "balance_loss_clip": 1.0569787,
      "balance_loss_mlp": 1.02542126,
      "epoch": 0.25876270065532375,
      "flos": 14793185980800.0,
      "grad_norm": 2.194096105092921,
      "language_loss": 0.86397839,
      "learning_rate": 3.475855035392444e-06,
      "loss": 0.88607144,
      "num_input_tokens_seen": 46033235,
      "step": 2152,
      "time_per_iteration": 3.6637401580810547
    },
    {
      "auxiliary_loss_clip": 0.01134491,
      "auxiliary_loss_mlp": 0.01027614,
      "balance_loss_clip": 1.05275917,
      "balance_loss_mlp": 1.01827955,
      "epoch": 0.25888294354596286,
      "flos": 60467821810560.0,
      "grad_norm": 2.1239627315234615,
      "language_loss": 0.7138356,
      "learning_rate": 3.475329211098029e-06,
      "loss": 0.73545665,
      "num_input_tokens_seen": 46056390,
      "step": 2153,
      "time_per_iteration": 3.238281488418579
    },
    {
      "auxiliary_loss_clip": 0.01155336,
      "auxiliary_loss_mlp": 0.0103493,
      "balance_loss_clip": 1.05355394,
      "balance_loss_mlp": 1.02564371,
      "epoch": 0.2590031864366019,
      "flos": 27851166771840.0,
      "grad_norm": 1.7417133918443026,
      "language_loss": 0.82448679,
      "learning_rate": 3.4748031629973453e-06,
      "loss": 0.84638941,
      "num_input_tokens_seen": 46077120,
      "step": 2154,
      "time_per_iteration": 2.975493907928467
    },
    {
      "auxiliary_loss_clip": 0.01051174,
      "auxiliary_loss_mlp": 0.01002258,
      "balance_loss_clip": 1.02084756,
      "balance_loss_mlp": 1.00006473,
      "epoch": 0.25912342932724103,
      "flos": 62422444206720.0,
      "grad_norm": 0.9112428620346538,
      "language_loss": 0.56599599,
      "learning_rate": 3.4742768911701944e-06,
      "loss": 0.58653033,
      "num_input_tokens_seen": 46139815,
      "step": 2155,
      "time_per_iteration": 3.4819154739379883
    },
    {
      "auxiliary_loss_clip": 0.01199918,
      "auxiliary_loss_mlp": 0.01042272,
      "balance_loss_clip": 1.06041312,
      "balance_loss_mlp": 1.03215694,
      "epoch": 0.25924367221788014,
      "flos": 12378839368320.0,
      "grad_norm": 3.1095609490669505,
      "language_loss": 0.70518059,
      "learning_rate": 3.4737503956964113e-06,
      "loss": 0.72760248,
      "num_input_tokens_seen": 46152120,
      "step": 2156,
      "time_per_iteration": 2.7426395416259766
    },
    {
      "auxiliary_loss_clip": 0.0116977,
      "auxiliary_loss_mlp": 0.01033402,
      "balance_loss_clip": 1.05192232,
      "balance_loss_mlp": 1.02307296,
      "epoch": 0.2593639151085192,
      "flos": 14575208296320.0,
      "grad_norm": 2.1910608382395202,
      "language_loss": 0.6756202,
      "learning_rate": 3.473223676655865e-06,
      "loss": 0.69765192,
      "num_input_tokens_seen": 46170120,
      "step": 2157,
      "time_per_iteration": 2.7701849937438965
    },
    {
      "auxiliary_loss_clip": 0.01173262,
      "auxiliary_loss_mlp": 0.01040302,
      "balance_loss_clip": 1.05379856,
      "balance_loss_mlp": 1.02979946,
      "epoch": 0.2594841579991583,
      "flos": 15230937029760.0,
      "grad_norm": 2.4930472947072735,
      "language_loss": 0.80002081,
      "learning_rate": 3.472696734128459e-06,
      "loss": 0.82215643,
      "num_input_tokens_seen": 46187985,
      "step": 2158,
      "time_per_iteration": 2.7413103580474854
    },
    {
      "auxiliary_loss_clip": 0.01194436,
      "auxiliary_loss_mlp": 0.01038129,
      "balance_loss_clip": 1.05688751,
      "balance_loss_mlp": 1.02812159,
      "epoch": 0.2596044008897974,
      "flos": 23623583650560.0,
      "grad_norm": 2.0899533275350226,
      "language_loss": 0.75534683,
      "learning_rate": 3.4721695681941286e-06,
      "loss": 0.77767247,
      "num_input_tokens_seen": 46207025,
      "step": 2159,
      "time_per_iteration": 2.8154211044311523
    },
    {
      "auxiliary_loss_clip": 0.0117781,
      "auxiliary_loss_mlp": 0.00764978,
      "balance_loss_clip": 1.05486274,
      "balance_loss_mlp": 1.00054526,
      "epoch": 0.25972464378043647,
      "flos": 13772281628160.0,
      "grad_norm": 2.1129042815780026,
      "language_loss": 0.82275307,
      "learning_rate": 3.471642178932845e-06,
      "loss": 0.84218097,
      "num_input_tokens_seen": 46225670,
      "step": 2160,
      "time_per_iteration": 2.7799184322357178
    },
    {
      "auxiliary_loss_clip": 0.01177577,
      "auxiliary_loss_mlp": 0.01038868,
      "balance_loss_clip": 1.05226755,
      "balance_loss_mlp": 1.02931297,
      "epoch": 0.2598448866710756,
      "flos": 19573578391680.0,
      "grad_norm": 1.936514405578926,
      "language_loss": 0.89751923,
      "learning_rate": 3.471114566424613e-06,
      "loss": 0.91968369,
      "num_input_tokens_seen": 46244130,
      "step": 2161,
      "time_per_iteration": 2.901782989501953
    },
    {
      "auxiliary_loss_clip": 0.01175666,
      "auxiliary_loss_mlp": 0.01029722,
      "balance_loss_clip": 1.05627453,
      "balance_loss_mlp": 1.02007186,
      "epoch": 0.25996512956171464,
      "flos": 21653237053440.0,
      "grad_norm": 2.2103103150702235,
      "language_loss": 0.7623933,
      "learning_rate": 3.4705867307494715e-06,
      "loss": 0.78444713,
      "num_input_tokens_seen": 46263200,
      "step": 2162,
      "time_per_iteration": 2.8005387783050537
    },
    {
      "auxiliary_loss_clip": 0.01193319,
      "auxiliary_loss_mlp": 0.01037348,
      "balance_loss_clip": 1.05383658,
      "balance_loss_mlp": 1.02706075,
      "epoch": 0.26008537245235375,
      "flos": 18223480869120.0,
      "grad_norm": 4.416136885711264,
      "language_loss": 0.8449707,
      "learning_rate": 3.470058671987492e-06,
      "loss": 0.86727738,
      "num_input_tokens_seen": 46281465,
      "step": 2163,
      "time_per_iteration": 2.743229627609253
    },
    {
      "auxiliary_loss_clip": 0.0119742,
      "auxiliary_loss_mlp": 0.01033849,
      "balance_loss_clip": 1.05565071,
      "balance_loss_mlp": 1.02339423,
      "epoch": 0.26020561534299286,
      "flos": 24645385843200.0,
      "grad_norm": 1.860966638672939,
      "language_loss": 0.84184545,
      "learning_rate": 3.4695303902187805e-06,
      "loss": 0.86415815,
      "num_input_tokens_seen": 46301020,
      "step": 2164,
      "time_per_iteration": 2.7318592071533203
    },
    {
      "auxiliary_loss_clip": 0.01158897,
      "auxiliary_loss_mlp": 0.01043429,
      "balance_loss_clip": 1.0489018,
      "balance_loss_mlp": 1.03354073,
      "epoch": 0.2603258582336319,
      "flos": 25773662926080.0,
      "grad_norm": 2.087856966405713,
      "language_loss": 0.78345549,
      "learning_rate": 3.469001885523478e-06,
      "loss": 0.80547881,
      "num_input_tokens_seen": 46321740,
      "step": 2165,
      "time_per_iteration": 2.8161497116088867
    },
    {
      "auxiliary_loss_clip": 0.01207479,
      "auxiliary_loss_mlp": 0.01028191,
      "balance_loss_clip": 1.05798817,
      "balance_loss_mlp": 1.01883256,
      "epoch": 0.260446101124271,
      "flos": 28766314506240.0,
      "grad_norm": 1.8186532244428,
      "language_loss": 0.8131355,
      "learning_rate": 3.4684731579817568e-06,
      "loss": 0.83549225,
      "num_input_tokens_seen": 46342730,
      "step": 2166,
      "time_per_iteration": 2.7604758739471436
    },
    {
      "auxiliary_loss_clip": 0.01133991,
      "auxiliary_loss_mlp": 0.01038768,
      "balance_loss_clip": 1.0512712,
      "balance_loss_mlp": 1.02888536,
      "epoch": 0.26056634401491013,
      "flos": 25666757072640.0,
      "grad_norm": 1.7063066253593668,
      "language_loss": 0.76923245,
      "learning_rate": 3.4679442076738247e-06,
      "loss": 0.79096007,
      "num_input_tokens_seen": 46362445,
      "step": 2167,
      "time_per_iteration": 2.871920585632324
    },
    {
      "auxiliary_loss_clip": 0.01211173,
      "auxiliary_loss_mlp": 0.01035963,
      "balance_loss_clip": 1.05940676,
      "balance_loss_mlp": 1.02558017,
      "epoch": 0.2606865869055492,
      "flos": 27052765217280.0,
      "grad_norm": 1.860217774491937,
      "language_loss": 0.83401871,
      "learning_rate": 3.4674150346799245e-06,
      "loss": 0.85649002,
      "num_input_tokens_seen": 46382145,
      "step": 2168,
      "time_per_iteration": 2.69301700592041
    },
    {
      "auxiliary_loss_clip": 0.01180665,
      "auxiliary_loss_mlp": 0.01030314,
      "balance_loss_clip": 1.05592287,
      "balance_loss_mlp": 1.02005029,
      "epoch": 0.2608068297961883,
      "flos": 17712615686400.0,
      "grad_norm": 2.1296615444067273,
      "language_loss": 0.80305302,
      "learning_rate": 3.4668856390803295e-06,
      "loss": 0.82516289,
      "num_input_tokens_seen": 46400025,
      "step": 2169,
      "time_per_iteration": 2.7036795616149902
    },
    {
      "auxiliary_loss_clip": 0.01182633,
      "auxiliary_loss_mlp": 0.01040415,
      "balance_loss_clip": 1.05450833,
      "balance_loss_mlp": 1.0300976,
      "epoch": 0.2609270726868274,
      "flos": 18551632544640.0,
      "grad_norm": 2.101467351468322,
      "language_loss": 0.90076733,
      "learning_rate": 3.4663560209553495e-06,
      "loss": 0.92299783,
      "num_input_tokens_seen": 46418090,
      "step": 2170,
      "time_per_iteration": 2.741187334060669
    },
    {
      "auxiliary_loss_clip": 0.01167436,
      "auxiliary_loss_mlp": 0.01035671,
      "balance_loss_clip": 1.05249321,
      "balance_loss_mlp": 1.02555037,
      "epoch": 0.26104731557746647,
      "flos": 21835699165440.0,
      "grad_norm": 1.5820962747393674,
      "language_loss": 0.79232526,
      "learning_rate": 3.4658261803853267e-06,
      "loss": 0.81435627,
      "num_input_tokens_seen": 46436015,
      "step": 2171,
      "time_per_iteration": 2.7068581581115723
    },
    {
      "auxiliary_loss_clip": 0.01175742,
      "auxiliary_loss_mlp": 0.01028674,
      "balance_loss_clip": 1.05617809,
      "balance_loss_mlp": 1.01877928,
      "epoch": 0.2611675584681056,
      "flos": 21689650465920.0,
      "grad_norm": 2.139791024455814,
      "language_loss": 0.80061936,
      "learning_rate": 3.4652961174506383e-06,
      "loss": 0.82266349,
      "num_input_tokens_seen": 46455885,
      "step": 2172,
      "time_per_iteration": 3.5975639820098877
    },
    {
      "auxiliary_loss_clip": 0.0108444,
      "auxiliary_loss_mlp": 0.01005786,
      "balance_loss_clip": 1.01999354,
      "balance_loss_mlp": 1.00323474,
      "epoch": 0.2612878013587447,
      "flos": 71862101389440.0,
      "grad_norm": 0.9615243342852774,
      "language_loss": 0.58069944,
      "learning_rate": 3.464765832231694e-06,
      "loss": 0.60160172,
      "num_input_tokens_seen": 46510050,
      "step": 2173,
      "time_per_iteration": 3.298933982849121
    },
    {
      "auxiliary_loss_clip": 0.01195969,
      "auxiliary_loss_mlp": 0.01036208,
      "balance_loss_clip": 1.05890679,
      "balance_loss_mlp": 1.02623582,
      "epoch": 0.26140804424938374,
      "flos": 20227511445120.0,
      "grad_norm": 2.2990585542003266,
      "language_loss": 0.70660454,
      "learning_rate": 3.4642353248089373e-06,
      "loss": 0.7289263,
      "num_input_tokens_seen": 46528810,
      "step": 2174,
      "time_per_iteration": 2.8091073036193848
    },
    {
      "auxiliary_loss_clip": 0.01171861,
      "auxiliary_loss_mlp": 0.01036182,
      "balance_loss_clip": 1.05329525,
      "balance_loss_mlp": 1.02585888,
      "epoch": 0.26152828714002285,
      "flos": 25557085872000.0,
      "grad_norm": 1.8034301111642723,
      "language_loss": 0.80496556,
      "learning_rate": 3.463704595262846e-06,
      "loss": 0.82704604,
      "num_input_tokens_seen": 46549690,
      "step": 2175,
      "time_per_iteration": 3.8178837299346924
    },
    {
      "auxiliary_loss_clip": 0.01164182,
      "auxiliary_loss_mlp": 0.01030556,
      "balance_loss_clip": 1.05607224,
      "balance_loss_mlp": 1.02093029,
      "epoch": 0.26164853003066196,
      "flos": 25446516831360.0,
      "grad_norm": 1.8763549566028066,
      "language_loss": 0.70642006,
      "learning_rate": 3.463173643673931e-06,
      "loss": 0.72836745,
      "num_input_tokens_seen": 46572215,
      "step": 2176,
      "time_per_iteration": 2.7861552238464355
    },
    {
      "auxiliary_loss_clip": 0.01094338,
      "auxiliary_loss_mlp": 0.01004648,
      "balance_loss_clip": 1.02289259,
      "balance_loss_mlp": 1.00233555,
      "epoch": 0.261768772921301,
      "flos": 53944580568960.0,
      "grad_norm": 0.8987036025135968,
      "language_loss": 0.63464659,
      "learning_rate": 3.4626424701227387e-06,
      "loss": 0.65563643,
      "num_input_tokens_seen": 46627275,
      "step": 2177,
      "time_per_iteration": 4.102824926376343
    },
    {
      "auxiliary_loss_clip": 0.0110583,
      "auxiliary_loss_mlp": 0.01009487,
      "balance_loss_clip": 1.02470589,
      "balance_loss_mlp": 1.00707853,
      "epoch": 0.26188901581194013,
      "flos": 70687606481280.0,
      "grad_norm": 0.8334593467847138,
      "language_loss": 0.55782431,
      "learning_rate": 3.4621110746898452e-06,
      "loss": 0.57897747,
      "num_input_tokens_seen": 46695135,
      "step": 2178,
      "time_per_iteration": 4.3226542472839355
    },
    {
      "auxiliary_loss_clip": 0.01198421,
      "auxiliary_loss_mlp": 0.01035405,
      "balance_loss_clip": 1.05922937,
      "balance_loss_mlp": 1.02536762,
      "epoch": 0.2620092587025792,
      "flos": 21069580959360.0,
      "grad_norm": 1.840232647129895,
      "language_loss": 0.75102186,
      "learning_rate": 3.4615794574558654e-06,
      "loss": 0.77336013,
      "num_input_tokens_seen": 46714145,
      "step": 2179,
      "time_per_iteration": 2.7355494499206543
    },
    {
      "auxiliary_loss_clip": 0.01175348,
      "auxiliary_loss_mlp": 0.01030893,
      "balance_loss_clip": 1.05278587,
      "balance_loss_mlp": 1.02123666,
      "epoch": 0.2621295015932183,
      "flos": 18369601395840.0,
      "grad_norm": 2.4706412017102948,
      "language_loss": 0.84754932,
      "learning_rate": 3.4610476185014436e-06,
      "loss": 0.86961174,
      "num_input_tokens_seen": 46731405,
      "step": 2180,
      "time_per_iteration": 2.6961328983306885
    },
    {
      "auxiliary_loss_clip": 0.01208733,
      "auxiliary_loss_mlp": 0.01041557,
      "balance_loss_clip": 1.05797112,
      "balance_loss_mlp": 1.0313046,
      "epoch": 0.2622497444838574,
      "flos": 23659997063040.0,
      "grad_norm": 1.9022042559792318,
      "language_loss": 0.79593945,
      "learning_rate": 3.4605155579072597e-06,
      "loss": 0.81844229,
      "num_input_tokens_seen": 46751260,
      "step": 2181,
      "time_per_iteration": 2.720080614089966
    },
    {
      "auxiliary_loss_clip": 0.0114067,
      "auxiliary_loss_mlp": 0.01036271,
      "balance_loss_clip": 1.04921651,
      "balance_loss_mlp": 1.02683568,
      "epoch": 0.26236998737449646,
      "flos": 22123810154880.0,
      "grad_norm": 1.7972152796016299,
      "language_loss": 0.7117247,
      "learning_rate": 3.459983275754027e-06,
      "loss": 0.7334941,
      "num_input_tokens_seen": 46770155,
      "step": 2182,
      "time_per_iteration": 2.837202787399292
    },
    {
      "auxiliary_loss_clip": 0.01207463,
      "auxiliary_loss_mlp": 0.01032889,
      "balance_loss_clip": 1.0573355,
      "balance_loss_mlp": 1.02303624,
      "epoch": 0.26249023026513557,
      "flos": 17895185539200.0,
      "grad_norm": 3.974735705417144,
      "language_loss": 0.79529798,
      "learning_rate": 3.4594507721224918e-06,
      "loss": 0.81770152,
      "num_input_tokens_seen": 46788805,
      "step": 2183,
      "time_per_iteration": 2.668518304824829
    },
    {
      "auxiliary_loss_clip": 0.01181747,
      "auxiliary_loss_mlp": 0.01041515,
      "balance_loss_clip": 1.05528641,
      "balance_loss_mlp": 1.03237212,
      "epoch": 0.2626104731557747,
      "flos": 18332936588160.0,
      "grad_norm": 1.6536594359770505,
      "language_loss": 0.82277644,
      "learning_rate": 3.4589180470934353e-06,
      "loss": 0.84500903,
      "num_input_tokens_seen": 46808670,
      "step": 2184,
      "time_per_iteration": 2.7255167961120605
    },
    {
      "auxiliary_loss_clip": 0.01197688,
      "auxiliary_loss_mlp": 0.0103631,
      "balance_loss_clip": 1.05397189,
      "balance_loss_mlp": 1.02615309,
      "epoch": 0.26273071604641374,
      "flos": 19317714837120.0,
      "grad_norm": 2.1611291128342334,
      "language_loss": 0.7702406,
      "learning_rate": 3.4583851007476713e-06,
      "loss": 0.7925806,
      "num_input_tokens_seen": 46827140,
      "step": 2185,
      "time_per_iteration": 2.7311878204345703
    },
    {
      "auxiliary_loss_clip": 0.01165102,
      "auxiliary_loss_mlp": 0.01040107,
      "balance_loss_clip": 1.05444121,
      "balance_loss_mlp": 1.0299207,
      "epoch": 0.26285095893705285,
      "flos": 18327477720960.0,
      "grad_norm": 2.135956953688338,
      "language_loss": 0.68645656,
      "learning_rate": 3.4578519331660464e-06,
      "loss": 0.70850861,
      "num_input_tokens_seen": 46844135,
      "step": 2186,
      "time_per_iteration": 2.7207775115966797
    },
    {
      "auxiliary_loss_clip": 0.01190914,
      "auxiliary_loss_mlp": 0.01034134,
      "balance_loss_clip": 1.05875897,
      "balance_loss_mlp": 1.0247643,
      "epoch": 0.26297120182769196,
      "flos": 20193827466240.0,
      "grad_norm": 17.809857089328183,
      "language_loss": 0.82067966,
      "learning_rate": 3.4573185444294426e-06,
      "loss": 0.84293014,
      "num_input_tokens_seen": 46862500,
      "step": 2187,
      "time_per_iteration": 2.7988717555999756
    },
    {
      "auxiliary_loss_clip": 0.01176632,
      "auxiliary_loss_mlp": 0.00765028,
      "balance_loss_clip": 1.0542202,
      "balance_loss_mlp": 1.00057769,
      "epoch": 0.263091444718331,
      "flos": 22418421505920.0,
      "grad_norm": 2.0510791474024535,
      "language_loss": 0.7874338,
      "learning_rate": 3.456784934618774e-06,
      "loss": 0.80685037,
      "num_input_tokens_seen": 46883665,
      "step": 2188,
      "time_per_iteration": 2.795130968093872
    },
    {
      "auxiliary_loss_clip": 0.01174268,
      "auxiliary_loss_mlp": 0.01039173,
      "balance_loss_clip": 1.05261922,
      "balance_loss_mlp": 1.02937949,
      "epoch": 0.2632116876089701,
      "flos": 19024827338880.0,
      "grad_norm": 2.102418552233927,
      "language_loss": 0.796381,
      "learning_rate": 3.4562511038149897e-06,
      "loss": 0.8185153,
      "num_input_tokens_seen": 46899160,
      "step": 2189,
      "time_per_iteration": 2.693253993988037
    },
    {
      "auxiliary_loss_clip": 0.01052286,
      "auxiliary_loss_mlp": 0.01008738,
      "balance_loss_clip": 1.02056551,
      "balance_loss_mlp": 1.0066514,
      "epoch": 0.26333193049960923,
      "flos": 67308054531840.0,
      "grad_norm": 0.8593881775456508,
      "language_loss": 0.57728714,
      "learning_rate": 3.4557170520990705e-06,
      "loss": 0.59789735,
      "num_input_tokens_seen": 46959835,
      "step": 2190,
      "time_per_iteration": 3.4310836791992188
    },
    {
      "auxiliary_loss_clip": 0.01188511,
      "auxiliary_loss_mlp": 0.01038746,
      "balance_loss_clip": 1.05677354,
      "balance_loss_mlp": 1.02988839,
      "epoch": 0.2634521733902483,
      "flos": 25048806468480.0,
      "grad_norm": 2.3401930449806394,
      "language_loss": 0.86477405,
      "learning_rate": 3.4551827795520324e-06,
      "loss": 0.88704669,
      "num_input_tokens_seen": 46982720,
      "step": 2191,
      "time_per_iteration": 2.7360470294952393
    },
    {
      "auxiliary_loss_clip": 0.01190251,
      "auxiliary_loss_mlp": 0.0103506,
      "balance_loss_clip": 1.05276489,
      "balance_loss_mlp": 1.02530289,
      "epoch": 0.2635724162808874,
      "flos": 20594985534720.0,
      "grad_norm": 1.7063172322199287,
      "language_loss": 0.85233325,
      "learning_rate": 3.4546482862549226e-06,
      "loss": 0.87458634,
      "num_input_tokens_seen": 47003035,
      "step": 2192,
      "time_per_iteration": 2.7484350204467773
    },
    {
      "auxiliary_loss_clip": 0.01159714,
      "auxiliary_loss_mlp": 0.0103439,
      "balance_loss_clip": 1.05284488,
      "balance_loss_mlp": 1.02428746,
      "epoch": 0.2636926591715265,
      "flos": 19244636616960.0,
      "grad_norm": 2.231681787812652,
      "language_loss": 0.79020196,
      "learning_rate": 3.4541135722888253e-06,
      "loss": 0.81214297,
      "num_input_tokens_seen": 47019625,
      "step": 2193,
      "time_per_iteration": 2.7736289501190186
    },
    {
      "auxiliary_loss_clip": 0.01208724,
      "auxiliary_loss_mlp": 0.01037253,
      "balance_loss_clip": 1.05882609,
      "balance_loss_mlp": 1.02780557,
      "epoch": 0.26381290206216557,
      "flos": 28804882734720.0,
      "grad_norm": 2.1465003968563545,
      "language_loss": 0.80461538,
      "learning_rate": 3.453578637734854e-06,
      "loss": 0.82707512,
      "num_input_tokens_seen": 47040815,
      "step": 2194,
      "time_per_iteration": 2.681612014770508
    },
    {
      "auxiliary_loss_clip": 0.01213133,
      "auxiliary_loss_mlp": 0.01038825,
      "balance_loss_clip": 1.06305838,
      "balance_loss_mlp": 1.02930617,
      "epoch": 0.2639331449528047,
      "flos": 25008909436800.0,
      "grad_norm": 1.6745718268148027,
      "language_loss": 0.78559887,
      "learning_rate": 3.4530434826741605e-06,
      "loss": 0.80811846,
      "num_input_tokens_seen": 47061755,
      "step": 2195,
      "time_per_iteration": 2.6600494384765625
    },
    {
      "auxiliary_loss_clip": 0.01173468,
      "auxiliary_loss_mlp": 0.01033287,
      "balance_loss_clip": 1.05574429,
      "balance_loss_mlp": 1.02404785,
      "epoch": 0.26405338784344373,
      "flos": 46535775465600.0,
      "grad_norm": 1.620102685890706,
      "language_loss": 0.69317663,
      "learning_rate": 3.452508107187926e-06,
      "loss": 0.71524417,
      "num_input_tokens_seen": 47085130,
      "step": 2196,
      "time_per_iteration": 3.0667669773101807
    },
    {
      "auxiliary_loss_clip": 0.01131302,
      "auxiliary_loss_mlp": 0.01035448,
      "balance_loss_clip": 1.04663563,
      "balance_loss_mlp": 1.02534461,
      "epoch": 0.26417363073408284,
      "flos": 21179467641600.0,
      "grad_norm": 1.9505966582222527,
      "language_loss": 0.77490026,
      "learning_rate": 3.451972511357366e-06,
      "loss": 0.7965678,
      "num_input_tokens_seen": 47104675,
      "step": 2197,
      "time_per_iteration": 3.7696046829223633
    },
    {
      "auxiliary_loss_clip": 0.01191808,
      "auxiliary_loss_mlp": 0.01039134,
      "balance_loss_clip": 1.0587039,
      "balance_loss_mlp": 1.02941227,
      "epoch": 0.26429387362472195,
      "flos": 22674751937280.0,
      "grad_norm": 1.7553847475644808,
      "language_loss": 0.85581696,
      "learning_rate": 3.45143669526373e-06,
      "loss": 0.87812638,
      "num_input_tokens_seen": 47124435,
      "step": 2198,
      "time_per_iteration": 2.949803352355957
    },
    {
      "auxiliary_loss_clip": 0.01083148,
      "auxiliary_loss_mlp": 0.01003335,
      "balance_loss_clip": 1.02117181,
      "balance_loss_mlp": 1.00087929,
      "epoch": 0.264414116515361,
      "flos": 67180534272000.0,
      "grad_norm": 0.7858887255874223,
      "language_loss": 0.63203478,
      "learning_rate": 3.450900658988302e-06,
      "loss": 0.65289956,
      "num_input_tokens_seen": 47185985,
      "step": 2199,
      "time_per_iteration": 3.2081148624420166
    },
    {
      "auxiliary_loss_clip": 0.0116877,
      "auxiliary_loss_mlp": 0.01030559,
      "balance_loss_clip": 1.05416107,
      "balance_loss_mlp": 1.02042007,
      "epoch": 0.2645343594060001,
      "flos": 25664709997440.0,
      "grad_norm": 1.925851027586999,
      "language_loss": 0.77978539,
      "learning_rate": 3.450364402612397e-06,
      "loss": 0.80177867,
      "num_input_tokens_seen": 47203140,
      "step": 2200,
      "time_per_iteration": 2.9211459159851074
    },
    {
      "auxiliary_loss_clip": 0.0117188,
      "auxiliary_loss_mlp": 0.01034775,
      "balance_loss_clip": 1.05399227,
      "balance_loss_mlp": 1.02505374,
      "epoch": 0.26465460229663923,
      "flos": 22491822948480.0,
      "grad_norm": 2.068813562591618,
      "language_loss": 0.84078729,
      "learning_rate": 3.449827926217366e-06,
      "loss": 0.86285383,
      "num_input_tokens_seen": 47222575,
      "step": 2201,
      "time_per_iteration": 3.695934295654297
    },
    {
      "auxiliary_loss_clip": 0.01178164,
      "auxiliary_loss_mlp": 0.01031537,
      "balance_loss_clip": 1.05020833,
      "balance_loss_mlp": 1.02198243,
      "epoch": 0.2647748451872783,
      "flos": 29388036038400.0,
      "grad_norm": 3.046940668421117,
      "language_loss": 0.80871421,
      "learning_rate": 3.449291229884591e-06,
      "loss": 0.83081126,
      "num_input_tokens_seen": 47243815,
      "step": 2202,
      "time_per_iteration": 2.7967514991760254
    },
    {
      "auxiliary_loss_clip": 0.01166177,
      "auxiliary_loss_mlp": 0.01039522,
      "balance_loss_clip": 1.0522182,
      "balance_loss_mlp": 1.02996111,
      "epoch": 0.2648950880779174,
      "flos": 26797799502720.0,
      "grad_norm": 2.1918455144218814,
      "language_loss": 0.86615229,
      "learning_rate": 3.4487543136954887e-06,
      "loss": 0.88820922,
      "num_input_tokens_seen": 47263435,
      "step": 2203,
      "time_per_iteration": 4.118595838546753
    },
    {
      "auxiliary_loss_clip": 0.01161024,
      "auxiliary_loss_mlp": 0.01038383,
      "balance_loss_clip": 1.05245149,
      "balance_loss_mlp": 1.02973485,
      "epoch": 0.2650153309685565,
      "flos": 28841008838400.0,
      "grad_norm": 3.1364960919281066,
      "language_loss": 0.90942639,
      "learning_rate": 3.448217177731509e-06,
      "loss": 0.93142051,
      "num_input_tokens_seen": 47283920,
      "step": 2204,
      "time_per_iteration": 3.7646875381469727
    },
    {
      "auxiliary_loss_clip": 0.01173943,
      "auxiliary_loss_mlp": 0.01034572,
      "balance_loss_clip": 1.05686116,
      "balance_loss_mlp": 1.02533329,
      "epoch": 0.26513557385919556,
      "flos": 20303247271680.0,
      "grad_norm": 2.3170343535227738,
      "language_loss": 0.77963054,
      "learning_rate": 3.4476798220741348e-06,
      "loss": 0.80171561,
      "num_input_tokens_seen": 47302800,
      "step": 2205,
      "time_per_iteration": 2.6515469551086426
    },
    {
      "auxiliary_loss_clip": 0.0120555,
      "auxiliary_loss_mlp": 0.01030253,
      "balance_loss_clip": 1.05951381,
      "balance_loss_mlp": 1.02199793,
      "epoch": 0.26525581674983467,
      "flos": 17676274101120.0,
      "grad_norm": 1.8979961268120034,
      "language_loss": 0.78407425,
      "learning_rate": 3.4471422468048826e-06,
      "loss": 0.80643231,
      "num_input_tokens_seen": 47321525,
      "step": 2206,
      "time_per_iteration": 2.7334678173065186
    },
    {
      "auxiliary_loss_clip": 0.01187018,
      "auxiliary_loss_mlp": 0.01029155,
      "balance_loss_clip": 1.05805933,
      "balance_loss_mlp": 1.0194037,
      "epoch": 0.2653760596404738,
      "flos": 26833746038400.0,
      "grad_norm": 2.277201840629732,
      "language_loss": 0.72842371,
      "learning_rate": 3.4466044520053022e-06,
      "loss": 0.75058544,
      "num_input_tokens_seen": 47340530,
      "step": 2207,
      "time_per_iteration": 2.6941213607788086
    },
    {
      "auxiliary_loss_clip": 0.01162428,
      "auxiliary_loss_mlp": 0.01029874,
      "balance_loss_clip": 1.0503912,
      "balance_loss_mlp": 1.02028346,
      "epoch": 0.26549630253111284,
      "flos": 22782160581120.0,
      "grad_norm": 1.8891307797456605,
      "language_loss": 0.60347766,
      "learning_rate": 3.446066437756977e-06,
      "loss": 0.62540066,
      "num_input_tokens_seen": 47359735,
      "step": 2208,
      "time_per_iteration": 2.639209508895874
    },
    {
      "auxiliary_loss_clip": 0.01174516,
      "auxiliary_loss_mlp": 0.01028589,
      "balance_loss_clip": 1.05377853,
      "balance_loss_mlp": 1.01922512,
      "epoch": 0.26561654542175195,
      "flos": 23550002640000.0,
      "grad_norm": 3.651381689863993,
      "language_loss": 0.75389469,
      "learning_rate": 3.4455282041415224e-06,
      "loss": 0.7759257,
      "num_input_tokens_seen": 47378945,
      "step": 2209,
      "time_per_iteration": 2.74253249168396
    },
    {
      "auxiliary_loss_clip": 0.01165353,
      "auxiliary_loss_mlp": 0.01025909,
      "balance_loss_clip": 1.05466533,
      "balance_loss_mlp": 1.01708794,
      "epoch": 0.265736788312391,
      "flos": 26906680604160.0,
      "grad_norm": 2.66029368421958,
      "language_loss": 0.87474716,
      "learning_rate": 3.4449897512405894e-06,
      "loss": 0.89665973,
      "num_input_tokens_seen": 47398095,
      "step": 2210,
      "time_per_iteration": 2.8266594409942627
    },
    {
      "auxiliary_loss_clip": 0.0112563,
      "auxiliary_loss_mlp": 0.00764771,
      "balance_loss_clip": 1.04431343,
      "balance_loss_mlp": 1.00049996,
      "epoch": 0.2658570312030301,
      "flos": 23477139901440.0,
      "grad_norm": 4.062567572063346,
      "language_loss": 0.75333297,
      "learning_rate": 3.444451079135859e-06,
      "loss": 0.77223694,
      "num_input_tokens_seen": 47417605,
      "step": 2211,
      "time_per_iteration": 2.7900304794311523
    },
    {
      "auxiliary_loss_clip": 0.0113752,
      "auxiliary_loss_mlp": 0.00765521,
      "balance_loss_clip": 1.04547262,
      "balance_loss_mlp": 1.00056267,
      "epoch": 0.2659772740936692,
      "flos": 21866402315520.0,
      "grad_norm": 2.343494770971865,
      "language_loss": 0.74437428,
      "learning_rate": 3.4439121879090493e-06,
      "loss": 0.76340473,
      "num_input_tokens_seen": 47435385,
      "step": 2212,
      "time_per_iteration": 2.788841724395752
    },
    {
      "auxiliary_loss_clip": 0.01180068,
      "auxiliary_loss_mlp": 0.01029973,
      "balance_loss_clip": 1.05415487,
      "balance_loss_mlp": 1.01999485,
      "epoch": 0.2660975169843083,
      "flos": 19793100360960.0,
      "grad_norm": 2.26005472447892,
      "language_loss": 0.83260524,
      "learning_rate": 3.4433730776419082e-06,
      "loss": 0.85470569,
      "num_input_tokens_seen": 47454310,
      "step": 2213,
      "time_per_iteration": 2.7120585441589355
    },
    {
      "auxiliary_loss_clip": 0.01194425,
      "auxiliary_loss_mlp": 0.00765094,
      "balance_loss_clip": 1.05515933,
      "balance_loss_mlp": 1.00060618,
      "epoch": 0.2662177598749474,
      "flos": 29018981750400.0,
      "grad_norm": 3.7683472434301657,
      "language_loss": 0.8070997,
      "learning_rate": 3.4428337484162183e-06,
      "loss": 0.82669485,
      "num_input_tokens_seen": 47475120,
      "step": 2214,
      "time_per_iteration": 2.6899020671844482
    },
    {
      "auxiliary_loss_clip": 0.01169915,
      "auxiliary_loss_mlp": 0.01033955,
      "balance_loss_clip": 1.05018306,
      "balance_loss_mlp": 1.02436447,
      "epoch": 0.2663380027655865,
      "flos": 21762549118080.0,
      "grad_norm": 2.4714374393625023,
      "language_loss": 0.84697616,
      "learning_rate": 3.442294200313797e-06,
      "loss": 0.86901486,
      "num_input_tokens_seen": 47493150,
      "step": 2215,
      "time_per_iteration": 2.771312713623047
    },
    {
      "auxiliary_loss_clip": 0.0110376,
      "auxiliary_loss_mlp": 0.01002996,
      "balance_loss_clip": 1.02397573,
      "balance_loss_mlp": 1.00074315,
      "epoch": 0.26645824565622556,
      "flos": 66980333819520.0,
      "grad_norm": 0.7580198241322796,
      "language_loss": 0.5273037,
      "learning_rate": 3.4417544334164916e-06,
      "loss": 0.54837132,
      "num_input_tokens_seen": 47557295,
      "step": 2216,
      "time_per_iteration": 3.213793992996216
    },
    {
      "auxiliary_loss_clip": 0.01163105,
      "auxiliary_loss_mlp": 0.01026035,
      "balance_loss_clip": 1.05635834,
      "balance_loss_mlp": 1.01627803,
      "epoch": 0.26657848854686467,
      "flos": 25264198373760.0,
      "grad_norm": 1.6001771199829078,
      "language_loss": 0.77367234,
      "learning_rate": 3.4412144478061854e-06,
      "loss": 0.7955637,
      "num_input_tokens_seen": 47579705,
      "step": 2217,
      "time_per_iteration": 2.823359251022339
    },
    {
      "auxiliary_loss_clip": 0.01099554,
      "auxiliary_loss_mlp": 0.01045631,
      "balance_loss_clip": 1.04512,
      "balance_loss_mlp": 1.03568304,
      "epoch": 0.2666987314375038,
      "flos": 23696769611520.0,
      "grad_norm": 2.318847722805783,
      "language_loss": 0.75470817,
      "learning_rate": 3.4406742435647925e-06,
      "loss": 0.77616,
      "num_input_tokens_seen": 47599770,
      "step": 2218,
      "time_per_iteration": 3.0209999084472656
    },
    {
      "auxiliary_loss_clip": 0.01191854,
      "auxiliary_loss_mlp": 0.01033939,
      "balance_loss_clip": 1.05799127,
      "balance_loss_mlp": 1.02515936,
      "epoch": 0.26681897432814283,
      "flos": 27048958375680.0,
      "grad_norm": 2.203887392903152,
      "language_loss": 0.7875753,
      "learning_rate": 3.440133820774263e-06,
      "loss": 0.80983323,
      "num_input_tokens_seen": 47619580,
      "step": 2219,
      "time_per_iteration": 2.919261932373047
    },
    {
      "auxiliary_loss_clip": 0.01180266,
      "auxiliary_loss_mlp": 0.01037246,
      "balance_loss_clip": 1.05545056,
      "balance_loss_mlp": 1.02709556,
      "epoch": 0.26693921721878194,
      "flos": 28985944216320.0,
      "grad_norm": 2.1575144226540366,
      "language_loss": 0.82079363,
      "learning_rate": 3.439593179516578e-06,
      "loss": 0.84296876,
      "num_input_tokens_seen": 47639490,
      "step": 2220,
      "time_per_iteration": 2.717928886413574
    },
    {
      "auxiliary_loss_clip": 0.01183341,
      "auxiliary_loss_mlp": 0.01032399,
      "balance_loss_clip": 1.05661511,
      "balance_loss_mlp": 1.02225447,
      "epoch": 0.26705946010942105,
      "flos": 21507834798720.0,
      "grad_norm": 1.8409073249817467,
      "language_loss": 0.81362921,
      "learning_rate": 3.4390523198737524e-06,
      "loss": 0.83578658,
      "num_input_tokens_seen": 47658650,
      "step": 2221,
      "time_per_iteration": 2.6761701107025146
    },
    {
      "auxiliary_loss_clip": 0.01208861,
      "auxiliary_loss_mlp": 0.00764551,
      "balance_loss_clip": 1.05828381,
      "balance_loss_mlp": 1.00053918,
      "epoch": 0.2671797030000601,
      "flos": 21471277731840.0,
      "grad_norm": 2.208974590899092,
      "language_loss": 0.73598623,
      "learning_rate": 3.4385112419278333e-06,
      "loss": 0.75572038,
      "num_input_tokens_seen": 47679875,
      "step": 2222,
      "time_per_iteration": 2.6392171382904053
    },
    {
      "auxiliary_loss_clip": 0.01094371,
      "auxiliary_loss_mlp": 0.01003475,
      "balance_loss_clip": 1.0234437,
      "balance_loss_mlp": 1.00135362,
      "epoch": 0.2672999458906992,
      "flos": 64189929767040.0,
      "grad_norm": 0.7903585930415805,
      "language_loss": 0.64762592,
      "learning_rate": 3.4379699457609033e-06,
      "loss": 0.66860437,
      "num_input_tokens_seen": 47737700,
      "step": 2223,
      "time_per_iteration": 4.2064690589904785
    },
    {
      "auxiliary_loss_clip": 0.01167187,
      "auxiliary_loss_mlp": 0.01038381,
      "balance_loss_clip": 1.05111909,
      "balance_loss_mlp": 1.02864802,
      "epoch": 0.26742018878133833,
      "flos": 16909042573440.0,
      "grad_norm": 3.5092957980360606,
      "language_loss": 0.90148485,
      "learning_rate": 3.4374284314550755e-06,
      "loss": 0.92354047,
      "num_input_tokens_seen": 47756740,
      "step": 2224,
      "time_per_iteration": 2.7145373821258545
    },
    {
      "auxiliary_loss_clip": 0.01206341,
      "auxiliary_loss_mlp": 0.01033135,
      "balance_loss_clip": 1.0579654,
      "balance_loss_mlp": 1.02369392,
      "epoch": 0.2675404316719774,
      "flos": 20667560964480.0,
      "grad_norm": 5.600253131259644,
      "language_loss": 0.81072879,
      "learning_rate": 3.436886699092498e-06,
      "loss": 0.83312356,
      "num_input_tokens_seen": 47775255,
      "step": 2225,
      "time_per_iteration": 2.6467885971069336
    },
    {
      "auxiliary_loss_clip": 0.01209723,
      "auxiliary_loss_mlp": 0.01033582,
      "balance_loss_clip": 1.05911064,
      "balance_loss_mlp": 1.02330661,
      "epoch": 0.2676606745626165,
      "flos": 17485013157120.0,
      "grad_norm": 3.6126822767378806,
      "language_loss": 0.71566981,
      "learning_rate": 3.4363447487553502e-06,
      "loss": 0.73810285,
      "num_input_tokens_seen": 47788570,
      "step": 2226,
      "time_per_iteration": 2.5871429443359375
    },
    {
      "auxiliary_loss_clip": 0.01174831,
      "auxiliary_loss_mlp": 0.01030288,
      "balance_loss_clip": 1.05610907,
      "balance_loss_mlp": 1.02025664,
      "epoch": 0.26778091745325555,
      "flos": 27852675143040.0,
      "grad_norm": 2.9344914465544645,
      "language_loss": 0.7838732,
      "learning_rate": 3.4358025805258455e-06,
      "loss": 0.80592442,
      "num_input_tokens_seen": 47808275,
      "step": 2227,
      "time_per_iteration": 3.6969895362854004
    },
    {
      "auxiliary_loss_clip": 0.01151781,
      "auxiliary_loss_mlp": 0.01033916,
      "balance_loss_clip": 1.05118525,
      "balance_loss_mlp": 1.02461147,
      "epoch": 0.26790116034389466,
      "flos": 20955995176320.0,
      "grad_norm": 2.074283011053654,
      "language_loss": 0.83585846,
      "learning_rate": 3.435260194486232e-06,
      "loss": 0.85771543,
      "num_input_tokens_seen": 47826245,
      "step": 2228,
      "time_per_iteration": 2.8312504291534424
    },
    {
      "auxiliary_loss_clip": 0.01175381,
      "auxiliary_loss_mlp": 0.01030128,
      "balance_loss_clip": 1.05546379,
      "balance_loss_mlp": 1.01988184,
      "epoch": 0.2680214032345338,
      "flos": 18040659621120.0,
      "grad_norm": 2.623568291515022,
      "language_loss": 0.82777762,
      "learning_rate": 3.4347175907187875e-06,
      "loss": 0.84983277,
      "num_input_tokens_seen": 47843235,
      "step": 2229,
      "time_per_iteration": 4.031090497970581
    },
    {
      "auxiliary_loss_clip": 0.01187474,
      "auxiliary_loss_mlp": 0.01040285,
      "balance_loss_clip": 1.05483842,
      "balance_loss_mlp": 1.03117788,
      "epoch": 0.26814164612517283,
      "flos": 22419427086720.0,
      "grad_norm": 2.2295506647598398,
      "language_loss": 0.87935054,
      "learning_rate": 3.4341747693058254e-06,
      "loss": 0.90162814,
      "num_input_tokens_seen": 47861710,
      "step": 2230,
      "time_per_iteration": 3.563784599304199
    },
    {
      "auxiliary_loss_clip": 0.01096257,
      "auxiliary_loss_mlp": 0.01027787,
      "balance_loss_clip": 1.0443368,
      "balance_loss_mlp": 1.01838779,
      "epoch": 0.26826188901581194,
      "flos": 35627371159680.0,
      "grad_norm": 1.8285461580452447,
      "language_loss": 0.77447498,
      "learning_rate": 3.4336317303296916e-06,
      "loss": 0.79571545,
      "num_input_tokens_seen": 47882685,
      "step": 2231,
      "time_per_iteration": 3.4041006565093994
    },
    {
      "auxiliary_loss_clip": 0.01185438,
      "auxiliary_loss_mlp": 0.01039052,
      "balance_loss_clip": 1.05485821,
      "balance_loss_mlp": 1.0299325,
      "epoch": 0.26838213190645105,
      "flos": 17639788861440.0,
      "grad_norm": 2.28244769017708,
      "language_loss": 0.75351381,
      "learning_rate": 3.4330884738727635e-06,
      "loss": 0.77575874,
      "num_input_tokens_seen": 47900860,
      "step": 2232,
      "time_per_iteration": 2.847151517868042
    },
    {
      "auxiliary_loss_clip": 0.01135259,
      "auxiliary_loss_mlp": 0.01030107,
      "balance_loss_clip": 1.05008101,
      "balance_loss_mlp": 1.02101183,
      "epoch": 0.2685023747970901,
      "flos": 22674823764480.0,
      "grad_norm": 2.4463649638151583,
      "language_loss": 0.70890176,
      "learning_rate": 3.4325450000174535e-06,
      "loss": 0.73055542,
      "num_input_tokens_seen": 47917500,
      "step": 2233,
      "time_per_iteration": 3.0780458450317383
    },
    {
      "auxiliary_loss_clip": 0.01137847,
      "auxiliary_loss_mlp": 0.0103523,
      "balance_loss_clip": 1.04939842,
      "balance_loss_mlp": 1.02524018,
      "epoch": 0.2686226176877292,
      "flos": 20120533764480.0,
      "grad_norm": 3.531227007619784,
      "language_loss": 0.74778259,
      "learning_rate": 3.4320013088462067e-06,
      "loss": 0.76951337,
      "num_input_tokens_seen": 47934860,
      "step": 2234,
      "time_per_iteration": 2.7203075885772705
    },
    {
      "auxiliary_loss_clip": 0.0116323,
      "auxiliary_loss_mlp": 0.01035187,
      "balance_loss_clip": 1.05042088,
      "balance_loss_mlp": 1.02623391,
      "epoch": 0.2687428605783683,
      "flos": 21872040750720.0,
      "grad_norm": 1.6287222567088606,
      "language_loss": 0.81778407,
      "learning_rate": 3.431457400441499e-06,
      "loss": 0.83976823,
      "num_input_tokens_seen": 47955255,
      "step": 2235,
      "time_per_iteration": 2.765872001647949
    },
    {
      "auxiliary_loss_clip": 0.01035556,
      "auxiliary_loss_mlp": 0.01006201,
      "balance_loss_clip": 1.01941466,
      "balance_loss_mlp": 1.00393569,
      "epoch": 0.2688631034690074,
      "flos": 69943320766080.0,
      "grad_norm": 0.9123079505956964,
      "language_loss": 0.60835618,
      "learning_rate": 3.4309132748858424e-06,
      "loss": 0.62877369,
      "num_input_tokens_seen": 48016245,
      "step": 2236,
      "time_per_iteration": 3.429778575897217
    },
    {
      "auxiliary_loss_clip": 0.01188564,
      "auxiliary_loss_mlp": 0.01030612,
      "balance_loss_clip": 1.05692816,
      "balance_loss_mlp": 1.02183867,
      "epoch": 0.2689833463596465,
      "flos": 22856639431680.0,
      "grad_norm": 1.66521320211076,
      "language_loss": 0.83395159,
      "learning_rate": 3.430368932261779e-06,
      "loss": 0.85614336,
      "num_input_tokens_seen": 48036600,
      "step": 2237,
      "time_per_iteration": 2.695408821105957
    },
    {
      "auxiliary_loss_clip": 0.01175367,
      "auxiliary_loss_mlp": 0.01032113,
      "balance_loss_clip": 1.05442691,
      "balance_loss_mlp": 1.02231395,
      "epoch": 0.2691035892502856,
      "flos": 17200242132480.0,
      "grad_norm": 3.503876587831936,
      "language_loss": 0.75005698,
      "learning_rate": 3.429824372651886e-06,
      "loss": 0.7721318,
      "num_input_tokens_seen": 48054750,
      "step": 2238,
      "time_per_iteration": 2.7807703018188477
    },
    {
      "auxiliary_loss_clip": 0.01156177,
      "auxiliary_loss_mlp": 0.01036308,
      "balance_loss_clip": 1.0519973,
      "balance_loss_mlp": 1.02585912,
      "epoch": 0.26922383214092466,
      "flos": 17747484814080.0,
      "grad_norm": 2.15517679234154,
      "language_loss": 0.8361038,
      "learning_rate": 3.4292795961387732e-06,
      "loss": 0.85802859,
      "num_input_tokens_seen": 48072650,
      "step": 2239,
      "time_per_iteration": 2.711087226867676
    },
    {
      "auxiliary_loss_clip": 0.01205176,
      "auxiliary_loss_mlp": 0.01033422,
      "balance_loss_clip": 1.05669045,
      "balance_loss_mlp": 1.02397442,
      "epoch": 0.26934407503156377,
      "flos": 16173376122240.0,
      "grad_norm": 2.0940325808336517,
      "language_loss": 0.8762455,
      "learning_rate": 3.4287346028050818e-06,
      "loss": 0.89863151,
      "num_input_tokens_seen": 48088720,
      "step": 2240,
      "time_per_iteration": 2.6220645904541016
    },
    {
      "auxiliary_loss_clip": 0.01172563,
      "auxiliary_loss_mlp": 0.01028457,
      "balance_loss_clip": 1.0520227,
      "balance_loss_mlp": 1.01971865,
      "epoch": 0.2694643179222028,
      "flos": 23732895715200.0,
      "grad_norm": 1.5767410987140882,
      "language_loss": 0.79628795,
      "learning_rate": 3.4281893927334866e-06,
      "loss": 0.8182981,
      "num_input_tokens_seen": 48108630,
      "step": 2241,
      "time_per_iteration": 2.731638193130493
    },
    {
      "auxiliary_loss_clip": 0.0119096,
      "auxiliary_loss_mlp": 0.01029256,
      "balance_loss_clip": 1.05524337,
      "balance_loss_mlp": 1.02008331,
      "epoch": 0.26958456081284193,
      "flos": 24718140840960.0,
      "grad_norm": 1.984485092082006,
      "language_loss": 0.75532472,
      "learning_rate": 3.4276439660066963e-06,
      "loss": 0.77752686,
      "num_input_tokens_seen": 48128330,
      "step": 2242,
      "time_per_iteration": 2.674990177154541
    },
    {
      "auxiliary_loss_clip": 0.01204347,
      "auxiliary_loss_mlp": 0.0103475,
      "balance_loss_clip": 1.05775428,
      "balance_loss_mlp": 1.02568984,
      "epoch": 0.26970480370348104,
      "flos": 18112588606080.0,
      "grad_norm": 2.7755946024011613,
      "language_loss": 0.84265792,
      "learning_rate": 3.427098322707452e-06,
      "loss": 0.86504889,
      "num_input_tokens_seen": 48144295,
      "step": 2243,
      "time_per_iteration": 2.563952684402466
    },
    {
      "auxiliary_loss_clip": 0.01194117,
      "auxiliary_loss_mlp": 0.01037633,
      "balance_loss_clip": 1.06196141,
      "balance_loss_mlp": 1.02812016,
      "epoch": 0.2698250465941201,
      "flos": 10816546250880.0,
      "grad_norm": 2.492226179288465,
      "language_loss": 0.89648795,
      "learning_rate": 3.426552462918526e-06,
      "loss": 0.91880548,
      "num_input_tokens_seen": 48162230,
      "step": 2244,
      "time_per_iteration": 2.6364941596984863
    },
    {
      "auxiliary_loss_clip": 0.01204324,
      "auxiliary_loss_mlp": 0.01031565,
      "balance_loss_clip": 1.05976498,
      "balance_loss_mlp": 1.02199209,
      "epoch": 0.2699452894847592,
      "flos": 17308117653120.0,
      "grad_norm": 2.253748281086285,
      "language_loss": 0.73136568,
      "learning_rate": 3.426006386722726e-06,
      "loss": 0.75372458,
      "num_input_tokens_seen": 48180290,
      "step": 2245,
      "time_per_iteration": 2.5815622806549072
    },
    {
      "auxiliary_loss_clip": 0.01169408,
      "auxiliary_loss_mlp": 0.0104032,
      "balance_loss_clip": 1.059587,
      "balance_loss_mlp": 1.03028297,
      "epoch": 0.2700655323753983,
      "flos": 18078150441600.0,
      "grad_norm": 1.8702319731934123,
      "language_loss": 0.92366099,
      "learning_rate": 3.4254600942028914e-06,
      "loss": 0.94575828,
      "num_input_tokens_seen": 48198165,
      "step": 2246,
      "time_per_iteration": 2.705512523651123
    },
    {
      "auxiliary_loss_clip": 0.01171411,
      "auxiliary_loss_mlp": 0.0103686,
      "balance_loss_clip": 1.05578923,
      "balance_loss_mlp": 1.02835476,
      "epoch": 0.2701857752660374,
      "flos": 18186636493440.0,
      "grad_norm": 2.4974395671781133,
      "language_loss": 0.82603467,
      "learning_rate": 3.424913585441893e-06,
      "loss": 0.84811741,
      "num_input_tokens_seen": 48216000,
      "step": 2247,
      "time_per_iteration": 2.63370680809021
    },
    {
      "auxiliary_loss_clip": 0.0118764,
      "auxiliary_loss_mlp": 0.01032383,
      "balance_loss_clip": 1.05633914,
      "balance_loss_mlp": 1.02319753,
      "epoch": 0.2703060181566765,
      "flos": 16319496648960.0,
      "grad_norm": 3.618570510720264,
      "language_loss": 0.8777504,
      "learning_rate": 3.4243668605226374e-06,
      "loss": 0.89995056,
      "num_input_tokens_seen": 48233025,
      "step": 2248,
      "time_per_iteration": 2.62382435798645
    },
    {
      "auxiliary_loss_clip": 0.01163402,
      "auxiliary_loss_mlp": 0.01037847,
      "balance_loss_clip": 1.05673623,
      "balance_loss_mlp": 1.02748132,
      "epoch": 0.2704262610473156,
      "flos": 19572357329280.0,
      "grad_norm": 2.5118718628518013,
      "language_loss": 0.82806599,
      "learning_rate": 3.423819919528061e-06,
      "loss": 0.85007852,
      "num_input_tokens_seen": 48251110,
      "step": 2249,
      "time_per_iteration": 3.8477587699890137
    },
    {
      "auxiliary_loss_clip": 0.01147069,
      "auxiliary_loss_mlp": 0.01027972,
      "balance_loss_clip": 1.04655266,
      "balance_loss_mlp": 1.01904321,
      "epoch": 0.27054650393795465,
      "flos": 20740746925440.0,
      "grad_norm": 1.7716723723822048,
      "language_loss": 0.78478092,
      "learning_rate": 3.4232727625411355e-06,
      "loss": 0.80653137,
      "num_input_tokens_seen": 48270215,
      "step": 2250,
      "time_per_iteration": 2.767078399658203
    },
    {
      "auxiliary_loss_clip": 0.01123326,
      "auxiliary_loss_mlp": 0.01031635,
      "balance_loss_clip": 1.0472697,
      "balance_loss_mlp": 1.02268267,
      "epoch": 0.27066674682859376,
      "flos": 18658322916480.0,
      "grad_norm": 1.8022671458723503,
      "language_loss": 0.868294,
      "learning_rate": 3.4227253896448626e-06,
      "loss": 0.88984358,
      "num_input_tokens_seen": 48288075,
      "step": 2251,
      "time_per_iteration": 2.787029504776001
    },
    {
      "auxiliary_loss_clip": 0.01204036,
      "auxiliary_loss_mlp": 0.01030839,
      "balance_loss_clip": 1.05746925,
      "balance_loss_mlp": 1.0222857,
      "epoch": 0.2707869897192329,
      "flos": 23002759958400.0,
      "grad_norm": 2.3350177002462567,
      "language_loss": 0.82068485,
      "learning_rate": 3.42217780092228e-06,
      "loss": 0.84303367,
      "num_input_tokens_seen": 48306415,
      "step": 2252,
      "time_per_iteration": 2.6354494094848633
    },
    {
      "auxiliary_loss_clip": 0.01075377,
      "auxiliary_loss_mlp": 0.01004319,
      "balance_loss_clip": 1.02228248,
      "balance_loss_mlp": 1.00208938,
      "epoch": 0.27090723260987193,
      "flos": 58323240293760.0,
      "grad_norm": 0.8024672355690275,
      "language_loss": 0.60266376,
      "learning_rate": 3.421629996456456e-06,
      "loss": 0.62346077,
      "num_input_tokens_seen": 48365035,
      "step": 2253,
      "time_per_iteration": 4.105928659439087
    },
    {
      "auxiliary_loss_clip": 0.01185328,
      "auxiliary_loss_mlp": 0.01035695,
      "balance_loss_clip": 1.05395365,
      "balance_loss_mlp": 1.02565181,
      "epoch": 0.27102747550051104,
      "flos": 11984540797440.0,
      "grad_norm": 2.072238863216963,
      "language_loss": 0.82984418,
      "learning_rate": 3.421081976330491e-06,
      "loss": 0.85205448,
      "num_input_tokens_seen": 48383550,
      "step": 2254,
      "time_per_iteration": 2.667833089828491
    },
    {
      "auxiliary_loss_clip": 0.01170353,
      "auxiliary_loss_mlp": 0.01044033,
      "balance_loss_clip": 1.05175066,
      "balance_loss_mlp": 1.03511012,
      "epoch": 0.27114771839115015,
      "flos": 19900401264000.0,
      "grad_norm": 2.9279864294154123,
      "language_loss": 0.87830806,
      "learning_rate": 3.4205337406275207e-06,
      "loss": 0.9004519,
      "num_input_tokens_seen": 48403670,
      "step": 2255,
      "time_per_iteration": 2.7073915004730225
    },
    {
      "auxiliary_loss_clip": 0.01200841,
      "auxiliary_loss_mlp": 0.01034671,
      "balance_loss_clip": 1.05403852,
      "balance_loss_mlp": 1.02533078,
      "epoch": 0.2712679612817892,
      "flos": 18331966920960.0,
      "grad_norm": 3.6973722134784395,
      "language_loss": 0.75817657,
      "learning_rate": 3.4199852894307114e-06,
      "loss": 0.78053164,
      "num_input_tokens_seen": 48420420,
      "step": 2256,
      "time_per_iteration": 4.436914920806885
    },
    {
      "auxiliary_loss_clip": 0.01130749,
      "auxiliary_loss_mlp": 0.01035851,
      "balance_loss_clip": 1.05104673,
      "balance_loss_mlp": 1.02642798,
      "epoch": 0.2713882041724283,
      "flos": 24460302038400.0,
      "grad_norm": 2.020935595226801,
      "language_loss": 0.78907758,
      "learning_rate": 3.419436622823262e-06,
      "loss": 0.81074357,
      "num_input_tokens_seen": 48441140,
      "step": 2257,
      "time_per_iteration": 2.8501999378204346
    },
    {
      "auxiliary_loss_clip": 0.01171159,
      "auxiliary_loss_mlp": 0.01029118,
      "balance_loss_clip": 1.0516876,
      "balance_loss_mlp": 1.02015352,
      "epoch": 0.27150844706306737,
      "flos": 23039317025280.0,
      "grad_norm": 1.5244851836775115,
      "language_loss": 0.74349898,
      "learning_rate": 3.4188877408884063e-06,
      "loss": 0.76550168,
      "num_input_tokens_seen": 48461845,
      "step": 2258,
      "time_per_iteration": 2.9187352657318115
    },
    {
      "auxiliary_loss_clip": 0.01165911,
      "auxiliary_loss_mlp": 0.01025104,
      "balance_loss_clip": 1.05272889,
      "balance_loss_mlp": 1.01618743,
      "epoch": 0.2716286899537065,
      "flos": 22563644192640.0,
      "grad_norm": 3.133842791265892,
      "language_loss": 0.6529209,
      "learning_rate": 3.4183386437094088e-06,
      "loss": 0.67483103,
      "num_input_tokens_seen": 48478510,
      "step": 2259,
      "time_per_iteration": 2.6737172603607178
    },
    {
      "auxiliary_loss_clip": 0.01175572,
      "auxiliary_loss_mlp": 0.01032411,
      "balance_loss_clip": 1.05327082,
      "balance_loss_mlp": 1.02337503,
      "epoch": 0.2717489328443456,
      "flos": 13115044523520.0,
      "grad_norm": 2.3353183897836,
      "language_loss": 0.82491642,
      "learning_rate": 3.417789331369565e-06,
      "loss": 0.84699625,
      "num_input_tokens_seen": 48494300,
      "step": 2260,
      "time_per_iteration": 2.7134249210357666
    },
    {
      "auxiliary_loss_clip": 0.01207014,
      "auxiliary_loss_mlp": 0.01034179,
      "balance_loss_clip": 1.0597049,
      "balance_loss_mlp": 1.02481556,
      "epoch": 0.27186917573498465,
      "flos": 29278688060160.0,
      "grad_norm": 2.0414070262122004,
      "language_loss": 0.91550297,
      "learning_rate": 3.4172398039522088e-06,
      "loss": 0.93791497,
      "num_input_tokens_seen": 48515585,
      "step": 2261,
      "time_per_iteration": 2.7078797817230225
    },
    {
      "auxiliary_loss_clip": 0.01189596,
      "auxiliary_loss_mlp": 0.01031936,
      "balance_loss_clip": 1.05665159,
      "balance_loss_mlp": 1.02305484,
      "epoch": 0.27198941862562376,
      "flos": 26032220000640.0,
      "grad_norm": 1.8310170756824937,
      "language_loss": 0.80002964,
      "learning_rate": 3.4166900615407e-06,
      "loss": 0.82224488,
      "num_input_tokens_seen": 48533500,
      "step": 2262,
      "time_per_iteration": 2.6920950412750244
    },
    {
      "auxiliary_loss_clip": 0.01185497,
      "auxiliary_loss_mlp": 0.01032027,
      "balance_loss_clip": 1.05559468,
      "balance_loss_mlp": 1.02317572,
      "epoch": 0.27210966151626287,
      "flos": 32780983760640.0,
      "grad_norm": 2.225004896837347,
      "language_loss": 0.74870867,
      "learning_rate": 3.416140104218436e-06,
      "loss": 0.77088392,
      "num_input_tokens_seen": 48552865,
      "step": 2263,
      "time_per_iteration": 2.7059764862060547
    },
    {
      "auxiliary_loss_clip": 0.0107623,
      "auxiliary_loss_mlp": 0.00755436,
      "balance_loss_clip": 1.02033305,
      "balance_loss_mlp": 1.00021708,
      "epoch": 0.2722299044069019,
      "flos": 65471043219840.0,
      "grad_norm": 0.8439753316452376,
      "language_loss": 0.69589019,
      "learning_rate": 3.4155899320688437e-06,
      "loss": 0.71420681,
      "num_input_tokens_seen": 48618940,
      "step": 2264,
      "time_per_iteration": 3.2804648876190186
    },
    {
      "auxiliary_loss_clip": 0.01131037,
      "auxiliary_loss_mlp": 0.01031735,
      "balance_loss_clip": 1.05193293,
      "balance_loss_mlp": 1.02236474,
      "epoch": 0.27235014729754103,
      "flos": 15334143782400.0,
      "grad_norm": 2.355352376633204,
      "language_loss": 0.74075484,
      "learning_rate": 3.415039545175384e-06,
      "loss": 0.76238251,
      "num_input_tokens_seen": 48634665,
      "step": 2265,
      "time_per_iteration": 2.8144209384918213
    },
    {
      "auxiliary_loss_clip": 0.01189467,
      "auxiliary_loss_mlp": 0.01036904,
      "balance_loss_clip": 1.05561638,
      "balance_loss_mlp": 1.0285002,
      "epoch": 0.27247039018818014,
      "flos": 21872363973120.0,
      "grad_norm": 2.3401154833609805,
      "language_loss": 0.653283,
      "learning_rate": 3.414488943621551e-06,
      "loss": 0.67554677,
      "num_input_tokens_seen": 48653330,
      "step": 2266,
      "time_per_iteration": 2.6721744537353516
    },
    {
      "auxiliary_loss_clip": 0.0118707,
      "auxiliary_loss_mlp": 0.01037487,
      "balance_loss_clip": 1.05684018,
      "balance_loss_mlp": 1.02828372,
      "epoch": 0.2725906330788192,
      "flos": 18695490514560.0,
      "grad_norm": 1.8646913072600009,
      "language_loss": 0.7355082,
      "learning_rate": 3.41393812749087e-06,
      "loss": 0.75775373,
      "num_input_tokens_seen": 48671375,
      "step": 2267,
      "time_per_iteration": 2.605172872543335
    },
    {
      "auxiliary_loss_clip": 0.01173285,
      "auxiliary_loss_mlp": 0.01032521,
      "balance_loss_clip": 1.05768967,
      "balance_loss_mlp": 1.02340794,
      "epoch": 0.2727108759694583,
      "flos": 17886099398400.0,
      "grad_norm": 2.7747946154326737,
      "language_loss": 0.71699882,
      "learning_rate": 3.4133870968668984e-06,
      "loss": 0.73905689,
      "num_input_tokens_seen": 48686175,
      "step": 2268,
      "time_per_iteration": 2.684117078781128
    },
    {
      "auxiliary_loss_clip": 0.01176263,
      "auxiliary_loss_mlp": 0.01033818,
      "balance_loss_clip": 1.05631661,
      "balance_loss_mlp": 1.02388823,
      "epoch": 0.2728311188600974,
      "flos": 24461666755200.0,
      "grad_norm": 2.042080036318221,
      "language_loss": 0.78988075,
      "learning_rate": 3.412835851833229e-06,
      "loss": 0.81198156,
      "num_input_tokens_seen": 48708370,
      "step": 2269,
      "time_per_iteration": 2.705493450164795
    },
    {
      "auxiliary_loss_clip": 0.01183614,
      "auxiliary_loss_mlp": 0.01038441,
      "balance_loss_clip": 1.05600286,
      "balance_loss_mlp": 1.03035855,
      "epoch": 0.2729513617507365,
      "flos": 30993314757120.0,
      "grad_norm": 2.129643050326694,
      "language_loss": 0.78208184,
      "learning_rate": 3.4122843924734834e-06,
      "loss": 0.80430245,
      "num_input_tokens_seen": 48730670,
      "step": 2270,
      "time_per_iteration": 2.7089879512786865
    },
    {
      "auxiliary_loss_clip": 0.01164888,
      "auxiliary_loss_mlp": 0.01029413,
      "balance_loss_clip": 1.05095243,
      "balance_loss_mlp": 1.02033544,
      "epoch": 0.2730716046413756,
      "flos": 19094637421440.0,
      "grad_norm": 2.0168288392842606,
      "language_loss": 0.8827309,
      "learning_rate": 3.411732718871319e-06,
      "loss": 0.90467387,
      "num_input_tokens_seen": 48746510,
      "step": 2271,
      "time_per_iteration": 2.6455936431884766
    },
    {
      "auxiliary_loss_clip": 0.0120326,
      "auxiliary_loss_mlp": 0.01032228,
      "balance_loss_clip": 1.06216168,
      "balance_loss_mlp": 1.02356124,
      "epoch": 0.27319184753201464,
      "flos": 26944566474240.0,
      "grad_norm": 2.022940801363115,
      "language_loss": 0.78935623,
      "learning_rate": 3.4111808311104227e-06,
      "loss": 0.81171107,
      "num_input_tokens_seen": 48768825,
      "step": 2272,
      "time_per_iteration": 2.6933422088623047
    },
    {
      "auxiliary_loss_clip": 0.01180154,
      "auxiliary_loss_mlp": 0.01030921,
      "balance_loss_clip": 1.05394912,
      "balance_loss_mlp": 1.02149749,
      "epoch": 0.27331209042265375,
      "flos": 31759828012800.0,
      "grad_norm": 1.9813013230545355,
      "language_loss": 0.69558036,
      "learning_rate": 3.410628729274517e-06,
      "loss": 0.71769106,
      "num_input_tokens_seen": 48790345,
      "step": 2273,
      "time_per_iteration": 2.743873119354248
    },
    {
      "auxiliary_loss_clip": 0.0116834,
      "auxiliary_loss_mlp": 0.00763334,
      "balance_loss_clip": 1.05335879,
      "balance_loss_mlp": 1.00037646,
      "epoch": 0.27343233331329286,
      "flos": 25739081107200.0,
      "grad_norm": 1.995230637962598,
      "language_loss": 0.828354,
      "learning_rate": 3.4100764134473546e-06,
      "loss": 0.84767079,
      "num_input_tokens_seen": 48809630,
      "step": 2274,
      "time_per_iteration": 2.7149124145507812
    },
    {
      "auxiliary_loss_clip": 0.01202647,
      "auxiliary_loss_mlp": 0.01037756,
      "balance_loss_clip": 1.05793571,
      "balance_loss_mlp": 1.02884531,
      "epoch": 0.2735525762039319,
      "flos": 24389414547840.0,
      "grad_norm": 2.3712941154444547,
      "language_loss": 0.84656417,
      "learning_rate": 3.4095238837127215e-06,
      "loss": 0.86896819,
      "num_input_tokens_seen": 48828770,
      "step": 2275,
      "time_per_iteration": 3.568774938583374
    },
    {
      "auxiliary_loss_clip": 0.01153155,
      "auxiliary_loss_mlp": 0.01032735,
      "balance_loss_clip": 1.0503633,
      "balance_loss_mlp": 1.02365732,
      "epoch": 0.27367281909457103,
      "flos": 14465357527680.0,
      "grad_norm": 2.069719949048533,
      "language_loss": 0.79552901,
      "learning_rate": 3.4089711401544355e-06,
      "loss": 0.81738794,
      "num_input_tokens_seen": 48846365,
      "step": 2276,
      "time_per_iteration": 2.6595826148986816
    },
    {
      "auxiliary_loss_clip": 0.01186803,
      "auxiliary_loss_mlp": 0.01032305,
      "balance_loss_clip": 1.05444407,
      "balance_loss_mlp": 1.02267241,
      "epoch": 0.27379306198521014,
      "flos": 23476996247040.0,
      "grad_norm": 2.571770339442371,
      "language_loss": 0.67240494,
      "learning_rate": 3.4084181828563486e-06,
      "loss": 0.69459605,
      "num_input_tokens_seen": 48863085,
      "step": 2277,
      "time_per_iteration": 2.74448823928833
    },
    {
      "auxiliary_loss_clip": 0.01140748,
      "auxiliary_loss_mlp": 0.01031877,
      "balance_loss_clip": 1.05022073,
      "balance_loss_mlp": 1.02293611,
      "epoch": 0.2739133048758492,
      "flos": 17458152762240.0,
      "grad_norm": 1.8460070607065087,
      "language_loss": 0.70763659,
      "learning_rate": 3.4078650119023428e-06,
      "loss": 0.72936285,
      "num_input_tokens_seen": 48881400,
      "step": 2278,
      "time_per_iteration": 2.705878496170044
    },
    {
      "auxiliary_loss_clip": 0.01128034,
      "auxiliary_loss_mlp": 0.01041348,
      "balance_loss_clip": 1.04487729,
      "balance_loss_mlp": 1.03194284,
      "epoch": 0.2740335477664883,
      "flos": 19273113123840.0,
      "grad_norm": 2.2697561567033633,
      "language_loss": 0.74011034,
      "learning_rate": 3.4073116273763337e-06,
      "loss": 0.76180416,
      "num_input_tokens_seen": 48895845,
      "step": 2279,
      "time_per_iteration": 3.639044761657715
    },
    {
      "auxiliary_loss_clip": 0.01176851,
      "auxiliary_loss_mlp": 0.01028069,
      "balance_loss_clip": 1.0533421,
      "balance_loss_mlp": 1.01866364,
      "epoch": 0.2741537906571274,
      "flos": 26104723603200.0,
      "grad_norm": 1.7712504389927055,
      "language_loss": 0.81206346,
      "learning_rate": 3.40675802936227e-06,
      "loss": 0.83411264,
      "num_input_tokens_seen": 48916630,
      "step": 2280,
      "time_per_iteration": 2.771557092666626
    },
    {
      "auxiliary_loss_clip": 0.01164632,
      "auxiliary_loss_mlp": 0.01030056,
      "balance_loss_clip": 1.05222321,
      "balance_loss_mlp": 1.02044153,
      "epoch": 0.27427403354776647,
      "flos": 34164190644480.0,
      "grad_norm": 2.1171290157856855,
      "language_loss": 0.72074854,
      "learning_rate": 3.4062042179441318e-06,
      "loss": 0.74269545,
      "num_input_tokens_seen": 48937100,
      "step": 2281,
      "time_per_iteration": 3.7796597480773926
    },
    {
      "auxiliary_loss_clip": 0.01186476,
      "auxiliary_loss_mlp": 0.01026508,
      "balance_loss_clip": 1.05874896,
      "balance_loss_mlp": 1.01798475,
      "epoch": 0.2743942764384056,
      "flos": 18766988536320.0,
      "grad_norm": 1.9256935997915743,
      "language_loss": 0.80767041,
      "learning_rate": 3.4056501932059314e-06,
      "loss": 0.82980025,
      "num_input_tokens_seen": 48955175,
      "step": 2282,
      "time_per_iteration": 3.6783294677734375
    },
    {
      "auxiliary_loss_clip": 0.01102571,
      "auxiliary_loss_mlp": 0.0100744,
      "balance_loss_clip": 1.02294993,
      "balance_loss_mlp": 1.0054251,
      "epoch": 0.2745145193290447,
      "flos": 64904048058240.0,
      "grad_norm": 0.7778266523865007,
      "language_loss": 0.58028936,
      "learning_rate": 3.405095955231715e-06,
      "loss": 0.60138941,
      "num_input_tokens_seen": 49006830,
      "step": 2283,
      "time_per_iteration": 3.1335926055908203
    },
    {
      "auxiliary_loss_clip": 0.0119102,
      "auxiliary_loss_mlp": 0.0103371,
      "balance_loss_clip": 1.05309367,
      "balance_loss_mlp": 1.02508521,
      "epoch": 0.27463476221968375,
      "flos": 16136926796160.0,
      "grad_norm": 3.021598428508459,
      "language_loss": 0.94548357,
      "learning_rate": 3.4045415041055585e-06,
      "loss": 0.96773088,
      "num_input_tokens_seen": 49022470,
      "step": 2284,
      "time_per_iteration": 2.6129205226898193
    },
    {
      "auxiliary_loss_clip": 0.01176211,
      "auxiliary_loss_mlp": 0.01032848,
      "balance_loss_clip": 1.05369663,
      "balance_loss_mlp": 1.02412748,
      "epoch": 0.27475500511032286,
      "flos": 10376712213120.0,
      "grad_norm": 6.768879137803772,
      "language_loss": 0.78925574,
      "learning_rate": 3.4039868399115728e-06,
      "loss": 0.81134629,
      "num_input_tokens_seen": 49037110,
      "step": 2285,
      "time_per_iteration": 2.6596314907073975
    },
    {
      "auxiliary_loss_clip": 0.01132711,
      "auxiliary_loss_mlp": 0.01032818,
      "balance_loss_clip": 1.05136299,
      "balance_loss_mlp": 1.02344251,
      "epoch": 0.27487524800096197,
      "flos": 17311062568320.0,
      "grad_norm": 2.0460277080554867,
      "language_loss": 0.80511183,
      "learning_rate": 3.4034319627339003e-06,
      "loss": 0.82676709,
      "num_input_tokens_seen": 49053975,
      "step": 2286,
      "time_per_iteration": 2.745696783065796
    },
    {
      "auxiliary_loss_clip": 0.01177862,
      "auxiliary_loss_mlp": 0.01034363,
      "balance_loss_clip": 1.05641699,
      "balance_loss_mlp": 1.02574372,
      "epoch": 0.274995490891601,
      "flos": 27120205002240.0,
      "grad_norm": 4.2806835233643294,
      "language_loss": 0.70085186,
      "learning_rate": 3.402876872656715e-06,
      "loss": 0.72297406,
      "num_input_tokens_seen": 49072295,
      "step": 2287,
      "time_per_iteration": 2.748152017593384
    },
    {
      "auxiliary_loss_clip": 0.01171177,
      "auxiliary_loss_mlp": 0.01033235,
      "balance_loss_clip": 1.05601978,
      "balance_loss_mlp": 1.02383542,
      "epoch": 0.27511573378224013,
      "flos": 23436093634560.0,
      "grad_norm": 2.046667845909836,
      "language_loss": 0.89463043,
      "learning_rate": 3.402321569764223e-06,
      "loss": 0.91667455,
      "num_input_tokens_seen": 49091600,
      "step": 2288,
      "time_per_iteration": 2.6951262950897217
    },
    {
      "auxiliary_loss_clip": 0.01144561,
      "auxiliary_loss_mlp": 0.00764761,
      "balance_loss_clip": 1.05102849,
      "balance_loss_mlp": 1.00059903,
      "epoch": 0.2752359766728792,
      "flos": 16722019434240.0,
      "grad_norm": 1.966592425854624,
      "language_loss": 0.83729303,
      "learning_rate": 3.4017660541406635e-06,
      "loss": 0.85638624,
      "num_input_tokens_seen": 49107665,
      "step": 2289,
      "time_per_iteration": 2.7898805141448975
    },
    {
      "auxiliary_loss_clip": 0.01178135,
      "auxiliary_loss_mlp": 0.01029645,
      "balance_loss_clip": 1.05188024,
      "balance_loss_mlp": 1.01997662,
      "epoch": 0.2753562195635183,
      "flos": 25297738698240.0,
      "grad_norm": 1.7808938890664154,
      "language_loss": 0.74668616,
      "learning_rate": 3.4012103258703092e-06,
      "loss": 0.76876396,
      "num_input_tokens_seen": 49126420,
      "step": 2290,
      "time_per_iteration": 2.7106032371520996
    },
    {
      "auxiliary_loss_clip": 0.01157952,
      "auxiliary_loss_mlp": 0.01033364,
      "balance_loss_clip": 1.05237091,
      "balance_loss_mlp": 1.02354109,
      "epoch": 0.2754764624541574,
      "flos": 27338972785920.0,
      "grad_norm": 1.9154796267311263,
      "language_loss": 0.83065927,
      "learning_rate": 3.4006543850374616e-06,
      "loss": 0.85257244,
      "num_input_tokens_seen": 49141470,
      "step": 2291,
      "time_per_iteration": 2.7091152667999268
    },
    {
      "auxiliary_loss_clip": 0.01188049,
      "auxiliary_loss_mlp": 0.01034514,
      "balance_loss_clip": 1.05309689,
      "balance_loss_mlp": 1.02522206,
      "epoch": 0.27559670534479647,
      "flos": 17238379397760.0,
      "grad_norm": 2.0387133373870276,
      "language_loss": 0.75360715,
      "learning_rate": 3.400098231726458e-06,
      "loss": 0.77583277,
      "num_input_tokens_seen": 49158570,
      "step": 2292,
      "time_per_iteration": 2.632749557495117
    },
    {
      "auxiliary_loss_clip": 0.0116225,
      "auxiliary_loss_mlp": 0.01025496,
      "balance_loss_clip": 1.05011165,
      "balance_loss_mlp": 1.01607227,
      "epoch": 0.2757169482354356,
      "flos": 21939085486080.0,
      "grad_norm": 2.2279482624046527,
      "language_loss": 0.86742681,
      "learning_rate": 3.3995418660216657e-06,
      "loss": 0.88930422,
      "num_input_tokens_seen": 49176025,
      "step": 2293,
      "time_per_iteration": 2.8080203533172607
    },
    {
      "auxiliary_loss_clip": 0.01209449,
      "auxiliary_loss_mlp": 0.01034496,
      "balance_loss_clip": 1.06007719,
      "balance_loss_mlp": 1.02443516,
      "epoch": 0.2758371911260747,
      "flos": 20850669521280.0,
      "grad_norm": 2.262884773003336,
      "language_loss": 0.80434501,
      "learning_rate": 3.3989852880074848e-06,
      "loss": 0.82678449,
      "num_input_tokens_seen": 49197455,
      "step": 2294,
      "time_per_iteration": 2.7196407318115234
    },
    {
      "auxiliary_loss_clip": 0.01083828,
      "auxiliary_loss_mlp": 0.01004149,
      "balance_loss_clip": 1.02812636,
      "balance_loss_mlp": 1.0021466,
      "epoch": 0.27595743401671374,
      "flos": 69269063592960.0,
      "grad_norm": 0.7476867123627904,
      "language_loss": 0.60615277,
      "learning_rate": 3.398428497768348e-06,
      "loss": 0.62703252,
      "num_input_tokens_seen": 49262625,
      "step": 2295,
      "time_per_iteration": 3.3849334716796875
    },
    {
      "auxiliary_loss_clip": 0.01166101,
      "auxiliary_loss_mlp": 0.01029497,
      "balance_loss_clip": 1.05083323,
      "balance_loss_mlp": 1.02059245,
      "epoch": 0.27607767690735285,
      "flos": 21215019127680.0,
      "grad_norm": 1.8463181454058246,
      "language_loss": 0.7201854,
      "learning_rate": 3.3978714953887205e-06,
      "loss": 0.74214137,
      "num_input_tokens_seen": 49282380,
      "step": 2296,
      "time_per_iteration": 2.7992069721221924
    },
    {
      "auxiliary_loss_clip": 0.01128037,
      "auxiliary_loss_mlp": 0.01033578,
      "balance_loss_clip": 1.04570699,
      "balance_loss_mlp": 1.02444053,
      "epoch": 0.27619791979799196,
      "flos": 24825334003200.0,
      "grad_norm": 1.801660517983336,
      "language_loss": 0.86351573,
      "learning_rate": 3.397314280953098e-06,
      "loss": 0.88513184,
      "num_input_tokens_seen": 49303205,
      "step": 2297,
      "time_per_iteration": 2.803788423538208
    },
    {
      "auxiliary_loss_clip": 0.01164808,
      "auxiliary_loss_mlp": 0.01036623,
      "balance_loss_clip": 1.05086195,
      "balance_loss_mlp": 1.02835631,
      "epoch": 0.276318162688631,
      "flos": 24753548672640.0,
      "grad_norm": 3.550408142048715,
      "language_loss": 0.80528313,
      "learning_rate": 3.3967568545460108e-06,
      "loss": 0.82729745,
      "num_input_tokens_seen": 49322745,
      "step": 2298,
      "time_per_iteration": 2.7134318351745605
    },
    {
      "auxiliary_loss_clip": 0.01182831,
      "auxiliary_loss_mlp": 0.01032407,
      "balance_loss_clip": 1.05352747,
      "balance_loss_mlp": 1.02340674,
      "epoch": 0.27643840557927013,
      "flos": 18150007599360.0,
      "grad_norm": 1.9498801075970262,
      "language_loss": 0.80673146,
      "learning_rate": 3.3961992162520185e-06,
      "loss": 0.82888377,
      "num_input_tokens_seen": 49341370,
      "step": 2299,
      "time_per_iteration": 2.6921231746673584
    },
    {
      "auxiliary_loss_clip": 0.0118633,
      "auxiliary_loss_mlp": 0.01032919,
      "balance_loss_clip": 1.05384958,
      "balance_loss_mlp": 1.02372825,
      "epoch": 0.27655864846990924,
      "flos": 24823933372800.0,
      "grad_norm": 2.876944248995244,
      "language_loss": 0.72201419,
      "learning_rate": 3.3956413661557156e-06,
      "loss": 0.74420667,
      "num_input_tokens_seen": 49361545,
      "step": 2300,
      "time_per_iteration": 2.660743474960327
    },
    {
      "auxiliary_loss_clip": 0.01165615,
      "auxiliary_loss_mlp": 0.01034657,
      "balance_loss_clip": 1.05230045,
      "balance_loss_mlp": 1.02528095,
      "epoch": 0.2766788913605483,
      "flos": 20266582464000.0,
      "grad_norm": 2.6109510457264475,
      "language_loss": 0.66770899,
      "learning_rate": 3.3950833043417273e-06,
      "loss": 0.68971169,
      "num_input_tokens_seen": 49379690,
      "step": 2301,
      "time_per_iteration": 3.641766309738159
    },
    {
      "auxiliary_loss_clip": 0.01192406,
      "auxiliary_loss_mlp": 0.010304,
      "balance_loss_clip": 1.059659,
      "balance_loss_mlp": 1.02082181,
      "epoch": 0.2767991342511874,
      "flos": 21470272151040.0,
      "grad_norm": 3.3700659403415996,
      "language_loss": 0.74108267,
      "learning_rate": 3.3945250308947105e-06,
      "loss": 0.76331079,
      "num_input_tokens_seen": 49395995,
      "step": 2302,
      "time_per_iteration": 2.651050090789795
    },
    {
      "auxiliary_loss_clip": 0.01090723,
      "auxiliary_loss_mlp": 0.01002214,
      "balance_loss_clip": 1.02027011,
      "balance_loss_mlp": 1.00011599,
      "epoch": 0.2769193771418265,
      "flos": 66002627571840.0,
      "grad_norm": 1.2237202648789334,
      "language_loss": 0.68300402,
      "learning_rate": 3.3939665458993556e-06,
      "loss": 0.70393342,
      "num_input_tokens_seen": 49450415,
      "step": 2303,
      "time_per_iteration": 3.1721460819244385
    },
    {
      "auxiliary_loss_clip": 0.0116337,
      "auxiliary_loss_mlp": 0.01030969,
      "balance_loss_clip": 1.05034792,
      "balance_loss_mlp": 1.02140808,
      "epoch": 0.27703962003246557,
      "flos": 20704441253760.0,
      "grad_norm": 2.0023148307354246,
      "language_loss": 0.76907074,
      "learning_rate": 3.3934078494403843e-06,
      "loss": 0.79101419,
      "num_input_tokens_seen": 49469990,
      "step": 2304,
      "time_per_iteration": 2.7728493213653564
    },
    {
      "auxiliary_loss_clip": 0.01111336,
      "auxiliary_loss_mlp": 0.00764713,
      "balance_loss_clip": 1.04737234,
      "balance_loss_mlp": 1.00050557,
      "epoch": 0.2771598629231047,
      "flos": 22929897219840.0,
      "grad_norm": 1.9987916024747108,
      "language_loss": 0.81838477,
      "learning_rate": 3.3928489416025495e-06,
      "loss": 0.83714527,
      "num_input_tokens_seen": 49490835,
      "step": 2305,
      "time_per_iteration": 3.7889559268951416
    },
    {
      "auxiliary_loss_clip": 0.0117374,
      "auxiliary_loss_mlp": 0.01034953,
      "balance_loss_clip": 1.05447483,
      "balance_loss_mlp": 1.02508235,
      "epoch": 0.27728010581374374,
      "flos": 18369457741440.0,
      "grad_norm": 2.308304761136325,
      "language_loss": 0.79066783,
      "learning_rate": 3.392289822470638e-06,
      "loss": 0.81275481,
      "num_input_tokens_seen": 49508815,
      "step": 2306,
      "time_per_iteration": 2.9027576446533203
    },
    {
      "auxiliary_loss_clip": 0.01171264,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 1.05332255,
      "balance_loss_mlp": 1.01851285,
      "epoch": 0.27740034870438285,
      "flos": 19427637432960.0,
      "grad_norm": 2.0856969119466915,
      "language_loss": 0.75945306,
      "learning_rate": 3.3917304921294674e-06,
      "loss": 0.78144509,
      "num_input_tokens_seen": 49526980,
      "step": 2307,
      "time_per_iteration": 3.660510778427124
    },
    {
      "auxiliary_loss_clip": 0.01188626,
      "auxiliary_loss_mlp": 0.01033141,
      "balance_loss_clip": 1.05529594,
      "balance_loss_mlp": 1.02359819,
      "epoch": 0.27752059159502196,
      "flos": 21614776565760.0,
      "grad_norm": 5.274908796967043,
      "language_loss": 0.80978274,
      "learning_rate": 3.3911709506638876e-06,
      "loss": 0.83200037,
      "num_input_tokens_seen": 49546290,
      "step": 2308,
      "time_per_iteration": 2.675731897354126
    },
    {
      "auxiliary_loss_clip": 0.01146595,
      "auxiliary_loss_mlp": 0.00763574,
      "balance_loss_clip": 1.04698062,
      "balance_loss_mlp": 1.00046074,
      "epoch": 0.277640834485661,
      "flos": 26608011016320.0,
      "grad_norm": 2.074614842925102,
      "language_loss": 0.81319058,
      "learning_rate": 3.390611198158781e-06,
      "loss": 0.83229226,
      "num_input_tokens_seen": 49564165,
      "step": 2309,
      "time_per_iteration": 2.836418628692627
    },
    {
      "auxiliary_loss_clip": 0.01208232,
      "auxiliary_loss_mlp": 0.01029721,
      "balance_loss_clip": 1.05940163,
      "balance_loss_mlp": 1.02025533,
      "epoch": 0.2777610773763001,
      "flos": 19492814661120.0,
      "grad_norm": 2.1799536867259186,
      "language_loss": 0.90261602,
      "learning_rate": 3.3900512346990612e-06,
      "loss": 0.9249956,
      "num_input_tokens_seen": 49580155,
      "step": 2310,
      "time_per_iteration": 2.6711266040802
    },
    {
      "auxiliary_loss_clip": 0.0114078,
      "auxiliary_loss_mlp": 0.01030434,
      "balance_loss_clip": 1.04602373,
      "balance_loss_mlp": 1.0208379,
      "epoch": 0.27788132026693924,
      "flos": 38290650001920.0,
      "grad_norm": 1.7731119353029887,
      "language_loss": 0.66141504,
      "learning_rate": 3.389491060369674e-06,
      "loss": 0.68312716,
      "num_input_tokens_seen": 49605830,
      "step": 2311,
      "time_per_iteration": 2.898549795150757
    },
    {
      "auxiliary_loss_clip": 0.0113666,
      "auxiliary_loss_mlp": 0.01043643,
      "balance_loss_clip": 1.04922903,
      "balance_loss_mlp": 1.03508377,
      "epoch": 0.2780015631575783,
      "flos": 22382546797440.0,
      "grad_norm": 1.890797936035661,
      "language_loss": 0.89483529,
      "learning_rate": 3.388930675255598e-06,
      "loss": 0.91663837,
      "num_input_tokens_seen": 49625680,
      "step": 2312,
      "time_per_iteration": 2.8286221027374268
    },
    {
      "auxiliary_loss_clip": 0.01177237,
      "auxiliary_loss_mlp": 0.01037616,
      "balance_loss_clip": 1.05383706,
      "balance_loss_mlp": 1.02828836,
      "epoch": 0.2781218060482174,
      "flos": 12203200840320.0,
      "grad_norm": 2.362091959738511,
      "language_loss": 0.78872275,
      "learning_rate": 3.388370079441843e-06,
      "loss": 0.8108713,
      "num_input_tokens_seen": 49641195,
      "step": 2313,
      "time_per_iteration": 2.7313311100006104
    },
    {
      "auxiliary_loss_clip": 0.01160948,
      "auxiliary_loss_mlp": 0.01038207,
      "balance_loss_clip": 1.05470157,
      "balance_loss_mlp": 1.02882516,
      "epoch": 0.2782420489388565,
      "flos": 18107632529280.0,
      "grad_norm": 2.039637090754787,
      "language_loss": 0.92850465,
      "learning_rate": 3.3878092730134505e-06,
      "loss": 0.9504962,
      "num_input_tokens_seen": 49659180,
      "step": 2314,
      "time_per_iteration": 2.6790573596954346
    },
    {
      "auxiliary_loss_clip": 0.01180886,
      "auxiliary_loss_mlp": 0.01035467,
      "balance_loss_clip": 1.0542587,
      "balance_loss_mlp": 1.02597177,
      "epoch": 0.27836229182949557,
      "flos": 18514752255360.0,
      "grad_norm": 2.203977772985784,
      "language_loss": 0.80386686,
      "learning_rate": 3.3872482560554947e-06,
      "loss": 0.82603043,
      "num_input_tokens_seen": 49677955,
      "step": 2315,
      "time_per_iteration": 2.7949914932250977
    },
    {
      "auxiliary_loss_clip": 0.01090378,
      "auxiliary_loss_mlp": 0.01007097,
      "balance_loss_clip": 1.0202837,
      "balance_loss_mlp": 1.00515354,
      "epoch": 0.2784825347201347,
      "flos": 67079230940160.0,
      "grad_norm": 0.8073154683488314,
      "language_loss": 0.56907481,
      "learning_rate": 3.386687028653082e-06,
      "loss": 0.59004956,
      "num_input_tokens_seen": 49740800,
      "step": 2316,
      "time_per_iteration": 3.2383580207824707
    },
    {
      "auxiliary_loss_clip": 0.01146828,
      "auxiliary_loss_mlp": 0.01034519,
      "balance_loss_clip": 1.05074799,
      "balance_loss_mlp": 1.02506614,
      "epoch": 0.2786027776107738,
      "flos": 22631119891200.0,
      "grad_norm": 1.8556607749606644,
      "language_loss": 0.85256493,
      "learning_rate": 3.386125590891349e-06,
      "loss": 0.87437838,
      "num_input_tokens_seen": 49757675,
      "step": 2317,
      "time_per_iteration": 2.7349190711975098
    },
    {
      "auxiliary_loss_clip": 0.01162487,
      "auxiliary_loss_mlp": 0.01035832,
      "balance_loss_clip": 1.05173254,
      "balance_loss_mlp": 1.02696252,
      "epoch": 0.27872302050141284,
      "flos": 15778826156160.0,
      "grad_norm": 2.468117294163745,
      "language_loss": 0.83242744,
      "learning_rate": 3.3855639428554657e-06,
      "loss": 0.85441065,
      "num_input_tokens_seen": 49775205,
      "step": 2318,
      "time_per_iteration": 2.6711957454681396
    },
    {
      "auxiliary_loss_clip": 0.01146873,
      "auxiliary_loss_mlp": 0.01029771,
      "balance_loss_clip": 1.0492065,
      "balance_loss_mlp": 1.02134848,
      "epoch": 0.27884326339205195,
      "flos": 22126970551680.0,
      "grad_norm": 2.156973575202536,
      "language_loss": 0.80043381,
      "learning_rate": 3.385002084630635e-06,
      "loss": 0.82220024,
      "num_input_tokens_seen": 49794175,
      "step": 2319,
      "time_per_iteration": 2.672119140625
    },
    {
      "auxiliary_loss_clip": 0.01191616,
      "auxiliary_loss_mlp": 0.01029939,
      "balance_loss_clip": 1.05629826,
      "balance_loss_mlp": 1.02042055,
      "epoch": 0.278963506282691,
      "flos": 20558715776640.0,
      "grad_norm": 3.986280193640279,
      "language_loss": 0.84840155,
      "learning_rate": 3.384440016302088e-06,
      "loss": 0.87061715,
      "num_input_tokens_seen": 49812850,
      "step": 2320,
      "time_per_iteration": 2.6638221740722656
    },
    {
      "auxiliary_loss_clip": 0.01185444,
      "auxiliary_loss_mlp": 0.01030785,
      "balance_loss_clip": 1.05580819,
      "balance_loss_mlp": 1.02159441,
      "epoch": 0.2790837491733301,
      "flos": 21942928241280.0,
      "grad_norm": 2.3954258695713113,
      "language_loss": 0.62434393,
      "learning_rate": 3.3838777379550923e-06,
      "loss": 0.64650619,
      "num_input_tokens_seen": 49832295,
      "step": 2321,
      "time_per_iteration": 2.6696975231170654
    },
    {
      "auxiliary_loss_clip": 0.01177054,
      "auxiliary_loss_mlp": 0.0102903,
      "balance_loss_clip": 1.05662096,
      "balance_loss_mlp": 1.02008331,
      "epoch": 0.27920399206396923,
      "flos": 26286790665600.0,
      "grad_norm": 16.88766770285102,
      "language_loss": 0.78691518,
      "learning_rate": 3.383315249674944e-06,
      "loss": 0.80897605,
      "num_input_tokens_seen": 49850860,
      "step": 2322,
      "time_per_iteration": 2.791940212249756
    },
    {
      "auxiliary_loss_clip": 0.01159606,
      "auxiliary_loss_mlp": 0.01035514,
      "balance_loss_clip": 1.05357122,
      "balance_loss_mlp": 1.02638221,
      "epoch": 0.2793242349546083,
      "flos": 25400981364480.0,
      "grad_norm": 4.325402427028171,
      "language_loss": 0.86075574,
      "learning_rate": 3.3827525515469715e-06,
      "loss": 0.882707,
      "num_input_tokens_seen": 49865765,
      "step": 2323,
      "time_per_iteration": 2.7910640239715576
    },
    {
      "auxiliary_loss_clip": 0.0114816,
      "auxiliary_loss_mlp": 0.01029771,
      "balance_loss_clip": 1.04846537,
      "balance_loss_mlp": 1.02025843,
      "epoch": 0.2794444778452474,
      "flos": 20850346298880.0,
      "grad_norm": 2.1144659409293363,
      "language_loss": 0.71305335,
      "learning_rate": 3.3821896436565367e-06,
      "loss": 0.7348327,
      "num_input_tokens_seen": 49885425,
      "step": 2324,
      "time_per_iteration": 2.6691856384277344
    },
    {
      "auxiliary_loss_clip": 0.0119335,
      "auxiliary_loss_mlp": 0.01032615,
      "balance_loss_clip": 1.06107306,
      "balance_loss_mlp": 1.02369833,
      "epoch": 0.2795647207358865,
      "flos": 21576244250880.0,
      "grad_norm": 1.8169946084430786,
      "language_loss": 0.69966763,
      "learning_rate": 3.381626526089032e-06,
      "loss": 0.72192729,
      "num_input_tokens_seen": 49904990,
      "step": 2325,
      "time_per_iteration": 2.621131420135498
    },
    {
      "auxiliary_loss_clip": 0.01168692,
      "auxiliary_loss_mlp": 0.01033927,
      "balance_loss_clip": 1.05027902,
      "balance_loss_mlp": 1.02502847,
      "epoch": 0.27968496362652556,
      "flos": 21471744608640.0,
      "grad_norm": 2.254284981771605,
      "language_loss": 0.79317021,
      "learning_rate": 3.3810631989298815e-06,
      "loss": 0.81519639,
      "num_input_tokens_seen": 49924600,
      "step": 2326,
      "time_per_iteration": 2.670252799987793
    },
    {
      "auxiliary_loss_clip": 0.01151497,
      "auxiliary_loss_mlp": 0.01039679,
      "balance_loss_clip": 1.05253959,
      "balance_loss_mlp": 1.02853858,
      "epoch": 0.2798052065171647,
      "flos": 23258695340160.0,
      "grad_norm": 3.3127688821182493,
      "language_loss": 0.84899724,
      "learning_rate": 3.3804996622645423e-06,
      "loss": 0.87090898,
      "num_input_tokens_seen": 49942600,
      "step": 2327,
      "time_per_iteration": 3.829312324523926
    },
    {
      "auxiliary_loss_clip": 0.01206231,
      "auxiliary_loss_mlp": 0.01030882,
      "balance_loss_clip": 1.05935216,
      "balance_loss_mlp": 1.02157831,
      "epoch": 0.2799254494078038,
      "flos": 21539328048000.0,
      "grad_norm": 1.771302012413825,
      "language_loss": 0.89434636,
      "learning_rate": 3.3799359161785015e-06,
      "loss": 0.91671753,
      "num_input_tokens_seen": 49962250,
      "step": 2328,
      "time_per_iteration": 2.6131808757781982
    },
    {
      "auxiliary_loss_clip": 0.01186839,
      "auxiliary_loss_mlp": 0.0103392,
      "balance_loss_clip": 1.05604041,
      "balance_loss_mlp": 1.02477694,
      "epoch": 0.28004569229844284,
      "flos": 26393912000640.0,
      "grad_norm": 1.895704559922919,
      "language_loss": 0.85690743,
      "learning_rate": 3.3793719607572798e-06,
      "loss": 0.87911505,
      "num_input_tokens_seen": 49983215,
      "step": 2329,
      "time_per_iteration": 2.67622709274292
    },
    {
      "auxiliary_loss_clip": 0.01159324,
      "auxiliary_loss_mlp": 0.01032256,
      "balance_loss_clip": 1.05124772,
      "balance_loss_mlp": 1.02271926,
      "epoch": 0.28016593518908195,
      "flos": 33547676584320.0,
      "grad_norm": 5.580542672259064,
      "language_loss": 0.77181137,
      "learning_rate": 3.378807796086428e-06,
      "loss": 0.79372722,
      "num_input_tokens_seen": 50006075,
      "step": 2330,
      "time_per_iteration": 3.6839306354522705
    },
    {
      "auxiliary_loss_clip": 0.01205082,
      "auxiliary_loss_mlp": 0.01030683,
      "balance_loss_clip": 1.0590477,
      "balance_loss_mlp": 1.02184379,
      "epoch": 0.28028617807972106,
      "flos": 15340823712000.0,
      "grad_norm": 2.317322704040704,
      "language_loss": 0.77230871,
      "learning_rate": 3.37824342225153e-06,
      "loss": 0.79466641,
      "num_input_tokens_seen": 50022495,
      "step": 2331,
      "time_per_iteration": 2.5543482303619385
    },
    {
      "auxiliary_loss_clip": 0.01147018,
      "auxiliary_loss_mlp": 0.01033185,
      "balance_loss_clip": 1.05393839,
      "balance_loss_mlp": 1.02450705,
      "epoch": 0.2804064209703601,
      "flos": 25520277409920.0,
      "grad_norm": 1.863933779593855,
      "language_loss": 0.77598453,
      "learning_rate": 3.3776788393382006e-06,
      "loss": 0.79778659,
      "num_input_tokens_seen": 50041975,
      "step": 2332,
      "time_per_iteration": 2.743546962738037
    },
    {
      "auxiliary_loss_clip": 0.01206149,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.05940235,
      "balance_loss_mlp": 1.02194798,
      "epoch": 0.2805266638609992,
      "flos": 29351766280320.0,
      "grad_norm": 7.7931706299215495,
      "language_loss": 0.77034098,
      "learning_rate": 3.3771140474320872e-06,
      "loss": 0.79272103,
      "num_input_tokens_seen": 50061925,
      "step": 2333,
      "time_per_iteration": 3.5931408405303955
    },
    {
      "auxiliary_loss_clip": 0.01167686,
      "auxiliary_loss_mlp": 0.01032477,
      "balance_loss_clip": 1.05489826,
      "balance_loss_mlp": 1.02263606,
      "epoch": 0.28064690675163834,
      "flos": 21463735875840.0,
      "grad_norm": 2.4326784765509024,
      "language_loss": 0.79771459,
      "learning_rate": 3.3765490466188664e-06,
      "loss": 0.81971622,
      "num_input_tokens_seen": 50079325,
      "step": 2334,
      "time_per_iteration": 3.6336872577667236
    },
    {
      "auxiliary_loss_clip": 0.01159598,
      "auxiliary_loss_mlp": 0.0103384,
      "balance_loss_clip": 1.05272365,
      "balance_loss_mlp": 1.02500677,
      "epoch": 0.2807671496422774,
      "flos": 20995640812800.0,
      "grad_norm": 3.1796229367392663,
      "language_loss": 0.74202758,
      "learning_rate": 3.3759838369842508e-06,
      "loss": 0.76396197,
      "num_input_tokens_seen": 50097400,
      "step": 2335,
      "time_per_iteration": 2.779179334640503
    },
    {
      "auxiliary_loss_clip": 0.01162616,
      "auxiliary_loss_mlp": 0.01025825,
      "balance_loss_clip": 1.05428648,
      "balance_loss_mlp": 1.01688981,
      "epoch": 0.2808873925329165,
      "flos": 21506577822720.0,
      "grad_norm": 1.9715535697772997,
      "language_loss": 0.73194915,
      "learning_rate": 3.375418418613981e-06,
      "loss": 0.75383353,
      "num_input_tokens_seen": 50116425,
      "step": 2336,
      "time_per_iteration": 2.7656161785125732
    },
    {
      "auxiliary_loss_clip": 0.01175548,
      "auxiliary_loss_mlp": 0.01030641,
      "balance_loss_clip": 1.05503392,
      "balance_loss_mlp": 1.02051425,
      "epoch": 0.28100763542355556,
      "flos": 16070815814400.0,
      "grad_norm": 5.340638372012982,
      "language_loss": 0.83243579,
      "learning_rate": 3.374852791593831e-06,
      "loss": 0.85449767,
      "num_input_tokens_seen": 50132625,
      "step": 2337,
      "time_per_iteration": 2.701455593109131
    },
    {
      "auxiliary_loss_clip": 0.01156317,
      "auxiliary_loss_mlp": 0.01029805,
      "balance_loss_clip": 1.05219746,
      "balance_loss_mlp": 1.01972568,
      "epoch": 0.28112787831419467,
      "flos": 19062605468160.0,
      "grad_norm": 3.6624035501222534,
      "language_loss": 0.54361433,
      "learning_rate": 3.374286956009605e-06,
      "loss": 0.56547558,
      "num_input_tokens_seen": 50151190,
      "step": 2338,
      "time_per_iteration": 2.6794064044952393
    },
    {
      "auxiliary_loss_clip": 0.01190467,
      "auxiliary_loss_mlp": 0.01033103,
      "balance_loss_clip": 1.06100678,
      "balance_loss_mlp": 1.02422786,
      "epoch": 0.2812481212048338,
      "flos": 12823629482880.0,
      "grad_norm": 2.430558146683905,
      "language_loss": 0.75278306,
      "learning_rate": 3.3737209119471405e-06,
      "loss": 0.77501881,
      "num_input_tokens_seen": 50167700,
      "step": 2339,
      "time_per_iteration": 2.6339893341064453
    },
    {
      "auxiliary_loss_clip": 0.01195815,
      "auxiliary_loss_mlp": 0.01034879,
      "balance_loss_clip": 1.05815053,
      "balance_loss_mlp": 1.02458572,
      "epoch": 0.28136836409547283,
      "flos": 15633064765440.0,
      "grad_norm": 3.08269738292447,
      "language_loss": 0.63587224,
      "learning_rate": 3.373154659492306e-06,
      "loss": 0.65817922,
      "num_input_tokens_seen": 50185840,
      "step": 2340,
      "time_per_iteration": 2.5992653369903564
    },
    {
      "auxiliary_loss_clip": 0.01180818,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 1.05659413,
      "balance_loss_mlp": 1.02147818,
      "epoch": 0.28148860698611194,
      "flos": 19933726106880.0,
      "grad_norm": 1.8227994109788508,
      "language_loss": 0.85256785,
      "learning_rate": 3.3725881987310016e-06,
      "loss": 0.87468833,
      "num_input_tokens_seen": 50203375,
      "step": 2341,
      "time_per_iteration": 2.6649835109710693
    },
    {
      "auxiliary_loss_clip": 0.01174456,
      "auxiliary_loss_mlp": 0.01031597,
      "balance_loss_clip": 1.05467331,
      "balance_loss_mlp": 1.0221082,
      "epoch": 0.28160884987675106,
      "flos": 17457219008640.0,
      "grad_norm": 1.8676261489450174,
      "language_loss": 0.87863839,
      "learning_rate": 3.372021529749159e-06,
      "loss": 0.9006989,
      "num_input_tokens_seen": 50222435,
      "step": 2342,
      "time_per_iteration": 2.6611344814300537
    },
    {
      "auxiliary_loss_clip": 0.01135256,
      "auxiliary_loss_mlp": 0.01038057,
      "balance_loss_clip": 1.05363441,
      "balance_loss_mlp": 1.02821612,
      "epoch": 0.2817290927673901,
      "flos": 16834743290880.0,
      "grad_norm": 4.108949708462741,
      "language_loss": 0.92194033,
      "learning_rate": 3.3714546526327405e-06,
      "loss": 0.94367349,
      "num_input_tokens_seen": 50240435,
      "step": 2343,
      "time_per_iteration": 2.71604585647583
    },
    {
      "auxiliary_loss_clip": 0.01165799,
      "auxiliary_loss_mlp": 0.01035659,
      "balance_loss_clip": 1.05146134,
      "balance_loss_mlp": 1.02593756,
      "epoch": 0.2818493356580292,
      "flos": 15414081500160.0,
      "grad_norm": 2.1266252811715276,
      "language_loss": 0.88349807,
      "learning_rate": 3.3708875674677423e-06,
      "loss": 0.90551269,
      "num_input_tokens_seen": 50258410,
      "step": 2344,
      "time_per_iteration": 2.7140326499938965
    },
    {
      "auxiliary_loss_clip": 0.01185868,
      "auxiliary_loss_mlp": 0.01030455,
      "balance_loss_clip": 1.05810797,
      "balance_loss_mlp": 1.02103162,
      "epoch": 0.28196957854866833,
      "flos": 20412451595520.0,
      "grad_norm": 2.8117928211943357,
      "language_loss": 0.8367548,
      "learning_rate": 3.37032027434019e-06,
      "loss": 0.85891795,
      "num_input_tokens_seen": 50277930,
      "step": 2345,
      "time_per_iteration": 2.6686089038848877
    },
    {
      "auxiliary_loss_clip": 0.01198196,
      "auxiliary_loss_mlp": 0.01036161,
      "balance_loss_clip": 1.0566932,
      "balance_loss_mlp": 1.02616501,
      "epoch": 0.2820898214393074,
      "flos": 19973120348160.0,
      "grad_norm": 4.824029872486958,
      "language_loss": 0.82920337,
      "learning_rate": 3.369752773336141e-06,
      "loss": 0.851547,
      "num_input_tokens_seen": 50297410,
      "step": 2346,
      "time_per_iteration": 2.6435608863830566
    },
    {
      "auxiliary_loss_clip": 0.01173213,
      "auxiliary_loss_mlp": 0.01033175,
      "balance_loss_clip": 1.05304217,
      "balance_loss_mlp": 1.02302432,
      "epoch": 0.2822100643299465,
      "flos": 22528308188160.0,
      "grad_norm": 1.9930865718489856,
      "language_loss": 0.78527427,
      "learning_rate": 3.3691850645416864e-06,
      "loss": 0.80733824,
      "num_input_tokens_seen": 50317120,
      "step": 2347,
      "time_per_iteration": 2.635512113571167
    },
    {
      "auxiliary_loss_clip": 0.01194288,
      "auxiliary_loss_mlp": 0.01038756,
      "balance_loss_clip": 1.05678844,
      "balance_loss_mlp": 1.02923727,
      "epoch": 0.2823303072205856,
      "flos": 11546682007680.0,
      "grad_norm": 2.740958457610266,
      "language_loss": 0.83556855,
      "learning_rate": 3.368617148042945e-06,
      "loss": 0.85789901,
      "num_input_tokens_seen": 50334790,
      "step": 2348,
      "time_per_iteration": 2.7242624759674072
    },
    {
      "auxiliary_loss_clip": 0.01170788,
      "auxiliary_loss_mlp": 0.01030641,
      "balance_loss_clip": 1.05206227,
      "balance_loss_mlp": 1.02066922,
      "epoch": 0.28245055011122466,
      "flos": 18259894281600.0,
      "grad_norm": 2.221702501327401,
      "language_loss": 0.8447262,
      "learning_rate": 3.368049023926071e-06,
      "loss": 0.86674047,
      "num_input_tokens_seen": 50353785,
      "step": 2349,
      "time_per_iteration": 2.6637260913848877
    },
    {
      "auxiliary_loss_clip": 0.01190609,
      "auxiliary_loss_mlp": 0.01030628,
      "balance_loss_clip": 1.0572927,
      "balance_loss_mlp": 1.02193165,
      "epoch": 0.2825707930018638,
      "flos": 24608110504320.0,
      "grad_norm": 1.8972147367739607,
      "language_loss": 0.83817601,
      "learning_rate": 3.3674806922772476e-06,
      "loss": 0.8603884,
      "num_input_tokens_seen": 50374670,
      "step": 2350,
      "time_per_iteration": 2.7485814094543457
    },
    {
      "auxiliary_loss_clip": 0.01166605,
      "auxiliary_loss_mlp": 0.01032933,
      "balance_loss_clip": 1.05191469,
      "balance_loss_mlp": 1.02313983,
      "epoch": 0.28269103589250283,
      "flos": 25226994862080.0,
      "grad_norm": 2.1930162062835366,
      "language_loss": 0.75180459,
      "learning_rate": 3.3669121531826904e-06,
      "loss": 0.77379996,
      "num_input_tokens_seen": 50395650,
      "step": 2351,
      "time_per_iteration": 2.7867517471313477
    },
    {
      "auxiliary_loss_clip": 0.0115689,
      "auxiliary_loss_mlp": 0.01032733,
      "balance_loss_clip": 1.05674386,
      "balance_loss_mlp": 1.02292776,
      "epoch": 0.28281127878314194,
      "flos": 19281552819840.0,
      "grad_norm": 2.309193021789076,
      "language_loss": 0.83663595,
      "learning_rate": 3.366343406728647e-06,
      "loss": 0.85853219,
      "num_input_tokens_seen": 50415100,
      "step": 2352,
      "time_per_iteration": 2.6951420307159424
    },
    {
      "auxiliary_loss_clip": 0.01180288,
      "auxiliary_loss_mlp": 0.01030778,
      "balance_loss_clip": 1.05128658,
      "balance_loss_mlp": 1.02186692,
      "epoch": 0.28293152167378105,
      "flos": 23878405710720.0,
      "grad_norm": 1.8597286148053394,
      "language_loss": 0.68668878,
      "learning_rate": 3.3657744530013946e-06,
      "loss": 0.70879948,
      "num_input_tokens_seen": 50434335,
      "step": 2353,
      "time_per_iteration": 3.5917623043060303
    },
    {
      "auxiliary_loss_clip": 0.01198688,
      "auxiliary_loss_mlp": 0.01029618,
      "balance_loss_clip": 1.05972528,
      "balance_loss_mlp": 1.01961613,
      "epoch": 0.2830517645644201,
      "flos": 43866965928960.0,
      "grad_norm": 2.642301119548798,
      "language_loss": 0.71953571,
      "learning_rate": 3.3652052920872437e-06,
      "loss": 0.74181879,
      "num_input_tokens_seen": 50457200,
      "step": 2354,
      "time_per_iteration": 2.8305740356445312
    },
    {
      "auxiliary_loss_clip": 0.0117896,
      "auxiliary_loss_mlp": 0.01035057,
      "balance_loss_clip": 1.05553424,
      "balance_loss_mlp": 1.02558625,
      "epoch": 0.2831720074550592,
      "flos": 26651750803200.0,
      "grad_norm": 1.8902573193071233,
      "language_loss": 0.85439122,
      "learning_rate": 3.3646359240725355e-06,
      "loss": 0.87653142,
      "num_input_tokens_seen": 50476390,
      "step": 2355,
      "time_per_iteration": 3.6528236865997314
    },
    {
      "auxiliary_loss_clip": 0.0118502,
      "auxiliary_loss_mlp": 0.00764729,
      "balance_loss_clip": 1.05549729,
      "balance_loss_mlp": 1.00055289,
      "epoch": 0.2832922503456983,
      "flos": 31029979564800.0,
      "grad_norm": 2.3434940527708483,
      "language_loss": 0.67822748,
      "learning_rate": 3.364066349043643e-06,
      "loss": 0.69772494,
      "num_input_tokens_seen": 50497595,
      "step": 2356,
      "time_per_iteration": 2.7397255897521973
    },
    {
      "auxiliary_loss_clip": 0.01170474,
      "auxiliary_loss_mlp": 0.01030258,
      "balance_loss_clip": 1.05227256,
      "balance_loss_mlp": 1.0217346,
      "epoch": 0.2834124932363374,
      "flos": 20405699838720.0,
      "grad_norm": 1.7449124731344097,
      "language_loss": 0.82115197,
      "learning_rate": 3.363496567086969e-06,
      "loss": 0.84315932,
      "num_input_tokens_seen": 50514690,
      "step": 2357,
      "time_per_iteration": 2.610638380050659
    },
    {
      "auxiliary_loss_clip": 0.01206112,
      "auxiliary_loss_mlp": 0.01037043,
      "balance_loss_clip": 1.05990195,
      "balance_loss_mlp": 1.02743506,
      "epoch": 0.2835327361269765,
      "flos": 39384848056320.0,
      "grad_norm": 1.9838583069343096,
      "language_loss": 0.75989985,
      "learning_rate": 3.3629265782889506e-06,
      "loss": 0.78233135,
      "num_input_tokens_seen": 50536515,
      "step": 2358,
      "time_per_iteration": 2.753045082092285
    },
    {
      "auxiliary_loss_clip": 0.01152291,
      "auxiliary_loss_mlp": 0.01032508,
      "balance_loss_clip": 1.04792845,
      "balance_loss_mlp": 1.02260804,
      "epoch": 0.2836529790176156,
      "flos": 30261598801920.0,
      "grad_norm": 2.089831767182233,
      "language_loss": 0.72149885,
      "learning_rate": 3.362356382736054e-06,
      "loss": 0.74334681,
      "num_input_tokens_seen": 50557120,
      "step": 2359,
      "time_per_iteration": 3.6589319705963135
    },
    {
      "auxiliary_loss_clip": 0.01158839,
      "auxiliary_loss_mlp": 0.01028384,
      "balance_loss_clip": 1.0491538,
      "balance_loss_mlp": 1.01865053,
      "epoch": 0.28377322190825466,
      "flos": 12677796264960.0,
      "grad_norm": 1.865901441084261,
      "language_loss": 0.91042376,
      "learning_rate": 3.361785980514777e-06,
      "loss": 0.93229598,
      "num_input_tokens_seen": 50573320,
      "step": 2360,
      "time_per_iteration": 3.597364664077759
    },
    {
      "auxiliary_loss_clip": 0.01125839,
      "auxiliary_loss_mlp": 0.01031932,
      "balance_loss_clip": 1.04859495,
      "balance_loss_mlp": 1.02259779,
      "epoch": 0.28389346479889377,
      "flos": 18296666830080.0,
      "grad_norm": 6.276199404584636,
      "language_loss": 0.7637834,
      "learning_rate": 3.361215371711649e-06,
      "loss": 0.78536111,
      "num_input_tokens_seen": 50592415,
      "step": 2361,
      "time_per_iteration": 2.7613251209259033
    },
    {
      "auxiliary_loss_clip": 0.01152738,
      "auxiliary_loss_mlp": 0.01027495,
      "balance_loss_clip": 1.05158329,
      "balance_loss_mlp": 1.0181725,
      "epoch": 0.2840137076895329,
      "flos": 20406992728320.0,
      "grad_norm": 1.832699020278592,
      "language_loss": 0.83609688,
      "learning_rate": 3.3606445564132326e-06,
      "loss": 0.85789925,
      "num_input_tokens_seen": 50609710,
      "step": 2362,
      "time_per_iteration": 2.6977882385253906
    },
    {
      "auxiliary_loss_clip": 0.01208895,
      "auxiliary_loss_mlp": 0.0076411,
      "balance_loss_clip": 1.06083751,
      "balance_loss_mlp": 1.00033307,
      "epoch": 0.28413395058017193,
      "flos": 20048030161920.0,
      "grad_norm": 2.0901393834707465,
      "language_loss": 0.82397544,
      "learning_rate": 3.360073534706118e-06,
      "loss": 0.84370553,
      "num_input_tokens_seen": 50626865,
      "step": 2363,
      "time_per_iteration": 2.642947196960449
    },
    {
      "auxiliary_loss_clip": 0.01179062,
      "auxiliary_loss_mlp": 0.01039641,
      "balance_loss_clip": 1.05773568,
      "balance_loss_mlp": 1.02938318,
      "epoch": 0.28425419347081105,
      "flos": 37663613256960.0,
      "grad_norm": 2.2255208551177668,
      "language_loss": 0.76164031,
      "learning_rate": 3.35950230667693e-06,
      "loss": 0.7838273,
      "num_input_tokens_seen": 50648560,
      "step": 2364,
      "time_per_iteration": 2.828190565109253
    },
    {
      "auxiliary_loss_clip": 0.01191679,
      "auxiliary_loss_mlp": 0.01034338,
      "balance_loss_clip": 1.05609524,
      "balance_loss_mlp": 1.02422333,
      "epoch": 0.28437443636145016,
      "flos": 13845072539520.0,
      "grad_norm": 2.90331377882048,
      "language_loss": 0.86436778,
      "learning_rate": 3.358930872412323e-06,
      "loss": 0.88662791,
      "num_input_tokens_seen": 50665725,
      "step": 2365,
      "time_per_iteration": 2.589320421218872
    },
    {
      "auxiliary_loss_clip": 0.01189504,
      "auxiliary_loss_mlp": 0.01034233,
      "balance_loss_clip": 1.05647612,
      "balance_loss_mlp": 1.02497661,
      "epoch": 0.2844946792520892,
      "flos": 22747794243840.0,
      "grad_norm": 1.9133376359335028,
      "language_loss": 0.81353688,
      "learning_rate": 3.3583592319989825e-06,
      "loss": 0.8357743,
      "num_input_tokens_seen": 50685095,
      "step": 2366,
      "time_per_iteration": 2.6664140224456787
    },
    {
      "auxiliary_loss_clip": 0.01203317,
      "auxiliary_loss_mlp": 0.0103846,
      "balance_loss_clip": 1.06035662,
      "balance_loss_mlp": 1.02759457,
      "epoch": 0.2846149221427283,
      "flos": 32415987709440.0,
      "grad_norm": 1.8777296327517488,
      "language_loss": 0.68705016,
      "learning_rate": 3.357787385523627e-06,
      "loss": 0.70946789,
      "num_input_tokens_seen": 50706500,
      "step": 2367,
      "time_per_iteration": 2.7343101501464844
    },
    {
      "auxiliary_loss_clip": 0.01141287,
      "auxiliary_loss_mlp": 0.01033645,
      "balance_loss_clip": 1.05096889,
      "balance_loss_mlp": 1.02422726,
      "epoch": 0.2847351650333674,
      "flos": 28475976873600.0,
      "grad_norm": 2.136752091524356,
      "language_loss": 0.82838303,
      "learning_rate": 3.3572153330730048e-06,
      "loss": 0.85013235,
      "num_input_tokens_seen": 50727595,
      "step": 2368,
      "time_per_iteration": 2.824789524078369
    },
    {
      "auxiliary_loss_clip": 0.01070108,
      "auxiliary_loss_mlp": 0.01009358,
      "balance_loss_clip": 1.02103043,
      "balance_loss_mlp": 1.00752199,
      "epoch": 0.2848554079240065,
      "flos": 55753399704960.0,
      "grad_norm": 0.8275862869796837,
      "language_loss": 0.64750534,
      "learning_rate": 3.3566430747338956e-06,
      "loss": 0.66829997,
      "num_input_tokens_seen": 50782800,
      "step": 2369,
      "time_per_iteration": 3.1232106685638428
    },
    {
      "auxiliary_loss_clip": 0.01190909,
      "auxiliary_loss_mlp": 0.01032935,
      "balance_loss_clip": 1.05415356,
      "balance_loss_mlp": 1.02364206,
      "epoch": 0.2849756508146456,
      "flos": 11836875985920.0,
      "grad_norm": 2.3433930803403826,
      "language_loss": 0.86371946,
      "learning_rate": 3.35607061059311e-06,
      "loss": 0.88595784,
      "num_input_tokens_seen": 50797730,
      "step": 2370,
      "time_per_iteration": 2.651404619216919
    },
    {
      "auxiliary_loss_clip": 0.01202915,
      "auxiliary_loss_mlp": 0.01032987,
      "balance_loss_clip": 1.05829644,
      "balance_loss_mlp": 1.02365875,
      "epoch": 0.28509589370528465,
      "flos": 25155209531520.0,
      "grad_norm": 2.200817479193329,
      "language_loss": 0.75336301,
      "learning_rate": 3.3554979407374917e-06,
      "loss": 0.77572203,
      "num_input_tokens_seen": 50819840,
      "step": 2371,
      "time_per_iteration": 2.699162483215332
    },
    {
      "auxiliary_loss_clip": 0.01191985,
      "auxiliary_loss_mlp": 0.01038596,
      "balance_loss_clip": 1.05625653,
      "balance_loss_mlp": 1.02965569,
      "epoch": 0.28521613659592376,
      "flos": 19974808287360.0,
      "grad_norm": 1.8804384194775674,
      "language_loss": 0.73713142,
      "learning_rate": 3.3549250652539134e-06,
      "loss": 0.7594372,
      "num_input_tokens_seen": 50838935,
      "step": 2372,
      "time_per_iteration": 2.6486406326293945
    },
    {
      "auxiliary_loss_clip": 0.01172165,
      "auxiliary_loss_mlp": 0.01034215,
      "balance_loss_clip": 1.05029178,
      "balance_loss_mlp": 1.02389109,
      "epoch": 0.2853363794865629,
      "flos": 23367971491200.0,
      "grad_norm": 2.4775226470520253,
      "language_loss": 0.81620634,
      "learning_rate": 3.3543519842292794e-06,
      "loss": 0.83827013,
      "num_input_tokens_seen": 50858590,
      "step": 2373,
      "time_per_iteration": 2.666287660598755
    },
    {
      "auxiliary_loss_clip": 0.01202244,
      "auxiliary_loss_mlp": 0.00764412,
      "balance_loss_clip": 1.05640221,
      "balance_loss_mlp": 1.00039399,
      "epoch": 0.28545662237720193,
      "flos": 19861940776320.0,
      "grad_norm": 1.9474675456926365,
      "language_loss": 0.83626366,
      "learning_rate": 3.353778697750527e-06,
      "loss": 0.85593027,
      "num_input_tokens_seen": 50876995,
      "step": 2374,
      "time_per_iteration": 2.611072301864624
    },
    {
      "auxiliary_loss_clip": 0.01167066,
      "auxiliary_loss_mlp": 0.0103257,
      "balance_loss_clip": 1.05200422,
      "balance_loss_mlp": 1.02310479,
      "epoch": 0.28557686526784104,
      "flos": 23879016241920.0,
      "grad_norm": 1.642586791783303,
      "language_loss": 0.89424503,
      "learning_rate": 3.353205205904622e-06,
      "loss": 0.91624141,
      "num_input_tokens_seen": 50896105,
      "step": 2375,
      "time_per_iteration": 2.678462266921997
    },
    {
      "auxiliary_loss_clip": 0.01174128,
      "auxiliary_loss_mlp": 0.01034095,
      "balance_loss_clip": 1.05490589,
      "balance_loss_mlp": 1.02508235,
      "epoch": 0.28569710815848015,
      "flos": 44890384233600.0,
      "grad_norm": 2.424116785006551,
      "language_loss": 0.719051,
      "learning_rate": 3.3526315087785637e-06,
      "loss": 0.74113321,
      "num_input_tokens_seen": 50917220,
      "step": 2376,
      "time_per_iteration": 2.8992092609405518
    },
    {
      "auxiliary_loss_clip": 0.01127671,
      "auxiliary_loss_mlp": 0.01036265,
      "balance_loss_clip": 1.04735553,
      "balance_loss_mlp": 1.0265851,
      "epoch": 0.2858173510491192,
      "flos": 26829759628800.0,
      "grad_norm": 1.8885922861898812,
      "language_loss": 0.80657971,
      "learning_rate": 3.3520576064593805e-06,
      "loss": 0.82821906,
      "num_input_tokens_seen": 50937175,
      "step": 2377,
      "time_per_iteration": 2.7777557373046875
    },
    {
      "auxiliary_loss_clip": 0.01196312,
      "auxiliary_loss_mlp": 0.01040952,
      "balance_loss_clip": 1.05845571,
      "balance_loss_mlp": 1.03117132,
      "epoch": 0.2859375939397583,
      "flos": 23148916398720.0,
      "grad_norm": 1.6856780030056797,
      "language_loss": 0.82096159,
      "learning_rate": 3.3514834990341337e-06,
      "loss": 0.8433342,
      "num_input_tokens_seen": 50957500,
      "step": 2378,
      "time_per_iteration": 3.6613659858703613
    },
    {
      "auxiliary_loss_clip": 0.011802,
      "auxiliary_loss_mlp": 0.01027797,
      "balance_loss_clip": 1.05416417,
      "balance_loss_mlp": 1.01849818,
      "epoch": 0.2860578368303974,
      "flos": 12129799397760.0,
      "grad_norm": 4.0444912001607225,
      "language_loss": 0.92894399,
      "learning_rate": 3.3509091865899144e-06,
      "loss": 0.95102394,
      "num_input_tokens_seen": 50972690,
      "step": 2379,
      "time_per_iteration": 2.6827635765075684
    },
    {
      "auxiliary_loss_clip": 0.01204371,
      "auxiliary_loss_mlp": 0.01034577,
      "balance_loss_clip": 1.05646777,
      "balance_loss_mlp": 1.02449203,
      "epoch": 0.2861780797210365,
      "flos": 19938035738880.0,
      "grad_norm": 4.067698797997826,
      "language_loss": 0.70658076,
      "learning_rate": 3.350334669213846e-06,
      "loss": 0.72897029,
      "num_input_tokens_seen": 50990095,
      "step": 2380,
      "time_per_iteration": 2.5419352054595947
    },
    {
      "auxiliary_loss_clip": 0.01191781,
      "auxiliary_loss_mlp": 0.0102952,
      "balance_loss_clip": 1.05905056,
      "balance_loss_mlp": 1.01994705,
      "epoch": 0.2862983226116756,
      "flos": 27563127609600.0,
      "grad_norm": 3.428783396565654,
      "language_loss": 0.75500453,
      "learning_rate": 3.3497599469930816e-06,
      "loss": 0.77721751,
      "num_input_tokens_seen": 51008305,
      "step": 2381,
      "time_per_iteration": 3.5748863220214844
    },
    {
      "auxiliary_loss_clip": 0.01205557,
      "auxiliary_loss_mlp": 0.01032438,
      "balance_loss_clip": 1.05640197,
      "balance_loss_mlp": 1.02322936,
      "epoch": 0.28641856550231465,
      "flos": 22053964158720.0,
      "grad_norm": 3.823761988096933,
      "language_loss": 0.83538461,
      "learning_rate": 3.349185020014807e-06,
      "loss": 0.85776448,
      "num_input_tokens_seen": 51025570,
      "step": 2382,
      "time_per_iteration": 2.544820547103882
    },
    {
      "auxiliary_loss_clip": 0.01189352,
      "auxiliary_loss_mlp": 0.01036556,
      "balance_loss_clip": 1.05187106,
      "balance_loss_mlp": 1.02620244,
      "epoch": 0.28653880839295376,
      "flos": 22378775869440.0,
      "grad_norm": 8.321271629868242,
      "language_loss": 0.74774051,
      "learning_rate": 3.348609888366237e-06,
      "loss": 0.76999956,
      "num_input_tokens_seen": 51044585,
      "step": 2383,
      "time_per_iteration": 2.624713182449341
    },
    {
      "auxiliary_loss_clip": 0.01126788,
      "auxiliary_loss_mlp": 0.01034081,
      "balance_loss_clip": 1.04797971,
      "balance_loss_mlp": 1.02477121,
      "epoch": 0.28665905128359287,
      "flos": 23367971491200.0,
      "grad_norm": 2.3556798549128173,
      "language_loss": 0.63184077,
      "learning_rate": 3.348034552134619e-06,
      "loss": 0.65344948,
      "num_input_tokens_seen": 51063990,
      "step": 2384,
      "time_per_iteration": 2.7318496704101562
    },
    {
      "auxiliary_loss_clip": 0.01137432,
      "auxiliary_loss_mlp": 0.01032716,
      "balance_loss_clip": 1.04958856,
      "balance_loss_mlp": 1.02360892,
      "epoch": 0.2867792941742319,
      "flos": 20881695893760.0,
      "grad_norm": 2.3551537662458877,
      "language_loss": 0.84326243,
      "learning_rate": 3.3474590114072316e-06,
      "loss": 0.86496389,
      "num_input_tokens_seen": 51081990,
      "step": 2385,
      "time_per_iteration": 4.521415948867798
    },
    {
      "auxiliary_loss_clip": 0.01156555,
      "auxiliary_loss_mlp": 0.01030791,
      "balance_loss_clip": 1.05448937,
      "balance_loss_mlp": 1.02105129,
      "epoch": 0.28689953706487104,
      "flos": 20664005518080.0,
      "grad_norm": 2.1050925993417704,
      "language_loss": 0.82653689,
      "learning_rate": 3.3468832662713836e-06,
      "loss": 0.84841037,
      "num_input_tokens_seen": 51100235,
      "step": 2386,
      "time_per_iteration": 2.728040933609009
    },
    {
      "auxiliary_loss_clip": 0.01160352,
      "auxiliary_loss_mlp": 0.01032063,
      "balance_loss_clip": 1.05347872,
      "balance_loss_mlp": 1.02263975,
      "epoch": 0.28701977995551015,
      "flos": 12675533708160.0,
      "grad_norm": 3.1015702629130644,
      "language_loss": 0.8372196,
      "learning_rate": 3.346307316814415e-06,
      "loss": 0.85914373,
      "num_input_tokens_seen": 51115405,
      "step": 2387,
      "time_per_iteration": 2.616672992706299
    },
    {
      "auxiliary_loss_clip": 0.01191479,
      "auxiliary_loss_mlp": 0.01034593,
      "balance_loss_clip": 1.05790448,
      "balance_loss_mlp": 1.02413273,
      "epoch": 0.2871400228461492,
      "flos": 21252366293760.0,
      "grad_norm": 2.1915325339577243,
      "language_loss": 0.75968874,
      "learning_rate": 3.3457311631236965e-06,
      "loss": 0.78194952,
      "num_input_tokens_seen": 51136390,
      "step": 2388,
      "time_per_iteration": 2.659308671951294
    },
    {
      "auxiliary_loss_clip": 0.01160862,
      "auxiliary_loss_mlp": 0.01029355,
      "balance_loss_clip": 1.04882979,
      "balance_loss_mlp": 1.01990736,
      "epoch": 0.2872602657367883,
      "flos": 25119262995840.0,
      "grad_norm": 2.152616623258441,
      "language_loss": 0.84515446,
      "learning_rate": 3.345154805286631e-06,
      "loss": 0.86705667,
      "num_input_tokens_seen": 51156650,
      "step": 2389,
      "time_per_iteration": 2.67872953414917
    },
    {
      "auxiliary_loss_clip": 0.01182006,
      "auxiliary_loss_mlp": 0.01034028,
      "balance_loss_clip": 1.05254364,
      "balance_loss_mlp": 1.02475905,
      "epoch": 0.2873805086274274,
      "flos": 16646606830080.0,
      "grad_norm": 2.527725482025029,
      "language_loss": 0.76417685,
      "learning_rate": 3.344578243390651e-06,
      "loss": 0.78633714,
      "num_input_tokens_seen": 51172210,
      "step": 2390,
      "time_per_iteration": 2.587218761444092
    },
    {
      "auxiliary_loss_clip": 0.01173762,
      "auxiliary_loss_mlp": 0.01031829,
      "balance_loss_clip": 1.05412269,
      "balance_loss_mlp": 1.02263761,
      "epoch": 0.2875007515180665,
      "flos": 17420123237760.0,
      "grad_norm": 3.0106549331136128,
      "language_loss": 0.78881091,
      "learning_rate": 3.3440014775232206e-06,
      "loss": 0.81086683,
      "num_input_tokens_seen": 51190265,
      "step": 2391,
      "time_per_iteration": 2.6404542922973633
    },
    {
      "auxiliary_loss_clip": 0.01161145,
      "auxiliary_loss_mlp": 0.01033726,
      "balance_loss_clip": 1.05148983,
      "balance_loss_mlp": 1.02486849,
      "epoch": 0.2876209944087056,
      "flos": 23434190213760.0,
      "grad_norm": 2.5821261794465475,
      "language_loss": 0.71506053,
      "learning_rate": 3.343424507771834e-06,
      "loss": 0.73700923,
      "num_input_tokens_seen": 51208475,
      "step": 2392,
      "time_per_iteration": 2.777653455734253
    },
    {
      "auxiliary_loss_clip": 0.01162866,
      "auxiliary_loss_mlp": 0.01028026,
      "balance_loss_clip": 1.05521345,
      "balance_loss_mlp": 1.01915669,
      "epoch": 0.2877412372993447,
      "flos": 13735509079680.0,
      "grad_norm": 2.4204264317444593,
      "language_loss": 0.86592531,
      "learning_rate": 3.342847334224018e-06,
      "loss": 0.88783419,
      "num_input_tokens_seen": 51225875,
      "step": 2393,
      "time_per_iteration": 2.7588393688201904
    },
    {
      "auxiliary_loss_clip": 0.01088028,
      "auxiliary_loss_mlp": 0.01003552,
      "balance_loss_clip": 1.01688385,
      "balance_loss_mlp": 1.00176382,
      "epoch": 0.28786148018998375,
      "flos": 58079695104000.0,
      "grad_norm": 0.9770628880035306,
      "language_loss": 0.62375307,
      "learning_rate": 3.342269956967329e-06,
      "loss": 0.64466888,
      "num_input_tokens_seen": 51287780,
      "step": 2394,
      "time_per_iteration": 3.241800546646118
    },
    {
      "auxiliary_loss_clip": 0.01191198,
      "auxiliary_loss_mlp": 0.01032443,
      "balance_loss_clip": 1.05471504,
      "balance_loss_mlp": 1.02258432,
      "epoch": 0.28798172308062286,
      "flos": 23435052140160.0,
      "grad_norm": 2.977586888667627,
      "language_loss": 0.71994209,
      "learning_rate": 3.341692376089355e-06,
      "loss": 0.74217856,
      "num_input_tokens_seen": 51303335,
      "step": 2395,
      "time_per_iteration": 2.6765310764312744
    },
    {
      "auxiliary_loss_clip": 0.01187749,
      "auxiliary_loss_mlp": 0.01029072,
      "balance_loss_clip": 1.05584168,
      "balance_loss_mlp": 1.01996422,
      "epoch": 0.288101965971262,
      "flos": 25110033200640.0,
      "grad_norm": 4.398265863393965,
      "language_loss": 0.84220457,
      "learning_rate": 3.3411145916777146e-06,
      "loss": 0.86437279,
      "num_input_tokens_seen": 51317495,
      "step": 2396,
      "time_per_iteration": 2.623741626739502
    },
    {
      "auxiliary_loss_clip": 0.01168146,
      "auxiliary_loss_mlp": 0.01031953,
      "balance_loss_clip": 1.05085969,
      "balance_loss_mlp": 1.02235651,
      "epoch": 0.28822220886190103,
      "flos": 16252559654400.0,
      "grad_norm": 2.7655052855531177,
      "language_loss": 0.91208339,
      "learning_rate": 3.3405366038200566e-06,
      "loss": 0.93408442,
      "num_input_tokens_seen": 51336430,
      "step": 2397,
      "time_per_iteration": 2.686751365661621
    },
    {
      "auxiliary_loss_clip": 0.01177373,
      "auxiliary_loss_mlp": 0.01033793,
      "balance_loss_clip": 1.05693805,
      "balance_loss_mlp": 1.02362466,
      "epoch": 0.28834245175254014,
      "flos": 24535642815360.0,
      "grad_norm": 2.2529085106580595,
      "language_loss": 0.84806389,
      "learning_rate": 3.3399584126040617e-06,
      "loss": 0.87017554,
      "num_input_tokens_seen": 51355930,
      "step": 2398,
      "time_per_iteration": 2.668748378753662
    },
    {
      "auxiliary_loss_clip": 0.01202342,
      "auxiliary_loss_mlp": 0.0076418,
      "balance_loss_clip": 1.0555377,
      "balance_loss_mlp": 1.00034642,
      "epoch": 0.2884626946431792,
      "flos": 24571445696640.0,
      "grad_norm": 2.9083317675200426,
      "language_loss": 0.91380656,
      "learning_rate": 3.339380018117441e-06,
      "loss": 0.93347174,
      "num_input_tokens_seen": 51376765,
      "step": 2399,
      "time_per_iteration": 2.700516700744629
    },
    {
      "auxiliary_loss_clip": 0.01184681,
      "auxiliary_loss_mlp": 0.01027963,
      "balance_loss_clip": 1.05560267,
      "balance_loss_mlp": 1.01876569,
      "epoch": 0.2885829375338183,
      "flos": 16544657053440.0,
      "grad_norm": 5.228414917706568,
      "language_loss": 0.78581941,
      "learning_rate": 3.3388014204479366e-06,
      "loss": 0.80794585,
      "num_input_tokens_seen": 51394570,
      "step": 2400,
      "time_per_iteration": 2.622215986251831
    },
    {
      "auxiliary_loss_clip": 0.01202886,
      "auxiliary_loss_mlp": 0.01033788,
      "balance_loss_clip": 1.05508375,
      "balance_loss_mlp": 1.02403045,
      "epoch": 0.2887031804244574,
      "flos": 24061226958720.0,
      "grad_norm": 4.516009038903702,
      "language_loss": 0.91647804,
      "learning_rate": 3.338222619683321e-06,
      "loss": 0.93884474,
      "num_input_tokens_seen": 51414535,
      "step": 2401,
      "time_per_iteration": 2.622659683227539
    },
    {
      "auxiliary_loss_clip": 0.01174317,
      "auxiliary_loss_mlp": 0.01035095,
      "balance_loss_clip": 1.05212212,
      "balance_loss_mlp": 1.02480745,
      "epoch": 0.2888234233150965,
      "flos": 23330696152320.0,
      "grad_norm": 2.3729915691173216,
      "language_loss": 0.74016118,
      "learning_rate": 3.337643615911398e-06,
      "loss": 0.76225531,
      "num_input_tokens_seen": 51434160,
      "step": 2402,
      "time_per_iteration": 2.7715706825256348
    },
    {
      "auxiliary_loss_clip": 0.0118831,
      "auxiliary_loss_mlp": 0.01032367,
      "balance_loss_clip": 1.05354106,
      "balance_loss_mlp": 1.0229975,
      "epoch": 0.2889436662057356,
      "flos": 22272767856000.0,
      "grad_norm": 3.683729262652092,
      "language_loss": 0.78875661,
      "learning_rate": 3.3370644092200026e-06,
      "loss": 0.81096339,
      "num_input_tokens_seen": 51451435,
      "step": 2403,
      "time_per_iteration": 2.652864933013916
    },
    {
      "auxiliary_loss_clip": 0.01143688,
      "auxiliary_loss_mlp": 0.01036459,
      "balance_loss_clip": 1.04569805,
      "balance_loss_mlp": 1.02731538,
      "epoch": 0.2890639090963747,
      "flos": 21616931381760.0,
      "grad_norm": 1.8529550867152302,
      "language_loss": 0.78243172,
      "learning_rate": 3.3364849996969985e-06,
      "loss": 0.80423319,
      "num_input_tokens_seen": 51471455,
      "step": 2404,
      "time_per_iteration": 2.7073814868927
    },
    {
      "auxiliary_loss_clip": 0.01189567,
      "auxiliary_loss_mlp": 0.01033572,
      "balance_loss_clip": 1.0562923,
      "balance_loss_mlp": 1.02442265,
      "epoch": 0.28918415198701375,
      "flos": 28585540333440.0,
      "grad_norm": 2.245727133306413,
      "language_loss": 0.85530466,
      "learning_rate": 3.335905387430283e-06,
      "loss": 0.87753606,
      "num_input_tokens_seen": 51492890,
      "step": 2405,
      "time_per_iteration": 3.5333008766174316
    },
    {
      "auxiliary_loss_clip": 0.01174395,
      "auxiliary_loss_mlp": 0.01033304,
      "balance_loss_clip": 1.04991603,
      "balance_loss_mlp": 1.02405906,
      "epoch": 0.28930439487765286,
      "flos": 21944688007680.0,
      "grad_norm": 1.9046157277898264,
      "language_loss": 0.83213198,
      "learning_rate": 3.335325572507782e-06,
      "loss": 0.85420901,
      "num_input_tokens_seen": 51513390,
      "step": 2406,
      "time_per_iteration": 2.6870779991149902
    },
    {
      "auxiliary_loss_clip": 0.01206539,
      "auxiliary_loss_mlp": 0.00764851,
      "balance_loss_clip": 1.06024718,
      "balance_loss_mlp": 1.00054097,
      "epoch": 0.28942463776829197,
      "flos": 19281911955840.0,
      "grad_norm": 2.8538677621758586,
      "language_loss": 0.742935,
      "learning_rate": 3.3347455550174537e-06,
      "loss": 0.76264894,
      "num_input_tokens_seen": 51532730,
      "step": 2407,
      "time_per_iteration": 2.5980687141418457
    },
    {
      "auxiliary_loss_clip": 0.01153283,
      "auxiliary_loss_mlp": 0.01027564,
      "balance_loss_clip": 1.04847312,
      "balance_loss_mlp": 1.01762211,
      "epoch": 0.289544880658931,
      "flos": 14645700737280.0,
      "grad_norm": 2.093581379731839,
      "language_loss": 0.68122798,
      "learning_rate": 3.3341653350472864e-06,
      "loss": 0.70303649,
      "num_input_tokens_seen": 51549560,
      "step": 2408,
      "time_per_iteration": 3.610114574432373
    },
    {
      "auxiliary_loss_clip": 0.01212234,
      "auxiliary_loss_mlp": 0.01040243,
      "balance_loss_clip": 1.05917954,
      "balance_loss_mlp": 1.02940094,
      "epoch": 0.28966512354957014,
      "flos": 28621881918720.0,
      "grad_norm": 2.7458599108900383,
      "language_loss": 0.6931839,
      "learning_rate": 3.333584912685298e-06,
      "loss": 0.71570867,
      "num_input_tokens_seen": 51568180,
      "step": 2409,
      "time_per_iteration": 2.6895904541015625
    },
    {
      "auxiliary_loss_clip": 0.01061372,
      "auxiliary_loss_mlp": 0.01001843,
      "balance_loss_clip": 1.01579273,
      "balance_loss_mlp": 0.99981624,
      "epoch": 0.28978536644020925,
      "flos": 64711784511360.0,
      "grad_norm": 0.8789924063458642,
      "language_loss": 0.55554307,
      "learning_rate": 3.3330042880195385e-06,
      "loss": 0.57617521,
      "num_input_tokens_seen": 51622530,
      "step": 2410,
      "time_per_iteration": 3.1711766719818115
    },
    {
      "auxiliary_loss_clip": 0.01171671,
      "auxiliary_loss_mlp": 0.01033285,
      "balance_loss_clip": 1.05197453,
      "balance_loss_mlp": 1.02424884,
      "epoch": 0.2899056093308483,
      "flos": 18624638937600.0,
      "grad_norm": 2.078522745397886,
      "language_loss": 0.78414595,
      "learning_rate": 3.3324234611380888e-06,
      "loss": 0.8061955,
      "num_input_tokens_seen": 51641260,
      "step": 2411,
      "time_per_iteration": 3.6079792976379395
    },
    {
      "auxiliary_loss_clip": 0.01153658,
      "auxiliary_loss_mlp": 0.01027833,
      "balance_loss_clip": 1.05251074,
      "balance_loss_mlp": 1.01834369,
      "epoch": 0.2900258522214874,
      "flos": 22893735202560.0,
      "grad_norm": 1.925096488460499,
      "language_loss": 0.81712049,
      "learning_rate": 3.3318424321290596e-06,
      "loss": 0.83893538,
      "num_input_tokens_seen": 51660975,
      "step": 2412,
      "time_per_iteration": 2.7638180255889893
    },
    {
      "auxiliary_loss_clip": 0.01062571,
      "auxiliary_loss_mlp": 0.01004533,
      "balance_loss_clip": 1.01530004,
      "balance_loss_mlp": 1.00267375,
      "epoch": 0.2901460951121265,
      "flos": 71106036013440.0,
      "grad_norm": 0.827019108291743,
      "language_loss": 0.59916425,
      "learning_rate": 3.3312612010805917e-06,
      "loss": 0.61983526,
      "num_input_tokens_seen": 51720550,
      "step": 2413,
      "time_per_iteration": 3.3167805671691895
    },
    {
      "auxiliary_loss_clip": 0.01163299,
      "auxiliary_loss_mlp": 0.01034659,
      "balance_loss_clip": 1.05216837,
      "balance_loss_mlp": 1.02545607,
      "epoch": 0.2902663380027656,
      "flos": 32160986081280.0,
      "grad_norm": 1.9139314560332206,
      "language_loss": 0.70266998,
      "learning_rate": 3.330679768080858e-06,
      "loss": 0.72464955,
      "num_input_tokens_seen": 51744435,
      "step": 2414,
      "time_per_iteration": 2.832214832305908
    },
    {
      "auxiliary_loss_clip": 0.01188876,
      "auxiliary_loss_mlp": 0.01035049,
      "balance_loss_clip": 1.0580399,
      "balance_loss_mlp": 1.0255425,
      "epoch": 0.2903865808934047,
      "flos": 29351658539520.0,
      "grad_norm": 2.315234144692434,
      "language_loss": 0.83911312,
      "learning_rate": 3.3300981332180627e-06,
      "loss": 0.86135238,
      "num_input_tokens_seen": 51763640,
      "step": 2415,
      "time_per_iteration": 2.6691818237304688
    },
    {
      "auxiliary_loss_clip": 0.01167299,
      "auxiliary_loss_mlp": 0.01039547,
      "balance_loss_clip": 1.05370235,
      "balance_loss_mlp": 1.03017116,
      "epoch": 0.29050682378404374,
      "flos": 17089026647040.0,
      "grad_norm": 2.0643627793197763,
      "language_loss": 0.79983377,
      "learning_rate": 3.3295162965804373e-06,
      "loss": 0.82190228,
      "num_input_tokens_seen": 51782135,
      "step": 2416,
      "time_per_iteration": 2.67122483253479
    },
    {
      "auxiliary_loss_clip": 0.01158996,
      "auxiliary_loss_mlp": 0.01028976,
      "balance_loss_clip": 1.05566764,
      "balance_loss_mlp": 1.02044654,
      "epoch": 0.29062706667468285,
      "flos": 17858233422720.0,
      "grad_norm": 2.813231956991232,
      "language_loss": 0.78821075,
      "learning_rate": 3.328934258256247e-06,
      "loss": 0.81009042,
      "num_input_tokens_seen": 51800200,
      "step": 2417,
      "time_per_iteration": 2.6374270915985107
    },
    {
      "auxiliary_loss_clip": 0.0118599,
      "auxiliary_loss_mlp": 0.0103367,
      "balance_loss_clip": 1.05178189,
      "balance_loss_mlp": 1.02424026,
      "epoch": 0.29074730956532197,
      "flos": 24279815174400.0,
      "grad_norm": 6.008853151699176,
      "language_loss": 0.67290795,
      "learning_rate": 3.3283520183337856e-06,
      "loss": 0.69510454,
      "num_input_tokens_seen": 51819905,
      "step": 2418,
      "time_per_iteration": 2.6699626445770264
    },
    {
      "auxiliary_loss_clip": 0.0117069,
      "auxiliary_loss_mlp": 0.01034913,
      "balance_loss_clip": 1.05231977,
      "balance_loss_mlp": 1.02625275,
      "epoch": 0.290867552455961,
      "flos": 22340961826560.0,
      "grad_norm": 1.9157815318305342,
      "language_loss": 0.6960302,
      "learning_rate": 3.3277695769013797e-06,
      "loss": 0.71808624,
      "num_input_tokens_seen": 51839350,
      "step": 2419,
      "time_per_iteration": 2.689283609390259
    },
    {
      "auxiliary_loss_clip": 0.01186664,
      "auxiliary_loss_mlp": 0.01035933,
      "balance_loss_clip": 1.05580473,
      "balance_loss_mlp": 1.02575898,
      "epoch": 0.29098779534660013,
      "flos": 23186155824000.0,
      "grad_norm": 2.953332792148371,
      "language_loss": 0.77559125,
      "learning_rate": 3.327186934047385e-06,
      "loss": 0.79781723,
      "num_input_tokens_seen": 51858045,
      "step": 2420,
      "time_per_iteration": 2.6506004333496094
    },
    {
      "auxiliary_loss_clip": 0.01160566,
      "auxiliary_loss_mlp": 0.01027843,
      "balance_loss_clip": 1.04702818,
      "balance_loss_mlp": 1.0190568,
      "epoch": 0.29110803823723924,
      "flos": 15304194817920.0,
      "grad_norm": 2.3753526577024453,
      "language_loss": 0.65266418,
      "learning_rate": 3.3266040898601877e-06,
      "loss": 0.67454827,
      "num_input_tokens_seen": 51875880,
      "step": 2421,
      "time_per_iteration": 2.6396050453186035
    },
    {
      "auxiliary_loss_clip": 0.01134845,
      "auxiliary_loss_mlp": 0.01029732,
      "balance_loss_clip": 1.04598784,
      "balance_loss_mlp": 1.02049351,
      "epoch": 0.2912282811278783,
      "flos": 22595352923520.0,
      "grad_norm": 1.9131875916078578,
      "language_loss": 0.78210151,
      "learning_rate": 3.3260210444282045e-06,
      "loss": 0.80374724,
      "num_input_tokens_seen": 51893835,
      "step": 2422,
      "time_per_iteration": 2.789106607437134
    },
    {
      "auxiliary_loss_clip": 0.01185207,
      "auxiliary_loss_mlp": 0.01032847,
      "balance_loss_clip": 1.05853915,
      "balance_loss_mlp": 1.02415085,
      "epoch": 0.2913485240185174,
      "flos": 24497900599680.0,
      "grad_norm": 2.930877205862439,
      "language_loss": 0.73549193,
      "learning_rate": 3.325437797839883e-06,
      "loss": 0.75767249,
      "num_input_tokens_seen": 51912205,
      "step": 2423,
      "time_per_iteration": 2.670621871948242
    },
    {
      "auxiliary_loss_clip": 0.01202732,
      "auxiliary_loss_mlp": 0.01040298,
      "balance_loss_clip": 1.05671811,
      "balance_loss_mlp": 1.03067207,
      "epoch": 0.2914687669091565,
      "flos": 17931024334080.0,
      "grad_norm": 2.5990482625669706,
      "language_loss": 0.75094366,
      "learning_rate": 3.3248543501837015e-06,
      "loss": 0.77337396,
      "num_input_tokens_seen": 51929410,
      "step": 2424,
      "time_per_iteration": 2.551671266555786
    },
    {
      "auxiliary_loss_clip": 0.01147915,
      "auxiliary_loss_mlp": 0.01034647,
      "balance_loss_clip": 1.05177855,
      "balance_loss_mlp": 1.02539682,
      "epoch": 0.2915890097997956,
      "flos": 22529313768960.0,
      "grad_norm": 2.2205908031764645,
      "language_loss": 0.77473003,
      "learning_rate": 3.3242707015481684e-06,
      "loss": 0.79655564,
      "num_input_tokens_seen": 51949345,
      "step": 2425,
      "time_per_iteration": 2.8151302337646484
    },
    {
      "auxiliary_loss_clip": 0.01168402,
      "auxiliary_loss_mlp": 0.01022389,
      "balance_loss_clip": 1.04795551,
      "balance_loss_mlp": 1.01378155,
      "epoch": 0.2917092526904347,
      "flos": 13845216193920.0,
      "grad_norm": 2.11826436669137,
      "language_loss": 0.80743587,
      "learning_rate": 3.323686852021823e-06,
      "loss": 0.8293438,
      "num_input_tokens_seen": 51966855,
      "step": 2426,
      "time_per_iteration": 2.6223642826080322
    },
    {
      "auxiliary_loss_clip": 0.01163239,
      "auxiliary_loss_mlp": 0.0103193,
      "balance_loss_clip": 1.04848599,
      "balance_loss_mlp": 1.02261376,
      "epoch": 0.2918294955810738,
      "flos": 22674859678080.0,
      "grad_norm": 6.340243945991299,
      "language_loss": 0.79522443,
      "learning_rate": 3.323102801693235e-06,
      "loss": 0.8171761,
      "num_input_tokens_seen": 51985620,
      "step": 2427,
      "time_per_iteration": 2.7346606254577637
    },
    {
      "auxiliary_loss_clip": 0.01181787,
      "auxiliary_loss_mlp": 0.01028046,
      "balance_loss_clip": 1.05289674,
      "balance_loss_mlp": 1.01906347,
      "epoch": 0.29194973847171285,
      "flos": 23438284364160.0,
      "grad_norm": 2.3304855886962836,
      "language_loss": 0.80914336,
      "learning_rate": 3.322518550651003e-06,
      "loss": 0.83124173,
      "num_input_tokens_seen": 52004930,
      "step": 2428,
      "time_per_iteration": 2.64074969291687
    },
    {
      "auxiliary_loss_clip": 0.01179844,
      "auxiliary_loss_mlp": 0.01029776,
      "balance_loss_clip": 1.05395555,
      "balance_loss_mlp": 1.02076399,
      "epoch": 0.29206998136235196,
      "flos": 21909064694400.0,
      "grad_norm": 1.741034378808215,
      "language_loss": 0.81406438,
      "learning_rate": 3.3219340989837586e-06,
      "loss": 0.83616066,
      "num_input_tokens_seen": 52024920,
      "step": 2429,
      "time_per_iteration": 2.6724853515625
    },
    {
      "auxiliary_loss_clip": 0.01170953,
      "auxiliary_loss_mlp": 0.01025551,
      "balance_loss_clip": 1.0533812,
      "balance_loss_mlp": 1.01739693,
      "epoch": 0.292190224252991,
      "flos": 23215925220480.0,
      "grad_norm": 1.8928266013504385,
      "language_loss": 0.80639541,
      "learning_rate": 3.3213494467801625e-06,
      "loss": 0.82836044,
      "num_input_tokens_seen": 52044095,
      "step": 2430,
      "time_per_iteration": 3.559154510498047
    },
    {
      "auxiliary_loss_clip": 0.01113348,
      "auxiliary_loss_mlp": 0.01026899,
      "balance_loss_clip": 1.04432905,
      "balance_loss_mlp": 1.01770246,
      "epoch": 0.2923104671436301,
      "flos": 20740818752640.0,
      "grad_norm": 2.1942537758196834,
      "language_loss": 0.71587175,
      "learning_rate": 3.3207645941289063e-06,
      "loss": 0.73727417,
      "num_input_tokens_seen": 52062440,
      "step": 2431,
      "time_per_iteration": 2.7955310344696045
    },
    {
      "auxiliary_loss_clip": 0.01188298,
      "auxiliary_loss_mlp": 0.00763447,
      "balance_loss_clip": 1.05808103,
      "balance_loss_mlp": 1.00031328,
      "epoch": 0.29243071003426924,
      "flos": 35809114999680.0,
      "grad_norm": 1.7560568733729187,
      "language_loss": 0.79991996,
      "learning_rate": 3.320179541118711e-06,
      "loss": 0.81943738,
      "num_input_tokens_seen": 52084940,
      "step": 2432,
      "time_per_iteration": 2.7660868167877197
    },
    {
      "auxiliary_loss_clip": 0.01091084,
      "auxiliary_loss_mlp": 0.01005642,
      "balance_loss_clip": 1.02035546,
      "balance_loss_mlp": 1.00383055,
      "epoch": 0.2925509529249083,
      "flos": 58081598524800.0,
      "grad_norm": 1.0015028548756435,
      "language_loss": 0.6032207,
      "learning_rate": 3.3195942878383293e-06,
      "loss": 0.62418807,
      "num_input_tokens_seen": 52141040,
      "step": 2433,
      "time_per_iteration": 4.134268045425415
    },
    {
      "auxiliary_loss_clip": 0.01185882,
      "auxiliary_loss_mlp": 0.01028517,
      "balance_loss_clip": 1.05534208,
      "balance_loss_mlp": 1.0182656,
      "epoch": 0.2926711958155474,
      "flos": 21397122103680.0,
      "grad_norm": 1.9107247588445682,
      "language_loss": 0.78021657,
      "learning_rate": 3.319008834376543e-06,
      "loss": 0.80236053,
      "num_input_tokens_seen": 52160730,
      "step": 2434,
      "time_per_iteration": 2.7216079235076904
    },
    {
      "auxiliary_loss_clip": 0.01162321,
      "auxiliary_loss_mlp": 0.01031888,
      "balance_loss_clip": 1.04842567,
      "balance_loss_mlp": 1.02218401,
      "epoch": 0.2927914387061865,
      "flos": 23185796688000.0,
      "grad_norm": 2.459901177462285,
      "language_loss": 0.88869387,
      "learning_rate": 3.3184231808221654e-06,
      "loss": 0.91063595,
      "num_input_tokens_seen": 52175055,
      "step": 2435,
      "time_per_iteration": 2.760983943939209
    },
    {
      "auxiliary_loss_clip": 0.01164509,
      "auxiliary_loss_mlp": 0.01030067,
      "balance_loss_clip": 1.05495834,
      "balance_loss_mlp": 1.02156782,
      "epoch": 0.29291168159682557,
      "flos": 22455553190400.0,
      "grad_norm": 2.74438637460682,
      "language_loss": 0.63145071,
      "learning_rate": 3.3178373272640394e-06,
      "loss": 0.65339649,
      "num_input_tokens_seen": 52194150,
      "step": 2436,
      "time_per_iteration": 2.812869071960449
    },
    {
      "auxiliary_loss_clip": 0.0119863,
      "auxiliary_loss_mlp": 0.01032737,
      "balance_loss_clip": 1.05552232,
      "balance_loss_mlp": 1.02412438,
      "epoch": 0.2930319244874647,
      "flos": 21170632896000.0,
      "grad_norm": 2.8328547184374244,
      "language_loss": 0.85114354,
      "learning_rate": 3.3172512737910387e-06,
      "loss": 0.87345725,
      "num_input_tokens_seen": 52211660,
      "step": 2437,
      "time_per_iteration": 3.573444128036499
    },
    {
      "auxiliary_loss_clip": 0.01186002,
      "auxiliary_loss_mlp": 0.01032698,
      "balance_loss_clip": 1.0532136,
      "balance_loss_mlp": 1.0240314,
      "epoch": 0.2931521673781038,
      "flos": 31357843931520.0,
      "grad_norm": 2.6416590212450197,
      "language_loss": 0.89017123,
      "learning_rate": 3.3166650204920674e-06,
      "loss": 0.91235828,
      "num_input_tokens_seen": 52232830,
      "step": 2438,
      "time_per_iteration": 3.6097118854522705
    },
    {
      "auxiliary_loss_clip": 0.01188376,
      "auxiliary_loss_mlp": 0.0103266,
      "balance_loss_clip": 1.05830407,
      "balance_loss_mlp": 1.02388072,
      "epoch": 0.29327241026874284,
      "flos": 24200990778240.0,
      "grad_norm": 1.8432197583996215,
      "language_loss": 0.81750166,
      "learning_rate": 3.316078567456059e-06,
      "loss": 0.83971202,
      "num_input_tokens_seen": 52250670,
      "step": 2439,
      "time_per_iteration": 2.6202855110168457
    },
    {
      "auxiliary_loss_clip": 0.01130926,
      "auxiliary_loss_mlp": 0.01025513,
      "balance_loss_clip": 1.04725409,
      "balance_loss_mlp": 1.01648855,
      "epoch": 0.29339265315938196,
      "flos": 24242611662720.0,
      "grad_norm": 3.1314712731901775,
      "language_loss": 0.75919116,
      "learning_rate": 3.3154919147719786e-06,
      "loss": 0.78075552,
      "num_input_tokens_seen": 52271685,
      "step": 2440,
      "time_per_iteration": 2.7564847469329834
    },
    {
      "auxiliary_loss_clip": 0.01186802,
      "auxiliary_loss_mlp": 0.01034739,
      "balance_loss_clip": 1.05441952,
      "balance_loss_mlp": 1.02563119,
      "epoch": 0.29351289605002107,
      "flos": 16946641134720.0,
      "grad_norm": 3.0156303243899405,
      "language_loss": 0.86625469,
      "learning_rate": 3.31490506252882e-06,
      "loss": 0.88847005,
      "num_input_tokens_seen": 52291065,
      "step": 2441,
      "time_per_iteration": 2.590186595916748
    },
    {
      "auxiliary_loss_clip": 0.01148656,
      "auxiliary_loss_mlp": 0.01032887,
      "balance_loss_clip": 1.04763675,
      "balance_loss_mlp": 1.02416658,
      "epoch": 0.2936331389406601,
      "flos": 19829082810240.0,
      "grad_norm": 1.8469770517470634,
      "language_loss": 0.8423996,
      "learning_rate": 3.31431801081561e-06,
      "loss": 0.86421502,
      "num_input_tokens_seen": 52310000,
      "step": 2442,
      "time_per_iteration": 2.7188212871551514
    },
    {
      "auxiliary_loss_clip": 0.01070704,
      "auxiliary_loss_mlp": 0.01003858,
      "balance_loss_clip": 1.01710725,
      "balance_loss_mlp": 1.00189137,
      "epoch": 0.29375338183129923,
      "flos": 71416844398080.0,
      "grad_norm": 0.8884605046579317,
      "language_loss": 0.67882746,
      "learning_rate": 3.313730759721402e-06,
      "loss": 0.6995731,
      "num_input_tokens_seen": 52372930,
      "step": 2443,
      "time_per_iteration": 3.3173913955688477
    },
    {
      "auxiliary_loss_clip": 0.01170234,
      "auxiliary_loss_mlp": 0.01030714,
      "balance_loss_clip": 1.05556679,
      "balance_loss_mlp": 1.02188039,
      "epoch": 0.29387362472193834,
      "flos": 22054502862720.0,
      "grad_norm": 2.0510558615251107,
      "language_loss": 0.87016106,
      "learning_rate": 3.313143309335282e-06,
      "loss": 0.89217055,
      "num_input_tokens_seen": 52391420,
      "step": 2444,
      "time_per_iteration": 2.685771942138672
    },
    {
      "auxiliary_loss_clip": 0.01157304,
      "auxiliary_loss_mlp": 0.01030519,
      "balance_loss_clip": 1.0530349,
      "balance_loss_mlp": 1.02213895,
      "epoch": 0.2939938676125774,
      "flos": 22966418373120.0,
      "grad_norm": 2.02889833744704,
      "language_loss": 0.85080636,
      "learning_rate": 3.3125556597463665e-06,
      "loss": 0.8726846,
      "num_input_tokens_seen": 52410725,
      "step": 2445,
      "time_per_iteration": 2.7091870307922363
    },
    {
      "auxiliary_loss_clip": 0.01186189,
      "auxiliary_loss_mlp": 0.01028195,
      "balance_loss_clip": 1.05737948,
      "balance_loss_mlp": 1.02026749,
      "epoch": 0.2941141105032165,
      "flos": 31358705857920.0,
      "grad_norm": 1.5668387550079224,
      "language_loss": 0.66323811,
      "learning_rate": 3.311967811043801e-06,
      "loss": 0.68538195,
      "num_input_tokens_seen": 52432645,
      "step": 2446,
      "time_per_iteration": 2.723324775695801
    },
    {
      "auxiliary_loss_clip": 0.01184052,
      "auxiliary_loss_mlp": 0.01028851,
      "balance_loss_clip": 1.05617642,
      "balance_loss_mlp": 1.02022004,
      "epoch": 0.29423435339385556,
      "flos": 23222138273280.0,
      "grad_norm": 2.8935287764486453,
      "language_loss": 0.82250184,
      "learning_rate": 3.3113797633167617e-06,
      "loss": 0.84463084,
      "num_input_tokens_seen": 52450940,
      "step": 2447,
      "time_per_iteration": 2.6904137134552
    },
    {
      "auxiliary_loss_clip": 0.01201157,
      "auxiliary_loss_mlp": 0.01027667,
      "balance_loss_clip": 1.05780005,
      "balance_loss_mlp": 1.01875067,
      "epoch": 0.2943545962844947,
      "flos": 26864054138880.0,
      "grad_norm": 2.418451203267424,
      "language_loss": 0.69242311,
      "learning_rate": 3.310791516654455e-06,
      "loss": 0.71471143,
      "num_input_tokens_seen": 52468000,
      "step": 2448,
      "time_per_iteration": 2.6190404891967773
    },
    {
      "auxiliary_loss_clip": 0.01163752,
      "auxiliary_loss_mlp": 0.0102917,
      "balance_loss_clip": 1.05236208,
      "balance_loss_mlp": 1.01939464,
      "epoch": 0.2944748391751338,
      "flos": 20231677422720.0,
      "grad_norm": 2.625249568010857,
      "language_loss": 0.79737365,
      "learning_rate": 3.3102030711461177e-06,
      "loss": 0.81930292,
      "num_input_tokens_seen": 52487575,
      "step": 2449,
      "time_per_iteration": 2.7124245166778564
    },
    {
      "auxiliary_loss_clip": 0.01158231,
      "auxiliary_loss_mlp": 0.01032777,
      "balance_loss_clip": 1.05126524,
      "balance_loss_mlp": 1.02368128,
      "epoch": 0.29459508206577284,
      "flos": 15960965045760.0,
      "grad_norm": 2.883461741914008,
      "language_loss": 0.68151021,
      "learning_rate": 3.3096144268810156e-06,
      "loss": 0.70342028,
      "num_input_tokens_seen": 52506335,
      "step": 2450,
      "time_per_iteration": 2.638737678527832
    },
    {
      "auxiliary_loss_clip": 0.01177764,
      "auxiliary_loss_mlp": 0.01029256,
      "balance_loss_clip": 1.05318737,
      "balance_loss_mlp": 1.01958179,
      "epoch": 0.29471532495641195,
      "flos": 20412882558720.0,
      "grad_norm": 1.9634107849911926,
      "language_loss": 0.73060155,
      "learning_rate": 3.3090255839484462e-06,
      "loss": 0.75267172,
      "num_input_tokens_seen": 52524330,
      "step": 2451,
      "time_per_iteration": 2.6517651081085205
    },
    {
      "auxiliary_loss_clip": 0.01172124,
      "auxiliary_loss_mlp": 0.01031472,
      "balance_loss_clip": 1.05168271,
      "balance_loss_mlp": 1.02216804,
      "epoch": 0.29483556784705106,
      "flos": 20376576887040.0,
      "grad_norm": 4.193378356667574,
      "language_loss": 0.85514486,
      "learning_rate": 3.3084365424377366e-06,
      "loss": 0.87718081,
      "num_input_tokens_seen": 52543095,
      "step": 2452,
      "time_per_iteration": 2.635852336883545
    },
    {
      "auxiliary_loss_clip": 0.01066202,
      "auxiliary_loss_mlp": 0.01009692,
      "balance_loss_clip": 1.02826595,
      "balance_loss_mlp": 1.00741529,
      "epoch": 0.2949558107376901,
      "flos": 68555660595840.0,
      "grad_norm": 0.7267308794142744,
      "language_loss": 0.55947506,
      "learning_rate": 3.307847302438245e-06,
      "loss": 0.58023405,
      "num_input_tokens_seen": 52597075,
      "step": 2453,
      "time_per_iteration": 3.2400941848754883
    },
    {
      "auxiliary_loss_clip": 0.01126889,
      "auxiliary_loss_mlp": 0.01033754,
      "balance_loss_clip": 1.0438627,
      "balance_loss_mlp": 1.02469444,
      "epoch": 0.2950760536283292,
      "flos": 16107085572480.0,
      "grad_norm": 2.574712245679792,
      "language_loss": 0.77621293,
      "learning_rate": 3.3072578640393562e-06,
      "loss": 0.79781938,
      "num_input_tokens_seen": 52614410,
      "step": 2454,
      "time_per_iteration": 2.7028069496154785
    },
    {
      "auxiliary_loss_clip": 0.01170393,
      "auxiliary_loss_mlp": 0.01032872,
      "balance_loss_clip": 1.05305529,
      "balance_loss_mlp": 1.02402687,
      "epoch": 0.29519629651896834,
      "flos": 20483626394880.0,
      "grad_norm": 2.0731907783948995,
      "language_loss": 0.79884654,
      "learning_rate": 3.3066682273304886e-06,
      "loss": 0.82087922,
      "num_input_tokens_seen": 52632055,
      "step": 2455,
      "time_per_iteration": 2.719906806945801
    },
    {
      "auxiliary_loss_clip": 0.01191615,
      "auxiliary_loss_mlp": 0.00763997,
      "balance_loss_clip": 1.05600107,
      "balance_loss_mlp": 1.00040317,
      "epoch": 0.2953165394096074,
      "flos": 18916484941440.0,
      "grad_norm": 2.0477286566281006,
      "language_loss": 0.78717607,
      "learning_rate": 3.3060783924010904e-06,
      "loss": 0.80673218,
      "num_input_tokens_seen": 52649980,
      "step": 2456,
      "time_per_iteration": 3.5529534816741943
    },
    {
      "auxiliary_loss_clip": 0.01157635,
      "auxiliary_loss_mlp": 0.01030897,
      "balance_loss_clip": 1.05199957,
      "balance_loss_mlp": 1.02137256,
      "epoch": 0.2954367823002465,
      "flos": 20624467622400.0,
      "grad_norm": 2.3640834019290162,
      "language_loss": 0.8504703,
      "learning_rate": 3.3054883593406387e-06,
      "loss": 0.87235558,
      "num_input_tokens_seen": 52664730,
      "step": 2457,
      "time_per_iteration": 2.7160181999206543
    },
    {
      "auxiliary_loss_clip": 0.01172088,
      "auxiliary_loss_mlp": 0.01034907,
      "balance_loss_clip": 1.05190444,
      "balance_loss_mlp": 1.02669358,
      "epoch": 0.2955570251908856,
      "flos": 31175525473920.0,
      "grad_norm": 2.132650035963358,
      "language_loss": 0.65337348,
      "learning_rate": 3.3048981282386404e-06,
      "loss": 0.67544341,
      "num_input_tokens_seen": 52686040,
      "step": 2458,
      "time_per_iteration": 2.7735068798065186
    },
    {
      "auxiliary_loss_clip": 0.01146172,
      "auxiliary_loss_mlp": 0.01031449,
      "balance_loss_clip": 1.05156922,
      "balance_loss_mlp": 1.02157879,
      "epoch": 0.29567726808152467,
      "flos": 21650328051840.0,
      "grad_norm": 1.9458781015591813,
      "language_loss": 0.82235259,
      "learning_rate": 3.304307699184634e-06,
      "loss": 0.84412879,
      "num_input_tokens_seen": 52704630,
      "step": 2459,
      "time_per_iteration": 3.623737096786499
    },
    {
      "auxiliary_loss_clip": 0.01175797,
      "auxiliary_loss_mlp": 0.01033505,
      "balance_loss_clip": 1.05615973,
      "balance_loss_mlp": 1.02485681,
      "epoch": 0.2957975109721638,
      "flos": 24243868638720.0,
      "grad_norm": 1.674108417774582,
      "language_loss": 0.78746378,
      "learning_rate": 3.3037170722681866e-06,
      "loss": 0.80955684,
      "num_input_tokens_seen": 52725465,
      "step": 2460,
      "time_per_iteration": 2.7167670726776123
    },
    {
      "auxiliary_loss_clip": 0.01150115,
      "auxiliary_loss_mlp": 0.01029564,
      "balance_loss_clip": 1.05109453,
      "balance_loss_mlp": 1.02042103,
      "epoch": 0.29591775386280283,
      "flos": 13479717352320.0,
      "grad_norm": 1.8576289432757902,
      "language_loss": 0.68161768,
      "learning_rate": 3.3031262475788956e-06,
      "loss": 0.70341444,
      "num_input_tokens_seen": 52742405,
      "step": 2461,
      "time_per_iteration": 2.7279629707336426
    },
    {
      "auxiliary_loss_clip": 0.01169856,
      "auxiliary_loss_mlp": 0.01027582,
      "balance_loss_clip": 1.05427384,
      "balance_loss_mlp": 1.0187906,
      "epoch": 0.29603799675344195,
      "flos": 17749783284480.0,
      "grad_norm": 2.1702633124896544,
      "language_loss": 0.73218375,
      "learning_rate": 3.3025352252063897e-06,
      "loss": 0.75415814,
      "num_input_tokens_seen": 52761100,
      "step": 2462,
      "time_per_iteration": 2.664778232574463
    },
    {
      "auxiliary_loss_clip": 0.01187036,
      "auxiliary_loss_mlp": 0.01031723,
      "balance_loss_clip": 1.05966806,
      "balance_loss_mlp": 1.02231109,
      "epoch": 0.29615823964408106,
      "flos": 22783920347520.0,
      "grad_norm": 1.7724278875271713,
      "language_loss": 0.75051969,
      "learning_rate": 3.3019440052403252e-06,
      "loss": 0.77270728,
      "num_input_tokens_seen": 52780965,
      "step": 2463,
      "time_per_iteration": 4.514355897903442
    },
    {
      "auxiliary_loss_clip": 0.01174261,
      "auxiliary_loss_mlp": 0.01031111,
      "balance_loss_clip": 1.05478764,
      "balance_loss_mlp": 1.02190197,
      "epoch": 0.2962784825347201,
      "flos": 23514199758720.0,
      "grad_norm": 2.2187383389293567,
      "language_loss": 0.70877004,
      "learning_rate": 3.30135258777039e-06,
      "loss": 0.73082376,
      "num_input_tokens_seen": 52800335,
      "step": 2464,
      "time_per_iteration": 2.7360997200012207
    },
    {
      "auxiliary_loss_clip": 0.01191077,
      "auxiliary_loss_mlp": 0.00764373,
      "balance_loss_clip": 1.05483794,
      "balance_loss_mlp": 1.00043404,
      "epoch": 0.2963987254253592,
      "flos": 16362769559040.0,
      "grad_norm": 2.304217941397056,
      "language_loss": 0.70460379,
      "learning_rate": 3.3007609728863024e-06,
      "loss": 0.72415829,
      "num_input_tokens_seen": 52818425,
      "step": 2465,
      "time_per_iteration": 2.5990383625030518
    },
    {
      "auxiliary_loss_clip": 0.01123274,
      "auxiliary_loss_mlp": 0.01027825,
      "balance_loss_clip": 1.05217564,
      "balance_loss_mlp": 1.0196588,
      "epoch": 0.29651896831599833,
      "flos": 33472263980160.0,
      "grad_norm": 2.126704081789709,
      "language_loss": 0.72858155,
      "learning_rate": 3.300169160677809e-06,
      "loss": 0.75009251,
      "num_input_tokens_seen": 52842340,
      "step": 2466,
      "time_per_iteration": 2.904372453689575
    },
    {
      "auxiliary_loss_clip": 0.0116655,
      "auxiliary_loss_mlp": 0.01032748,
      "balance_loss_clip": 1.05389416,
      "balance_loss_mlp": 1.02320552,
      "epoch": 0.2966392112066374,
      "flos": 23805363404160.0,
      "grad_norm": 2.5885375924364507,
      "language_loss": 0.77659857,
      "learning_rate": 3.2995771512346878e-06,
      "loss": 0.79859161,
      "num_input_tokens_seen": 52860690,
      "step": 2467,
      "time_per_iteration": 2.687779426574707
    },
    {
      "auxiliary_loss_clip": 0.01208041,
      "auxiliary_loss_mlp": 0.00764918,
      "balance_loss_clip": 1.06073987,
      "balance_loss_mlp": 1.0004915,
      "epoch": 0.2967594540972765,
      "flos": 19938466702080.0,
      "grad_norm": 1.9764046160098536,
      "language_loss": 0.73144019,
      "learning_rate": 3.298984944646746e-06,
      "loss": 0.7511698,
      "num_input_tokens_seen": 52879370,
      "step": 2468,
      "time_per_iteration": 2.6537508964538574
    },
    {
      "auxiliary_loss_clip": 0.01192347,
      "auxiliary_loss_mlp": 0.00764528,
      "balance_loss_clip": 1.05869532,
      "balance_loss_mlp": 1.0005095,
      "epoch": 0.2968796969879156,
      "flos": 23732823888000.0,
      "grad_norm": 2.6906080870447906,
      "language_loss": 0.82018733,
      "learning_rate": 3.298392541003822e-06,
      "loss": 0.83975607,
      "num_input_tokens_seen": 52898775,
      "step": 2469,
      "time_per_iteration": 2.6473257541656494
    },
    {
      "auxiliary_loss_clip": 0.0117555,
      "auxiliary_loss_mlp": 0.01020398,
      "balance_loss_clip": 1.05707383,
      "balance_loss_mlp": 1.01229215,
      "epoch": 0.29699993987855466,
      "flos": 22893699288960.0,
      "grad_norm": 1.740324877208916,
      "language_loss": 0.89414406,
      "learning_rate": 3.2977999403957806e-06,
      "loss": 0.9161036,
      "num_input_tokens_seen": 52917535,
      "step": 2470,
      "time_per_iteration": 2.7564756870269775
    },
    {
      "auxiliary_loss_clip": 0.01207034,
      "auxiliary_loss_mlp": 0.01033166,
      "balance_loss_clip": 1.06130755,
      "balance_loss_mlp": 1.02386749,
      "epoch": 0.2971201827691938,
      "flos": 33832555349760.0,
      "grad_norm": 4.263796001732715,
      "language_loss": 0.67156446,
      "learning_rate": 3.2972071429125207e-06,
      "loss": 0.69396651,
      "num_input_tokens_seen": 52938755,
      "step": 2471,
      "time_per_iteration": 2.7117512226104736
    },
    {
      "auxiliary_loss_clip": 0.01154827,
      "auxiliary_loss_mlp": 0.01035167,
      "balance_loss_clip": 1.05210972,
      "balance_loss_mlp": 1.02593446,
      "epoch": 0.2972404256598329,
      "flos": 22054359208320.0,
      "grad_norm": 2.748850826992151,
      "language_loss": 0.88424659,
      "learning_rate": 3.2966141486439682e-06,
      "loss": 0.90614653,
      "num_input_tokens_seen": 52957945,
      "step": 2472,
      "time_per_iteration": 2.7050881385803223
    },
    {
      "auxiliary_loss_clip": 0.01131424,
      "auxiliary_loss_mlp": 0.01033015,
      "balance_loss_clip": 1.04792929,
      "balance_loss_mlp": 1.02300107,
      "epoch": 0.29736066855047194,
      "flos": 31978595796480.0,
      "grad_norm": 2.1290290323555947,
      "language_loss": 0.65068579,
      "learning_rate": 3.29602095768008e-06,
      "loss": 0.67233014,
      "num_input_tokens_seen": 52978460,
      "step": 2473,
      "time_per_iteration": 2.854241132736206
    },
    {
      "auxiliary_loss_clip": 0.01167947,
      "auxiliary_loss_mlp": 0.01037977,
      "balance_loss_clip": 1.05530834,
      "balance_loss_mlp": 1.02937031,
      "epoch": 0.29748091144111105,
      "flos": 33510401245440.0,
      "grad_norm": 2.078410767951869,
      "language_loss": 0.64167035,
      "learning_rate": 3.2954275701108437e-06,
      "loss": 0.66372961,
      "num_input_tokens_seen": 52999640,
      "step": 2474,
      "time_per_iteration": 2.7921977043151855
    },
    {
      "auxiliary_loss_clip": 0.01140443,
      "auxiliary_loss_mlp": 0.01036199,
      "balance_loss_clip": 1.05080807,
      "balance_loss_mlp": 1.02644181,
      "epoch": 0.29760115433175016,
      "flos": 41283373409280.0,
      "grad_norm": 2.3633618237283227,
      "language_loss": 0.68829334,
      "learning_rate": 3.294833986026275e-06,
      "loss": 0.71005976,
      "num_input_tokens_seen": 53022880,
      "step": 2475,
      "time_per_iteration": 2.965322732925415
    },
    {
      "auxiliary_loss_clip": 0.01154853,
      "auxiliary_loss_mlp": 0.01031186,
      "balance_loss_clip": 1.0541501,
      "balance_loss_mlp": 1.02172065,
      "epoch": 0.2977213972223892,
      "flos": 24493339572480.0,
      "grad_norm": 2.1006562233543282,
      "language_loss": 0.85395849,
      "learning_rate": 3.29424020551642e-06,
      "loss": 0.87581885,
      "num_input_tokens_seen": 53041515,
      "step": 2476,
      "time_per_iteration": 2.7666096687316895
    },
    {
      "auxiliary_loss_clip": 0.0120782,
      "auxiliary_loss_mlp": 0.01036559,
      "balance_loss_clip": 1.05911422,
      "balance_loss_mlp": 1.02594924,
      "epoch": 0.2978416401130283,
      "flos": 21285116519040.0,
      "grad_norm": 2.5788125152917916,
      "language_loss": 0.72052622,
      "learning_rate": 3.2936462286713546e-06,
      "loss": 0.74296999,
      "num_input_tokens_seen": 53059865,
      "step": 2477,
      "time_per_iteration": 2.707519769668579
    },
    {
      "auxiliary_loss_clip": 0.01194282,
      "auxiliary_loss_mlp": 0.01032991,
      "balance_loss_clip": 1.05930793,
      "balance_loss_mlp": 1.02353144,
      "epoch": 0.2979618830036674,
      "flos": 25772154554880.0,
      "grad_norm": 1.8856476438125973,
      "language_loss": 0.7715075,
      "learning_rate": 3.2930520555811846e-06,
      "loss": 0.79378021,
      "num_input_tokens_seen": 53079490,
      "step": 2478,
      "time_per_iteration": 2.653155565261841
    },
    {
      "auxiliary_loss_clip": 0.01094414,
      "auxiliary_loss_mlp": 0.00765358,
      "balance_loss_clip": 1.04681492,
      "balance_loss_mlp": 1.00039649,
      "epoch": 0.2980821258943065,
      "flos": 23476996247040.0,
      "grad_norm": 1.8163278360308488,
      "language_loss": 0.79809916,
      "learning_rate": 3.292457686336046e-06,
      "loss": 0.81669688,
      "num_input_tokens_seen": 53098810,
      "step": 2479,
      "time_per_iteration": 3.0359456539154053
    },
    {
      "auxiliary_loss_clip": 0.01084514,
      "auxiliary_loss_mlp": 0.01001606,
      "balance_loss_clip": 1.02242446,
      "balance_loss_mlp": 0.99990124,
      "epoch": 0.2982023687849456,
      "flos": 69752314195200.0,
      "grad_norm": 0.8571747934078684,
      "language_loss": 0.61238003,
      "learning_rate": 3.291863121026105e-06,
      "loss": 0.6332413,
      "num_input_tokens_seen": 53162590,
      "step": 2480,
      "time_per_iteration": 3.634397029876709
    },
    {
      "auxiliary_loss_clip": 0.01191348,
      "auxiliary_loss_mlp": 0.01031083,
      "balance_loss_clip": 1.05806947,
      "balance_loss_mlp": 1.02173078,
      "epoch": 0.29832261167558466,
      "flos": 29825930741760.0,
      "grad_norm": 2.2938195479561903,
      "language_loss": 0.76813704,
      "learning_rate": 3.2912683597415547e-06,
      "loss": 0.79036134,
      "num_input_tokens_seen": 53186675,
      "step": 2481,
      "time_per_iteration": 2.7483150959014893
    },
    {
      "auxiliary_loss_clip": 0.0116298,
      "auxiliary_loss_mlp": 0.01030249,
      "balance_loss_clip": 1.05538726,
      "balance_loss_mlp": 1.02091515,
      "epoch": 0.29844285456622377,
      "flos": 33910158683520.0,
      "grad_norm": 3.692901810665582,
      "language_loss": 0.78226292,
      "learning_rate": 3.2906734025726213e-06,
      "loss": 0.80419517,
      "num_input_tokens_seen": 53205940,
      "step": 2482,
      "time_per_iteration": 3.8432607650756836
    },
    {
      "auxiliary_loss_clip": 0.01200753,
      "auxiliary_loss_mlp": 0.01039803,
      "balance_loss_clip": 1.0614028,
      "balance_loss_mlp": 1.03062391,
      "epoch": 0.2985630974568629,
      "flos": 23876933253120.0,
      "grad_norm": 2.5906711410169554,
      "language_loss": 0.88180351,
      "learning_rate": 3.290078249609559e-06,
      "loss": 0.90420902,
      "num_input_tokens_seen": 53225360,
      "step": 2483,
      "time_per_iteration": 2.753770351409912
    },
    {
      "auxiliary_loss_clip": 0.01191353,
      "auxiliary_loss_mlp": 0.01034447,
      "balance_loss_clip": 1.061728,
      "balance_loss_mlp": 1.02514887,
      "epoch": 0.29868334034750194,
      "flos": 21799106184960.0,
      "grad_norm": 2.0282227161839375,
      "language_loss": 0.87848169,
      "learning_rate": 3.2894829009426514e-06,
      "loss": 0.90073967,
      "num_input_tokens_seen": 53243195,
      "step": 2484,
      "time_per_iteration": 2.683485984802246
    },
    {
      "auxiliary_loss_clip": 0.01190725,
      "auxiliary_loss_mlp": 0.01036006,
      "balance_loss_clip": 1.05981696,
      "balance_loss_mlp": 1.02706575,
      "epoch": 0.29880358323814105,
      "flos": 25666649331840.0,
      "grad_norm": 1.9836738464607877,
      "language_loss": 0.77715284,
      "learning_rate": 3.288887356662213e-06,
      "loss": 0.79942018,
      "num_input_tokens_seen": 53264530,
      "step": 2485,
      "time_per_iteration": 3.6818442344665527
    },
    {
      "auxiliary_loss_clip": 0.01086721,
      "auxiliary_loss_mlp": 0.0100174,
      "balance_loss_clip": 1.02255845,
      "balance_loss_mlp": 1.00002337,
      "epoch": 0.29892382612878016,
      "flos": 71005846003200.0,
      "grad_norm": 0.7890467341990293,
      "language_loss": 0.59736454,
      "learning_rate": 3.288291616858588e-06,
      "loss": 0.61824918,
      "num_input_tokens_seen": 53319920,
      "step": 2486,
      "time_per_iteration": 3.0973880290985107
    },
    {
      "auxiliary_loss_clip": 0.01140716,
      "auxiliary_loss_mlp": 0.01036075,
      "balance_loss_clip": 1.05508351,
      "balance_loss_mlp": 1.02626419,
      "epoch": 0.2990440690194192,
      "flos": 25481134563840.0,
      "grad_norm": 2.5765667930303087,
      "language_loss": 0.77001148,
      "learning_rate": 3.287695681622149e-06,
      "loss": 0.7917794,
      "num_input_tokens_seen": 53339270,
      "step": 2487,
      "time_per_iteration": 2.8812294006347656
    },
    {
      "auxiliary_loss_clip": 0.01180501,
      "auxiliary_loss_mlp": 0.0102833,
      "balance_loss_clip": 1.05634618,
      "balance_loss_mlp": 1.01946688,
      "epoch": 0.2991643119100583,
      "flos": 23732357011200.0,
      "grad_norm": 2.1717729596790516,
      "language_loss": 0.81096554,
      "learning_rate": 3.2870995510432982e-06,
      "loss": 0.83305383,
      "num_input_tokens_seen": 53357750,
      "step": 2488,
      "time_per_iteration": 2.7497847080230713
    },
    {
      "auxiliary_loss_clip": 0.0118655,
      "auxiliary_loss_mlp": 0.01026703,
      "balance_loss_clip": 1.05871451,
      "balance_loss_mlp": 1.01830459,
      "epoch": 0.29928455480069743,
      "flos": 27417545786880.0,
      "grad_norm": 2.255519732523873,
      "language_loss": 0.7736901,
      "learning_rate": 3.2865032252124697e-06,
      "loss": 0.79582256,
      "num_input_tokens_seen": 53378265,
      "step": 2489,
      "time_per_iteration": 3.827218532562256
    },
    {
      "auxiliary_loss_clip": 0.01172374,
      "auxiliary_loss_mlp": 0.01027463,
      "balance_loss_clip": 1.05274212,
      "balance_loss_mlp": 1.01907063,
      "epoch": 0.2994047976913365,
      "flos": 33692935184640.0,
      "grad_norm": 1.5463425157771602,
      "language_loss": 0.77925354,
      "learning_rate": 3.2859067042201243e-06,
      "loss": 0.80125189,
      "num_input_tokens_seen": 53400305,
      "step": 2490,
      "time_per_iteration": 2.8704824447631836
    },
    {
      "auxiliary_loss_clip": 0.01107774,
      "auxiliary_loss_mlp": 0.01027948,
      "balance_loss_clip": 1.04541278,
      "balance_loss_mlp": 1.01922798,
      "epoch": 0.2995250405819756,
      "flos": 16763963541120.0,
      "grad_norm": 1.8098043887933326,
      "language_loss": 0.7814616,
      "learning_rate": 3.2853099881567544e-06,
      "loss": 0.80281883,
      "num_input_tokens_seen": 53418705,
      "step": 2491,
      "time_per_iteration": 2.7769112586975098
    },
    {
      "auxiliary_loss_clip": 0.01201444,
      "auxiliary_loss_mlp": 0.01026891,
      "balance_loss_clip": 1.05985343,
      "balance_loss_mlp": 1.01867795,
      "epoch": 0.29964528347261465,
      "flos": 22963976248320.0,
      "grad_norm": 1.8063621969857508,
      "language_loss": 0.79249883,
      "learning_rate": 3.284713077112881e-06,
      "loss": 0.8147822,
      "num_input_tokens_seen": 53438135,
      "step": 2492,
      "time_per_iteration": 2.737680673599243
    },
    {
      "auxiliary_loss_clip": 0.01167681,
      "auxiliary_loss_mlp": 0.01036417,
      "balance_loss_clip": 1.05673313,
      "balance_loss_mlp": 1.02584863,
      "epoch": 0.29976552636325376,
      "flos": 16938021870720.0,
      "grad_norm": 3.7142739908279965,
      "language_loss": 0.86570978,
      "learning_rate": 3.284115971179056e-06,
      "loss": 0.8877508,
      "num_input_tokens_seen": 53452165,
      "step": 2493,
      "time_per_iteration": 2.6638522148132324
    },
    {
      "auxiliary_loss_clip": 0.0113572,
      "auxiliary_loss_mlp": 0.01030007,
      "balance_loss_clip": 1.05198574,
      "balance_loss_mlp": 1.0214417,
      "epoch": 0.2998857692538929,
      "flos": 17056455989760.0,
      "grad_norm": 1.703724364981277,
      "language_loss": 0.78776354,
      "learning_rate": 3.283518670445859e-06,
      "loss": 0.80942082,
      "num_input_tokens_seen": 53470075,
      "step": 2494,
      "time_per_iteration": 2.7404215335845947
    },
    {
      "auxiliary_loss_clip": 0.01074703,
      "auxiliary_loss_mlp": 0.00754919,
      "balance_loss_clip": 1.02317786,
      "balance_loss_mlp": 1.00025368,
      "epoch": 0.30000601214453193,
      "flos": 68831528025600.0,
      "grad_norm": 0.6988227591603532,
      "language_loss": 0.54335272,
      "learning_rate": 3.2829211750038995e-06,
      "loss": 0.56164896,
      "num_input_tokens_seen": 53538705,
      "step": 2495,
      "time_per_iteration": 3.3492865562438965
    },
    {
      "auxiliary_loss_clip": 0.01156434,
      "auxiliary_loss_mlp": 0.01028526,
      "balance_loss_clip": 1.05181515,
      "balance_loss_mlp": 1.01996708,
      "epoch": 0.30012625503517104,
      "flos": 17603267708160.0,
      "grad_norm": 1.8786488999244617,
      "language_loss": 0.89301306,
      "learning_rate": 3.2823234849438183e-06,
      "loss": 0.91486263,
      "num_input_tokens_seen": 53556740,
      "step": 2496,
      "time_per_iteration": 2.687960386276245
    },
    {
      "auxiliary_loss_clip": 0.01180836,
      "auxiliary_loss_mlp": 0.01035388,
      "balance_loss_clip": 1.0561837,
      "balance_loss_mlp": 1.02617931,
      "epoch": 0.30024649792581015,
      "flos": 21252581775360.0,
      "grad_norm": 2.4030353844697596,
      "language_loss": 0.75814128,
      "learning_rate": 3.2817256003562836e-06,
      "loss": 0.78030348,
      "num_input_tokens_seen": 53577115,
      "step": 2497,
      "time_per_iteration": 2.750623941421509
    },
    {
      "auxiliary_loss_clip": 0.01135251,
      "auxiliary_loss_mlp": 0.01028595,
      "balance_loss_clip": 1.05164969,
      "balance_loss_mlp": 1.01893258,
      "epoch": 0.3003667408164492,
      "flos": 23003262748800.0,
      "grad_norm": 2.142463032523493,
      "language_loss": 0.66236413,
      "learning_rate": 3.281127521331995e-06,
      "loss": 0.68400264,
      "num_input_tokens_seen": 53598295,
      "step": 2498,
      "time_per_iteration": 2.8403232097625732
    },
    {
      "auxiliary_loss_clip": 0.0110212,
      "auxiliary_loss_mlp": 0.0100579,
      "balance_loss_clip": 1.02249146,
      "balance_loss_mlp": 1.00404918,
      "epoch": 0.3004869837070883,
      "flos": 64232340750720.0,
      "grad_norm": 0.8753208160624256,
      "language_loss": 0.60642016,
      "learning_rate": 3.2805292479616798e-06,
      "loss": 0.62749922,
      "num_input_tokens_seen": 53657160,
      "step": 2499,
      "time_per_iteration": 3.1126906871795654
    },
    {
      "auxiliary_loss_clip": 0.01176343,
      "auxiliary_loss_mlp": 0.01030848,
      "balance_loss_clip": 1.05574179,
      "balance_loss_mlp": 1.02199662,
      "epoch": 0.30060722659772743,
      "flos": 26248653400320.0,
      "grad_norm": 2.213076221387785,
      "language_loss": 0.92408389,
      "learning_rate": 3.2799307803360955e-06,
      "loss": 0.94615579,
      "num_input_tokens_seen": 53673090,
      "step": 2500,
      "time_per_iteration": 2.758390426635742
    },
    {
      "auxiliary_loss_clip": 0.0120174,
      "auxiliary_loss_mlp": 0.01028484,
      "balance_loss_clip": 1.05699325,
      "balance_loss_mlp": 1.02024031,
      "epoch": 0.3007274694883665,
      "flos": 24970879912320.0,
      "grad_norm": 1.500481568395846,
      "language_loss": 0.81473511,
      "learning_rate": 3.27933211854603e-06,
      "loss": 0.83703744,
      "num_input_tokens_seen": 53692145,
      "step": 2501,
      "time_per_iteration": 2.6924455165863037
    },
    {
      "auxiliary_loss_clip": 0.01177146,
      "auxiliary_loss_mlp": 0.01032422,
      "balance_loss_clip": 1.05751657,
      "balance_loss_mlp": 1.02323675,
      "epoch": 0.3008477123790056,
      "flos": 17055845458560.0,
      "grad_norm": 1.7515453063742352,
      "language_loss": 0.87011474,
      "learning_rate": 3.278733262682299e-06,
      "loss": 0.89221042,
      "num_input_tokens_seen": 53710000,
      "step": 2502,
      "time_per_iteration": 2.694704532623291
    },
    {
      "auxiliary_loss_clip": 0.01205529,
      "auxiliary_loss_mlp": 0.01036929,
      "balance_loss_clip": 1.05855966,
      "balance_loss_mlp": 1.02702248,
      "epoch": 0.3009679552696447,
      "flos": 21506398254720.0,
      "grad_norm": 2.0728986678160717,
      "language_loss": 0.82664561,
      "learning_rate": 3.2781342128357484e-06,
      "loss": 0.84907019,
      "num_input_tokens_seen": 53729355,
      "step": 2503,
      "time_per_iteration": 2.624593496322632
    },
    {
      "auxiliary_loss_clip": 0.0115846,
      "auxiliary_loss_mlp": 0.01028952,
      "balance_loss_clip": 1.05176854,
      "balance_loss_mlp": 1.02004671,
      "epoch": 0.30108819816028376,
      "flos": 21134004001920.0,
      "grad_norm": 2.5907989320226097,
      "language_loss": 0.807302,
      "learning_rate": 3.2775349690972547e-06,
      "loss": 0.82917613,
      "num_input_tokens_seen": 53743505,
      "step": 2504,
      "time_per_iteration": 2.7438576221466064
    },
    {
      "auxiliary_loss_clip": 0.01082332,
      "auxiliary_loss_mlp": 0.01002214,
      "balance_loss_clip": 1.01881611,
      "balance_loss_mlp": 1.00053334,
      "epoch": 0.30120844105092287,
      "flos": 71126434938240.0,
      "grad_norm": 0.7678316732222886,
      "language_loss": 0.51830083,
      "learning_rate": 3.276935531557722e-06,
      "loss": 0.5391463,
      "num_input_tokens_seen": 53808725,
      "step": 2505,
      "time_per_iteration": 3.3351778984069824
    },
    {
      "auxiliary_loss_clip": 0.01149407,
      "auxiliary_loss_mlp": 0.01032905,
      "balance_loss_clip": 1.05079353,
      "balance_loss_mlp": 1.02293301,
      "epoch": 0.301328683941562,
      "flos": 20264571302400.0,
      "grad_norm": 5.413042466296938,
      "language_loss": 0.79869533,
      "learning_rate": 3.2763359003080837e-06,
      "loss": 0.82051843,
      "num_input_tokens_seen": 53825680,
      "step": 2506,
      "time_per_iteration": 2.7837071418762207
    },
    {
      "auxiliary_loss_clip": 0.01079266,
      "auxiliary_loss_mlp": 0.0100237,
      "balance_loss_clip": 1.02116084,
      "balance_loss_mlp": 1.00046241,
      "epoch": 0.30144892683220104,
      "flos": 70648212240000.0,
      "grad_norm": 0.8543175300556911,
      "language_loss": 0.62473786,
      "learning_rate": 3.2757360754393047e-06,
      "loss": 0.64555424,
      "num_input_tokens_seen": 53889750,
      "step": 2507,
      "time_per_iteration": 4.191912889480591
    },
    {
      "auxiliary_loss_clip": 0.01186469,
      "auxiliary_loss_mlp": 0.01034054,
      "balance_loss_clip": 1.05433905,
      "balance_loss_mlp": 1.02445805,
      "epoch": 0.30156916972284015,
      "flos": 22820549241600.0,
      "grad_norm": 3.356270739661911,
      "language_loss": 0.64561367,
      "learning_rate": 3.2751360570423767e-06,
      "loss": 0.6678189,
      "num_input_tokens_seen": 53908135,
      "step": 2508,
      "time_per_iteration": 2.7321279048919678
    },
    {
      "auxiliary_loss_clip": 0.01173203,
      "auxiliary_loss_mlp": 0.01025885,
      "balance_loss_clip": 1.0546006,
      "balance_loss_mlp": 1.01722491,
      "epoch": 0.3016894126134792,
      "flos": 29899188529920.0,
      "grad_norm": 2.3887724132560177,
      "language_loss": 0.75779271,
      "learning_rate": 3.2745358452083236e-06,
      "loss": 0.77978355,
      "num_input_tokens_seen": 53931035,
      "step": 2509,
      "time_per_iteration": 2.7862813472747803
    },
    {
      "auxiliary_loss_clip": 0.01185595,
      "auxiliary_loss_mlp": 0.01028649,
      "balance_loss_clip": 1.05607796,
      "balance_loss_mlp": 1.02075732,
      "epoch": 0.3018096555041183,
      "flos": 21546331200000.0,
      "grad_norm": 1.4974425420085826,
      "language_loss": 0.82163697,
      "learning_rate": 3.2739354400281955e-06,
      "loss": 0.84377944,
      "num_input_tokens_seen": 53952255,
      "step": 2510,
      "time_per_iteration": 2.6842591762542725
    },
    {
      "auxiliary_loss_clip": 0.01068214,
      "auxiliary_loss_mlp": 0.00755169,
      "balance_loss_clip": 1.02035618,
      "balance_loss_mlp": 1.00016773,
      "epoch": 0.3019298983947574,
      "flos": 59136294597120.0,
      "grad_norm": 0.8607184144590732,
      "language_loss": 0.63678193,
      "learning_rate": 3.2733348415930744e-06,
      "loss": 0.65501571,
      "num_input_tokens_seen": 54014125,
      "step": 2511,
      "time_per_iteration": 4.225912809371948
    },
    {
      "auxiliary_loss_clip": 0.01155835,
      "auxiliary_loss_mlp": 0.01028997,
      "balance_loss_clip": 1.0549984,
      "balance_loss_mlp": 1.02040803,
      "epoch": 0.3020501412853965,
      "flos": 34423070941440.0,
      "grad_norm": 3.5904280977935503,
      "language_loss": 0.80539465,
      "learning_rate": 3.27273404999407e-06,
      "loss": 0.82724297,
      "num_input_tokens_seen": 54036345,
      "step": 2512,
      "time_per_iteration": 2.8455705642700195
    },
    {
      "auxiliary_loss_clip": 0.01079077,
      "auxiliary_loss_mlp": 0.01003105,
      "balance_loss_clip": 1.01898885,
      "balance_loss_mlp": 1.00128078,
      "epoch": 0.3021703841760356,
      "flos": 71008288128000.0,
      "grad_norm": 0.7968676786434831,
      "language_loss": 0.60457325,
      "learning_rate": 3.272133065322322e-06,
      "loss": 0.62539506,
      "num_input_tokens_seen": 54094615,
      "step": 2513,
      "time_per_iteration": 3.2093701362609863
    },
    {
      "auxiliary_loss_clip": 0.01199369,
      "auxiliary_loss_mlp": 0.01033631,
      "balance_loss_clip": 1.05607796,
      "balance_loss_mlp": 1.02530408,
      "epoch": 0.3022906270666747,
      "flos": 21510528318720.0,
      "grad_norm": 1.900517210764832,
      "language_loss": 0.79629225,
      "learning_rate": 3.271531887669e-06,
      "loss": 0.81862223,
      "num_input_tokens_seen": 54114675,
      "step": 2514,
      "time_per_iteration": 2.6656150817871094
    },
    {
      "auxiliary_loss_clip": 0.01149241,
      "auxiliary_loss_mlp": 0.01033164,
      "balance_loss_clip": 1.05070782,
      "balance_loss_mlp": 1.02387774,
      "epoch": 0.30241086995731375,
      "flos": 31132001168640.0,
      "grad_norm": 2.2723411582577677,
      "language_loss": 0.63235587,
      "learning_rate": 3.2709305171253015e-06,
      "loss": 0.65417993,
      "num_input_tokens_seen": 54134795,
      "step": 2515,
      "time_per_iteration": 3.8105626106262207
    },
    {
      "auxiliary_loss_clip": 0.01188737,
      "auxiliary_loss_mlp": 0.01037594,
      "balance_loss_clip": 1.05778158,
      "balance_loss_mlp": 1.02888548,
      "epoch": 0.30253111284795287,
      "flos": 23511542152320.0,
      "grad_norm": 2.174041968346941,
      "language_loss": 0.78224921,
      "learning_rate": 3.2703289537824536e-06,
      "loss": 0.80451262,
      "num_input_tokens_seen": 54154595,
      "step": 2516,
      "time_per_iteration": 3.5870609283447266
    },
    {
      "auxiliary_loss_clip": 0.01147314,
      "auxiliary_loss_mlp": 0.0103644,
      "balance_loss_clip": 1.05164325,
      "balance_loss_mlp": 1.02794683,
      "epoch": 0.302651355738592,
      "flos": 18725367651840.0,
      "grad_norm": 3.060411624782866,
      "language_loss": 0.79172039,
      "learning_rate": 3.269727197731714e-06,
      "loss": 0.81355792,
      "num_input_tokens_seen": 54167360,
      "step": 2517,
      "time_per_iteration": 2.765566110610962
    },
    {
      "auxiliary_loss_clip": 0.01136225,
      "auxiliary_loss_mlp": 0.01025685,
      "balance_loss_clip": 1.05282211,
      "balance_loss_mlp": 1.0171262,
      "epoch": 0.30277159862923103,
      "flos": 22418888382720.0,
      "grad_norm": 2.4456448979672216,
      "language_loss": 0.77986515,
      "learning_rate": 3.269125249064367e-06,
      "loss": 0.80148423,
      "num_input_tokens_seen": 54187055,
      "step": 2518,
      "time_per_iteration": 2.7988667488098145
    },
    {
      "auxiliary_loss_clip": 0.01202,
      "auxiliary_loss_mlp": 0.0102895,
      "balance_loss_clip": 1.05745769,
      "balance_loss_mlp": 1.01981282,
      "epoch": 0.30289184151987014,
      "flos": 22273126992000.0,
      "grad_norm": 1.6987891023528463,
      "language_loss": 0.83344972,
      "learning_rate": 3.2685231078717297e-06,
      "loss": 0.85575926,
      "num_input_tokens_seen": 54207245,
      "step": 2519,
      "time_per_iteration": 2.7061755657196045
    },
    {
      "auxiliary_loss_clip": 0.01149664,
      "auxiliary_loss_mlp": 0.00764356,
      "balance_loss_clip": 1.05249119,
      "balance_loss_mlp": 1.00040722,
      "epoch": 0.30301208441050925,
      "flos": 25225594231680.0,
      "grad_norm": 2.105991573089684,
      "language_loss": 0.75633693,
      "learning_rate": 3.267920774245145e-06,
      "loss": 0.77547717,
      "num_input_tokens_seen": 54226650,
      "step": 2520,
      "time_per_iteration": 2.7748045921325684
    },
    {
      "auxiliary_loss_clip": 0.01187463,
      "auxiliary_loss_mlp": 0.01035517,
      "balance_loss_clip": 1.05669475,
      "balance_loss_mlp": 1.02579522,
      "epoch": 0.3031323273011483,
      "flos": 23039245198080.0,
      "grad_norm": 2.1368505671351214,
      "language_loss": 0.84306633,
      "learning_rate": 3.2673182482759876e-06,
      "loss": 0.86529607,
      "num_input_tokens_seen": 54245765,
      "step": 2521,
      "time_per_iteration": 2.689213514328003
    },
    {
      "auxiliary_loss_clip": 0.01186175,
      "auxiliary_loss_mlp": 0.01025656,
      "balance_loss_clip": 1.05663514,
      "balance_loss_mlp": 1.01777601,
      "epoch": 0.3032525701917874,
      "flos": 18876695650560.0,
      "grad_norm": 1.985118739777771,
      "language_loss": 0.66315633,
      "learning_rate": 3.266715530055659e-06,
      "loss": 0.68527466,
      "num_input_tokens_seen": 54263915,
      "step": 2522,
      "time_per_iteration": 2.6304843425750732
    },
    {
      "auxiliary_loss_clip": 0.01179387,
      "auxiliary_loss_mlp": 0.01032416,
      "balance_loss_clip": 1.05462635,
      "balance_loss_mlp": 1.02378535,
      "epoch": 0.30337281308242653,
      "flos": 17782641250560.0,
      "grad_norm": 2.238378980812776,
      "language_loss": 0.80554807,
      "learning_rate": 3.2661126196755927e-06,
      "loss": 0.82766616,
      "num_input_tokens_seen": 54283025,
      "step": 2523,
      "time_per_iteration": 2.6478872299194336
    },
    {
      "auxiliary_loss_clip": 0.01099328,
      "auxiliary_loss_mlp": 0.01003083,
      "balance_loss_clip": 1.02046442,
      "balance_loss_mlp": 1.00135469,
      "epoch": 0.3034930559730656,
      "flos": 57824298426240.0,
      "grad_norm": 0.7816601314695385,
      "language_loss": 0.55879402,
      "learning_rate": 3.265509517227248e-06,
      "loss": 0.57981813,
      "num_input_tokens_seen": 54339840,
      "step": 2524,
      "time_per_iteration": 3.203942060470581
    },
    {
      "auxiliary_loss_clip": 0.01168959,
      "auxiliary_loss_mlp": 0.01029411,
      "balance_loss_clip": 1.05125678,
      "balance_loss_mlp": 1.02125156,
      "epoch": 0.3036132988637047,
      "flos": 14755587419520.0,
      "grad_norm": 1.9100876056966332,
      "language_loss": 0.8088308,
      "learning_rate": 3.264906222802115e-06,
      "loss": 0.83081448,
      "num_input_tokens_seen": 54357690,
      "step": 2525,
      "time_per_iteration": 2.680345296859741
    },
    {
      "auxiliary_loss_clip": 0.01202144,
      "auxiliary_loss_mlp": 0.01030922,
      "balance_loss_clip": 1.05718446,
      "balance_loss_mlp": 1.02211833,
      "epoch": 0.30373354175434375,
      "flos": 21033203460480.0,
      "grad_norm": 2.870908680385623,
      "language_loss": 0.78035665,
      "learning_rate": 3.264302736491715e-06,
      "loss": 0.80268729,
      "num_input_tokens_seen": 54377810,
      "step": 2526,
      "time_per_iteration": 2.612384557723999
    },
    {
      "auxiliary_loss_clip": 0.01188616,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 1.06028306,
      "balance_loss_mlp": 1.02248907,
      "epoch": 0.30385378464498286,
      "flos": 21143233797120.0,
      "grad_norm": 2.5467490701723325,
      "language_loss": 0.87875843,
      "learning_rate": 3.263699058387594e-06,
      "loss": 0.90095288,
      "num_input_tokens_seen": 54395245,
      "step": 2527,
      "time_per_iteration": 2.6477293968200684
    },
    {
      "auxiliary_loss_clip": 0.01153217,
      "auxiliary_loss_mlp": 0.01034518,
      "balance_loss_clip": 1.05004728,
      "balance_loss_mlp": 1.02539277,
      "epoch": 0.30397402753562197,
      "flos": 20629244131200.0,
      "grad_norm": 2.2197968156811747,
      "language_loss": 0.89905256,
      "learning_rate": 3.2630951885813315e-06,
      "loss": 0.92092991,
      "num_input_tokens_seen": 54412640,
      "step": 2528,
      "time_per_iteration": 2.732787609100342
    },
    {
      "auxiliary_loss_clip": 0.01173083,
      "auxiliary_loss_mlp": 0.01028987,
      "balance_loss_clip": 1.05082035,
      "balance_loss_mlp": 1.02044559,
      "epoch": 0.304094270426261,
      "flos": 15085678429440.0,
      "grad_norm": 2.0364507872405646,
      "language_loss": 0.78031754,
      "learning_rate": 3.262491127164533e-06,
      "loss": 0.80233824,
      "num_input_tokens_seen": 54431455,
      "step": 2529,
      "time_per_iteration": 2.6846084594726562
    },
    {
      "auxiliary_loss_clip": 0.01180586,
      "auxiliary_loss_mlp": 0.00764064,
      "balance_loss_clip": 1.05590427,
      "balance_loss_mlp": 1.00057173,
      "epoch": 0.30421451331690014,
      "flos": 13845216193920.0,
      "grad_norm": 2.7802427377644543,
      "language_loss": 0.80142665,
      "learning_rate": 3.2618868742288337e-06,
      "loss": 0.82087314,
      "num_input_tokens_seen": 54448380,
      "step": 2530,
      "time_per_iteration": 2.7011606693267822
    },
    {
      "auxiliary_loss_clip": 0.01188221,
      "auxiliary_loss_mlp": 0.01029013,
      "balance_loss_clip": 1.05889666,
      "balance_loss_mlp": 1.01973224,
      "epoch": 0.30433475620753925,
      "flos": 17384212615680.0,
      "grad_norm": 2.623697958218013,
      "language_loss": 0.72592634,
      "learning_rate": 3.261282429865899e-06,
      "loss": 0.74809873,
      "num_input_tokens_seen": 54466385,
      "step": 2531,
      "time_per_iteration": 2.6986594200134277
    },
    {
      "auxiliary_loss_clip": 0.01177399,
      "auxiliary_loss_mlp": 0.00763666,
      "balance_loss_clip": 1.0570972,
      "balance_loss_mlp": 1.00045156,
      "epoch": 0.3044549990981783,
      "flos": 18916951818240.0,
      "grad_norm": 2.0196712735308093,
      "language_loss": 0.72308493,
      "learning_rate": 3.2606777941674225e-06,
      "loss": 0.7424956,
      "num_input_tokens_seen": 54485040,
      "step": 2532,
      "time_per_iteration": 2.747828483581543
    },
    {
      "auxiliary_loss_clip": 0.01134296,
      "auxiliary_loss_mlp": 0.01030882,
      "balance_loss_clip": 1.05026889,
      "balance_loss_mlp": 1.02225077,
      "epoch": 0.3045752419888174,
      "flos": 21068431724160.0,
      "grad_norm": 2.148986370739615,
      "language_loss": 0.8448025,
      "learning_rate": 3.2600729672251276e-06,
      "loss": 0.86645424,
      "num_input_tokens_seen": 54502755,
      "step": 2533,
      "time_per_iteration": 3.7471518516540527
    },
    {
      "auxiliary_loss_clip": 0.01203273,
      "auxiliary_loss_mlp": 0.0076386,
      "balance_loss_clip": 1.05990338,
      "balance_loss_mlp": 1.00040722,
      "epoch": 0.3046954848794565,
      "flos": 29096405516160.0,
      "grad_norm": 2.2088525381443502,
      "language_loss": 0.66078496,
      "learning_rate": 3.259467949130765e-06,
      "loss": 0.68045628,
      "num_input_tokens_seen": 54524165,
      "step": 2534,
      "time_per_iteration": 2.7948508262634277
    },
    {
      "auxiliary_loss_clip": 0.01176037,
      "auxiliary_loss_mlp": 0.01029919,
      "balance_loss_clip": 1.05819368,
      "balance_loss_mlp": 1.02074611,
      "epoch": 0.3048157277700956,
      "flos": 20295346279680.0,
      "grad_norm": 2.756156218830885,
      "language_loss": 0.82855809,
      "learning_rate": 3.2588627399761164e-06,
      "loss": 0.85061771,
      "num_input_tokens_seen": 54540160,
      "step": 2535,
      "time_per_iteration": 2.713184118270874
    },
    {
      "auxiliary_loss_clip": 0.01174256,
      "auxiliary_loss_mlp": 0.0103117,
      "balance_loss_clip": 1.05795288,
      "balance_loss_mlp": 1.02237868,
      "epoch": 0.3049359706607347,
      "flos": 22739929165440.0,
      "grad_norm": 1.9973814974975044,
      "language_loss": 0.70652384,
      "learning_rate": 3.2582573398529903e-06,
      "loss": 0.72857809,
      "num_input_tokens_seen": 54557515,
      "step": 2536,
      "time_per_iteration": 2.7494189739227295
    },
    {
      "auxiliary_loss_clip": 0.01158783,
      "auxiliary_loss_mlp": 0.01033747,
      "balance_loss_clip": 1.05181146,
      "balance_loss_mlp": 1.0247705,
      "epoch": 0.3050562135513738,
      "flos": 18434634969600.0,
      "grad_norm": 2.2338474890683324,
      "language_loss": 0.74292922,
      "learning_rate": 3.2576517488532265e-06,
      "loss": 0.76485455,
      "num_input_tokens_seen": 54573865,
      "step": 2537,
      "time_per_iteration": 3.6021199226379395
    },
    {
      "auxiliary_loss_clip": 0.0118657,
      "auxiliary_loss_mlp": 0.01026596,
      "balance_loss_clip": 1.05521274,
      "balance_loss_mlp": 1.01798892,
      "epoch": 0.30517645644201286,
      "flos": 20370327920640.0,
      "grad_norm": 1.870424540584384,
      "language_loss": 0.87736702,
      "learning_rate": 3.257045967068692e-06,
      "loss": 0.89949864,
      "num_input_tokens_seen": 54593120,
      "step": 2538,
      "time_per_iteration": 2.7130579948425293
    },
    {
      "auxiliary_loss_clip": 0.01204406,
      "auxiliary_loss_mlp": 0.01033336,
      "balance_loss_clip": 1.05968511,
      "balance_loss_mlp": 1.0242517,
      "epoch": 0.30529669933265197,
      "flos": 21945118970880.0,
      "grad_norm": 1.9850244089208842,
      "language_loss": 0.82122552,
      "learning_rate": 3.2564399945912848e-06,
      "loss": 0.8436029,
      "num_input_tokens_seen": 54612910,
      "step": 2539,
      "time_per_iteration": 2.6539738178253174
    },
    {
      "auxiliary_loss_clip": 0.0114717,
      "auxiliary_loss_mlp": 0.01034352,
      "balance_loss_clip": 1.05190563,
      "balance_loss_mlp": 1.02570391,
      "epoch": 0.305416942223291,
      "flos": 21835411856640.0,
      "grad_norm": 2.9930199525763252,
      "language_loss": 0.82151151,
      "learning_rate": 3.2558338315129287e-06,
      "loss": 0.84332675,
      "num_input_tokens_seen": 54631055,
      "step": 2540,
      "time_per_iteration": 3.7890100479125977
    },
    {
      "auxiliary_loss_clip": 0.01181407,
      "auxiliary_loss_mlp": 0.01028476,
      "balance_loss_clip": 1.05699944,
      "balance_loss_mlp": 1.02019048,
      "epoch": 0.30553718511393013,
      "flos": 33911810709120.0,
      "grad_norm": 3.6114232261528993,
      "language_loss": 0.7593928,
      "learning_rate": 3.2552274779255785e-06,
      "loss": 0.78149164,
      "num_input_tokens_seen": 54651985,
      "step": 2541,
      "time_per_iteration": 3.6540184020996094
    },
    {
      "auxiliary_loss_clip": 0.01186341,
      "auxiliary_loss_mlp": 0.01030843,
      "balance_loss_clip": 1.05877066,
      "balance_loss_mlp": 1.02228975,
      "epoch": 0.30565742800456924,
      "flos": 22268530051200.0,
      "grad_norm": 2.270128575365527,
      "language_loss": 0.77071536,
      "learning_rate": 3.2546209339212184e-06,
      "loss": 0.79288715,
      "num_input_tokens_seen": 54671005,
      "step": 2542,
      "time_per_iteration": 2.69398832321167
    },
    {
      "auxiliary_loss_clip": 0.01173758,
      "auxiliary_loss_mlp": 0.01027191,
      "balance_loss_clip": 1.0535332,
      "balance_loss_mlp": 1.01909649,
      "epoch": 0.3057776708952083,
      "flos": 22565044823040.0,
      "grad_norm": 1.7821031026213785,
      "language_loss": 0.77578527,
      "learning_rate": 3.25401419959186e-06,
      "loss": 0.79779476,
      "num_input_tokens_seen": 54691615,
      "step": 2543,
      "time_per_iteration": 2.783055067062378
    },
    {
      "auxiliary_loss_clip": 0.01186232,
      "auxiliary_loss_mlp": 0.01032921,
      "balance_loss_clip": 1.06113386,
      "balance_loss_mlp": 1.0245347,
      "epoch": 0.3058979137858474,
      "flos": 21799213925760.0,
      "grad_norm": 2.0929887151446263,
      "language_loss": 0.76067483,
      "learning_rate": 3.253407275029545e-06,
      "loss": 0.78286636,
      "num_input_tokens_seen": 54710520,
      "step": 2544,
      "time_per_iteration": 2.7258830070495605
    },
    {
      "auxiliary_loss_clip": 0.01162606,
      "auxiliary_loss_mlp": 0.0103511,
      "balance_loss_clip": 1.05553555,
      "balance_loss_mlp": 1.02574563,
      "epoch": 0.3060181566764865,
      "flos": 26979435601920.0,
      "grad_norm": 1.8991332383649282,
      "language_loss": 0.80180836,
      "learning_rate": 3.2528001603263425e-06,
      "loss": 0.82378554,
      "num_input_tokens_seen": 54732590,
      "step": 2545,
      "time_per_iteration": 2.785421371459961
    },
    {
      "auxiliary_loss_clip": 0.01190103,
      "auxiliary_loss_mlp": 0.01032186,
      "balance_loss_clip": 1.06193876,
      "balance_loss_mlp": 1.02335262,
      "epoch": 0.3061383995671256,
      "flos": 19865101173120.0,
      "grad_norm": 2.3072770908602207,
      "language_loss": 0.81228584,
      "learning_rate": 3.2521928555743514e-06,
      "loss": 0.83450872,
      "num_input_tokens_seen": 54749935,
      "step": 2546,
      "time_per_iteration": 2.6757330894470215
    },
    {
      "auxiliary_loss_clip": 0.01167876,
      "auxiliary_loss_mlp": 0.00763363,
      "balance_loss_clip": 1.05493462,
      "balance_loss_mlp": 1.00052094,
      "epoch": 0.3062586424577647,
      "flos": 22127509255680.0,
      "grad_norm": 2.2559636464682122,
      "language_loss": 0.66995209,
      "learning_rate": 3.2515853608657e-06,
      "loss": 0.68926454,
      "num_input_tokens_seen": 54767935,
      "step": 2547,
      "time_per_iteration": 2.732675790786743
    },
    {
      "auxiliary_loss_clip": 0.01184236,
      "auxiliary_loss_mlp": 0.01034656,
      "balance_loss_clip": 1.05810332,
      "balance_loss_mlp": 1.0254178,
      "epoch": 0.3063788853484038,
      "flos": 20845497962880.0,
      "grad_norm": 3.181600729558309,
      "language_loss": 0.75058985,
      "learning_rate": 3.250977676292545e-06,
      "loss": 0.77277875,
      "num_input_tokens_seen": 54786175,
      "step": 2548,
      "time_per_iteration": 2.7223854064941406
    },
    {
      "auxiliary_loss_clip": 0.01176378,
      "auxiliary_loss_mlp": 0.01035136,
      "balance_loss_clip": 1.05557203,
      "balance_loss_mlp": 1.02632082,
      "epoch": 0.30649912823904285,
      "flos": 16209717707520.0,
      "grad_norm": 2.1858008360544545,
      "language_loss": 0.79537177,
      "learning_rate": 3.2503698019470712e-06,
      "loss": 0.81748694,
      "num_input_tokens_seen": 54801945,
      "step": 2549,
      "time_per_iteration": 2.711002826690674
    },
    {
      "auxiliary_loss_clip": 0.01185126,
      "auxiliary_loss_mlp": 0.0103239,
      "balance_loss_clip": 1.05447507,
      "balance_loss_mlp": 1.02398002,
      "epoch": 0.30661937112968196,
      "flos": 18617815353600.0,
      "grad_norm": 2.0406707051112374,
      "language_loss": 0.78466588,
      "learning_rate": 3.249761737921492e-06,
      "loss": 0.80684102,
      "num_input_tokens_seen": 54818475,
      "step": 2550,
      "time_per_iteration": 2.643123149871826
    },
    {
      "auxiliary_loss_clip": 0.01171799,
      "auxiliary_loss_mlp": 0.01035411,
      "balance_loss_clip": 1.05829358,
      "balance_loss_mlp": 1.02638125,
      "epoch": 0.30673961402032107,
      "flos": 31390809638400.0,
      "grad_norm": 2.1476164101227813,
      "language_loss": 0.74512768,
      "learning_rate": 3.249153484308051e-06,
      "loss": 0.76719975,
      "num_input_tokens_seen": 54837090,
      "step": 2551,
      "time_per_iteration": 2.7386698722839355
    },
    {
      "auxiliary_loss_clip": 0.01130321,
      "auxiliary_loss_mlp": 0.01030342,
      "balance_loss_clip": 1.04781103,
      "balance_loss_mlp": 1.02137709,
      "epoch": 0.3068598569109601,
      "flos": 20229809915520.0,
      "grad_norm": 2.5325233156785147,
      "language_loss": 0.77970266,
      "learning_rate": 3.2485450411990194e-06,
      "loss": 0.80130935,
      "num_input_tokens_seen": 54856445,
      "step": 2552,
      "time_per_iteration": 2.7836077213287354
    },
    {
      "auxiliary_loss_clip": 0.01202424,
      "auxiliary_loss_mlp": 0.01028879,
      "balance_loss_clip": 1.05735028,
      "balance_loss_mlp": 1.02009964,
      "epoch": 0.30698009980159924,
      "flos": 29601991399680.0,
      "grad_norm": 3.1293389870743047,
      "language_loss": 0.82261252,
      "learning_rate": 3.2479364086866983e-06,
      "loss": 0.84492558,
      "num_input_tokens_seen": 54876700,
      "step": 2553,
      "time_per_iteration": 2.6642520427703857
    },
    {
      "auxiliary_loss_clip": 0.01177213,
      "auxiliary_loss_mlp": 0.00764112,
      "balance_loss_clip": 1.05974722,
      "balance_loss_mlp": 1.00072467,
      "epoch": 0.30710034269223835,
      "flos": 23842423261440.0,
      "grad_norm": 1.900181989570923,
      "language_loss": 0.8152833,
      "learning_rate": 3.247327586863416e-06,
      "loss": 0.83469653,
      "num_input_tokens_seen": 54897580,
      "step": 2554,
      "time_per_iteration": 2.75164532661438
    },
    {
      "auxiliary_loss_clip": 0.01163208,
      "auxiliary_loss_mlp": 0.010293,
      "balance_loss_clip": 1.05371976,
      "balance_loss_mlp": 1.01984692,
      "epoch": 0.3072205855828774,
      "flos": 25884986152320.0,
      "grad_norm": 2.2947512867564877,
      "language_loss": 0.76993906,
      "learning_rate": 3.2467185758215304e-06,
      "loss": 0.7918641,
      "num_input_tokens_seen": 54917320,
      "step": 2555,
      "time_per_iteration": 2.7952990531921387
    },
    {
      "auxiliary_loss_clip": 0.01164664,
      "auxiliary_loss_mlp": 0.00763116,
      "balance_loss_clip": 1.05754578,
      "balance_loss_mlp": 1.00062752,
      "epoch": 0.3073408284735165,
      "flos": 22236390357120.0,
      "grad_norm": 2.3535564430192712,
      "language_loss": 0.85708416,
      "learning_rate": 3.246109375653428e-06,
      "loss": 0.87636197,
      "num_input_tokens_seen": 54934085,
      "step": 2556,
      "time_per_iteration": 2.7071268558502197
    },
    {
      "auxiliary_loss_clip": 0.01201734,
      "auxiliary_loss_mlp": 0.01035999,
      "balance_loss_clip": 1.0585382,
      "balance_loss_mlp": 1.0272193,
      "epoch": 0.30746107136415557,
      "flos": 19500284689920.0,
      "grad_norm": 2.9537839152332617,
      "language_loss": 0.78637147,
      "learning_rate": 3.2454999864515243e-06,
      "loss": 0.80874878,
      "num_input_tokens_seen": 54953460,
      "step": 2557,
      "time_per_iteration": 2.588740825653076
    },
    {
      "auxiliary_loss_clip": 0.0116739,
      "auxiliary_loss_mlp": 0.00764454,
      "balance_loss_clip": 1.05395174,
      "balance_loss_mlp": 1.00051045,
      "epoch": 0.3075813142547947,
      "flos": 21724806902400.0,
      "grad_norm": 1.9159072587725827,
      "language_loss": 0.69212139,
      "learning_rate": 3.244890408308263e-06,
      "loss": 0.71143985,
      "num_input_tokens_seen": 54974165,
      "step": 2558,
      "time_per_iteration": 2.6535444259643555
    },
    {
      "auxiliary_loss_clip": 0.01143364,
      "auxiliary_loss_mlp": 0.01025364,
      "balance_loss_clip": 1.04961431,
      "balance_loss_mlp": 1.01696527,
      "epoch": 0.3077015571454338,
      "flos": 24097963593600.0,
      "grad_norm": 2.0314973443564632,
      "language_loss": 0.61608171,
      "learning_rate": 3.2442806413161165e-06,
      "loss": 0.63776898,
      "num_input_tokens_seen": 54993810,
      "step": 2559,
      "time_per_iteration": 3.679738759994507
    },
    {
      "auxiliary_loss_clip": 0.0115003,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 1.05430663,
      "balance_loss_mlp": 1.01621985,
      "epoch": 0.30782180003607285,
      "flos": 18405476104320.0,
      "grad_norm": 2.4819398896672857,
      "language_loss": 0.75627804,
      "learning_rate": 3.243670685567586e-06,
      "loss": 0.77802455,
      "num_input_tokens_seen": 55011210,
      "step": 2560,
      "time_per_iteration": 2.707723617553711
    },
    {
      "auxiliary_loss_clip": 0.01173032,
      "auxiliary_loss_mlp": 0.00764056,
      "balance_loss_clip": 1.05682278,
      "balance_loss_mlp": 1.00051355,
      "epoch": 0.30794204292671196,
      "flos": 23878549365120.0,
      "grad_norm": 2.746445225789566,
      "language_loss": 0.7982195,
      "learning_rate": 3.2430605411552012e-06,
      "loss": 0.81759042,
      "num_input_tokens_seen": 55031325,
      "step": 2561,
      "time_per_iteration": 2.740462303161621
    },
    {
      "auxiliary_loss_clip": 0.01076164,
      "auxiliary_loss_mlp": 0.01008684,
      "balance_loss_clip": 1.02472615,
      "balance_loss_mlp": 1.00696778,
      "epoch": 0.30806228581735107,
      "flos": 67927800816000.0,
      "grad_norm": 0.8941780687140051,
      "language_loss": 0.70603007,
      "learning_rate": 3.2424502081715205e-06,
      "loss": 0.72687852,
      "num_input_tokens_seen": 55094440,
      "step": 2562,
      "time_per_iteration": 3.3082668781280518
    },
    {
      "auxiliary_loss_clip": 0.01176855,
      "auxiliary_loss_mlp": 0.01032376,
      "balance_loss_clip": 1.05657792,
      "balance_loss_mlp": 1.02404356,
      "epoch": 0.3081825287079901,
      "flos": 23843213360640.0,
      "grad_norm": 2.396378388871925,
      "language_loss": 0.78102612,
      "learning_rate": 3.241839686709132e-06,
      "loss": 0.80311847,
      "num_input_tokens_seen": 55115375,
      "step": 2563,
      "time_per_iteration": 3.642937421798706
    },
    {
      "auxiliary_loss_clip": 0.01184281,
      "auxiliary_loss_mlp": 0.0102845,
      "balance_loss_clip": 1.05328131,
      "balance_loss_mlp": 1.01915121,
      "epoch": 0.30830277159862923,
      "flos": 16209969102720.0,
      "grad_norm": 2.426865793171944,
      "language_loss": 0.82155353,
      "learning_rate": 3.2412289768606495e-06,
      "loss": 0.84368086,
      "num_input_tokens_seen": 55131945,
      "step": 2564,
      "time_per_iteration": 2.639086961746216
    },
    {
      "auxiliary_loss_clip": 0.01190844,
      "auxiliary_loss_mlp": 0.01032871,
      "balance_loss_clip": 1.05612886,
      "balance_loss_mlp": 1.02413273,
      "epoch": 0.30842301448926834,
      "flos": 29349503723520.0,
      "grad_norm": 2.163959129570945,
      "language_loss": 0.82894349,
      "learning_rate": 3.240618078718718e-06,
      "loss": 0.85118067,
      "num_input_tokens_seen": 55153405,
      "step": 2565,
      "time_per_iteration": 2.6950745582580566
    },
    {
      "auxiliary_loss_clip": 0.01157836,
      "auxiliary_loss_mlp": 0.0103485,
      "balance_loss_clip": 1.05172563,
      "balance_loss_mlp": 1.02545643,
      "epoch": 0.3085432573799074,
      "flos": 21945190798080.0,
      "grad_norm": 2.1595072409700844,
      "language_loss": 0.74630737,
      "learning_rate": 3.240006992376011e-06,
      "loss": 0.76823425,
      "num_input_tokens_seen": 55173030,
      "step": 2566,
      "time_per_iteration": 3.673495292663574
    },
    {
      "auxiliary_loss_clip": 0.01179735,
      "auxiliary_loss_mlp": 0.0103074,
      "balance_loss_clip": 1.05790925,
      "balance_loss_mlp": 1.02188241,
      "epoch": 0.3086635002705465,
      "flos": 22054718344320.0,
      "grad_norm": 2.817728080030157,
      "language_loss": 0.75982308,
      "learning_rate": 3.2393957179252284e-06,
      "loss": 0.78192782,
      "num_input_tokens_seen": 55189565,
      "step": 2567,
      "time_per_iteration": 3.53012752532959
    },
    {
      "auxiliary_loss_clip": 0.01203306,
      "auxiliary_loss_mlp": 0.01033142,
      "balance_loss_clip": 1.05854177,
      "balance_loss_mlp": 1.02387977,
      "epoch": 0.3087837431611856,
      "flos": 32665925520000.0,
      "grad_norm": 2.043182069824605,
      "language_loss": 0.80890149,
      "learning_rate": 3.2387842554591016e-06,
      "loss": 0.83126593,
      "num_input_tokens_seen": 55210380,
      "step": 2568,
      "time_per_iteration": 2.682248115539551
    },
    {
      "auxiliary_loss_clip": 0.01205315,
      "auxiliary_loss_mlp": 0.01034431,
      "balance_loss_clip": 1.05966926,
      "balance_loss_mlp": 1.02535868,
      "epoch": 0.3089039860518247,
      "flos": 17599245384960.0,
      "grad_norm": 2.261043135377255,
      "language_loss": 0.88242084,
      "learning_rate": 3.238172605070388e-06,
      "loss": 0.9048183,
      "num_input_tokens_seen": 55225795,
      "step": 2569,
      "time_per_iteration": 2.5525317192077637
    },
    {
      "auxiliary_loss_clip": 0.01189687,
      "auxiliary_loss_mlp": 0.00764496,
      "balance_loss_clip": 1.0595243,
      "balance_loss_mlp": 1.00047648,
      "epoch": 0.3090242289424638,
      "flos": 14383839611520.0,
      "grad_norm": 8.008105498194258,
      "language_loss": 0.78411329,
      "learning_rate": 3.2375607668518745e-06,
      "loss": 0.80365515,
      "num_input_tokens_seen": 55238830,
      "step": 2570,
      "time_per_iteration": 2.591534376144409
    },
    {
      "auxiliary_loss_clip": 0.01164644,
      "auxiliary_loss_mlp": 0.01031063,
      "balance_loss_clip": 1.0544374,
      "balance_loss_mlp": 1.0217768,
      "epoch": 0.30914447183310284,
      "flos": 16068625084800.0,
      "grad_norm": 2.1591151238592325,
      "language_loss": 0.90010691,
      "learning_rate": 3.236948740896377e-06,
      "loss": 0.92206395,
      "num_input_tokens_seen": 55253630,
      "step": 2571,
      "time_per_iteration": 2.6720001697540283
    },
    {
      "auxiliary_loss_clip": 0.01189599,
      "auxiliary_loss_mlp": 0.01038132,
      "balance_loss_clip": 1.05844569,
      "balance_loss_mlp": 1.02811861,
      "epoch": 0.30926471472374195,
      "flos": 32230221546240.0,
      "grad_norm": 33.418514357449105,
      "language_loss": 0.84684944,
      "learning_rate": 3.2363365272967384e-06,
      "loss": 0.8691268,
      "num_input_tokens_seen": 55276200,
      "step": 2572,
      "time_per_iteration": 2.722968101501465
    },
    {
      "auxiliary_loss_clip": 0.0119176,
      "auxiliary_loss_mlp": 0.01039983,
      "balance_loss_clip": 1.06210589,
      "balance_loss_mlp": 1.02982664,
      "epoch": 0.30938495761438106,
      "flos": 20370722970240.0,
      "grad_norm": 2.2945231579371947,
      "language_loss": 0.81879592,
      "learning_rate": 3.235724126145832e-06,
      "loss": 0.84111333,
      "num_input_tokens_seen": 55292235,
      "step": 2573,
      "time_per_iteration": 2.678511381149292
    },
    {
      "auxiliary_loss_clip": 0.01177926,
      "auxiliary_loss_mlp": 0.01031551,
      "balance_loss_clip": 1.05394983,
      "balance_loss_mlp": 1.02251506,
      "epoch": 0.3095052005050201,
      "flos": 24061155131520.0,
      "grad_norm": 1.7425188555628297,
      "language_loss": 0.78034097,
      "learning_rate": 3.235111537536558e-06,
      "loss": 0.80243576,
      "num_input_tokens_seen": 55313050,
      "step": 2574,
      "time_per_iteration": 2.7052993774414062
    },
    {
      "auxiliary_loss_clip": 0.0118958,
      "auxiliary_loss_mlp": 0.010319,
      "balance_loss_clip": 1.05735552,
      "balance_loss_mlp": 1.02317953,
      "epoch": 0.30962544339565923,
      "flos": 23401547729280.0,
      "grad_norm": 2.4435278207919637,
      "language_loss": 0.83063269,
      "learning_rate": 3.2344987615618456e-06,
      "loss": 0.85284752,
      "num_input_tokens_seen": 55332885,
      "step": 2575,
      "time_per_iteration": 2.6447343826293945
    },
    {
      "auxiliary_loss_clip": 0.01159932,
      "auxiliary_loss_mlp": 0.01035848,
      "balance_loss_clip": 1.05717945,
      "balance_loss_mlp": 1.02760434,
      "epoch": 0.30974568628629834,
      "flos": 33799984692480.0,
      "grad_norm": 1.8565731782668493,
      "language_loss": 0.78786826,
      "learning_rate": 3.2338857983146533e-06,
      "loss": 0.80982608,
      "num_input_tokens_seen": 55354385,
      "step": 2576,
      "time_per_iteration": 2.9394004344940186
    },
    {
      "auxiliary_loss_clip": 0.01164945,
      "auxiliary_loss_mlp": 0.01033012,
      "balance_loss_clip": 1.05483437,
      "balance_loss_mlp": 1.02386248,
      "epoch": 0.3098659291769374,
      "flos": 20229594433920.0,
      "grad_norm": 1.8312249764822368,
      "language_loss": 0.76357073,
      "learning_rate": 3.233272647887966e-06,
      "loss": 0.7855503,
      "num_input_tokens_seen": 55373275,
      "step": 2577,
      "time_per_iteration": 2.6968865394592285
    },
    {
      "auxiliary_loss_clip": 0.01205478,
      "auxiliary_loss_mlp": 0.01030554,
      "balance_loss_clip": 1.06135154,
      "balance_loss_mlp": 1.02136874,
      "epoch": 0.3099861720675765,
      "flos": 24748556682240.0,
      "grad_norm": 1.6927839550756358,
      "language_loss": 0.90434593,
      "learning_rate": 3.2326593103747985e-06,
      "loss": 0.92670631,
      "num_input_tokens_seen": 55392290,
      "step": 2578,
      "time_per_iteration": 2.6784839630126953
    },
    {
      "auxiliary_loss_clip": 0.01188863,
      "auxiliary_loss_mlp": 0.01030454,
      "balance_loss_clip": 1.06004572,
      "balance_loss_mlp": 1.02151322,
      "epoch": 0.3101064149582156,
      "flos": 11765485704960.0,
      "grad_norm": 2.6965111785922296,
      "language_loss": 0.8487767,
      "learning_rate": 3.2320457858681936e-06,
      "loss": 0.87096989,
      "num_input_tokens_seen": 55410680,
      "step": 2579,
      "time_per_iteration": 2.641016960144043
    },
    {
      "auxiliary_loss_clip": 0.01171592,
      "auxiliary_loss_mlp": 0.01034938,
      "balance_loss_clip": 1.05468464,
      "balance_loss_mlp": 1.02694511,
      "epoch": 0.31022665784885467,
      "flos": 23033247626880.0,
      "grad_norm": 3.71921468736521,
      "language_loss": 0.85400718,
      "learning_rate": 3.2314320744612228e-06,
      "loss": 0.87607253,
      "num_input_tokens_seen": 55425980,
      "step": 2580,
      "time_per_iteration": 2.734492540359497
    },
    {
      "auxiliary_loss_clip": 0.01184975,
      "auxiliary_loss_mlp": 0.01032553,
      "balance_loss_clip": 1.05757487,
      "balance_loss_mlp": 1.02384496,
      "epoch": 0.3103469007394938,
      "flos": 16289188548480.0,
      "grad_norm": 1.650737736578647,
      "language_loss": 0.76373357,
      "learning_rate": 3.2308181762469854e-06,
      "loss": 0.78590882,
      "num_input_tokens_seen": 55443925,
      "step": 2581,
      "time_per_iteration": 2.591212272644043
    },
    {
      "auxiliary_loss_clip": 0.01206415,
      "auxiliary_loss_mlp": 0.01031096,
      "balance_loss_clip": 1.05939579,
      "balance_loss_mlp": 1.02225721,
      "epoch": 0.3104671436301329,
      "flos": 30515271626880.0,
      "grad_norm": 2.09458243121925,
      "language_loss": 0.78497112,
      "learning_rate": 3.230204091318609e-06,
      "loss": 0.80734622,
      "num_input_tokens_seen": 55464465,
      "step": 2582,
      "time_per_iteration": 2.651573657989502
    },
    {
      "auxiliary_loss_clip": 0.01204779,
      "auxiliary_loss_mlp": 0.00764298,
      "balance_loss_clip": 1.06062913,
      "balance_loss_mlp": 1.00055909,
      "epoch": 0.31058738652077195,
      "flos": 20047240062720.0,
      "grad_norm": 2.0390617803814455,
      "language_loss": 0.84664947,
      "learning_rate": 3.2295898197692503e-06,
      "loss": 0.86634022,
      "num_input_tokens_seen": 55483425,
      "step": 2583,
      "time_per_iteration": 2.5570414066314697
    },
    {
      "auxiliary_loss_clip": 0.01201967,
      "auxiliary_loss_mlp": 0.01034719,
      "balance_loss_clip": 1.05916786,
      "balance_loss_mlp": 1.02615333,
      "epoch": 0.31070762941141106,
      "flos": 28074639237120.0,
      "grad_norm": 1.769656889703578,
      "language_loss": 0.79186171,
      "learning_rate": 3.228975361692094e-06,
      "loss": 0.81422859,
      "num_input_tokens_seen": 55504445,
      "step": 2584,
      "time_per_iteration": 2.6346144676208496
    },
    {
      "auxiliary_loss_clip": 0.01192792,
      "auxiliary_loss_mlp": 0.00764988,
      "balance_loss_clip": 1.05662787,
      "balance_loss_mlp": 1.00067759,
      "epoch": 0.31082787230205017,
      "flos": 20521907314560.0,
      "grad_norm": 2.2628548258909347,
      "language_loss": 0.80352134,
      "learning_rate": 3.228360717180352e-06,
      "loss": 0.82309914,
      "num_input_tokens_seen": 55521970,
      "step": 2585,
      "time_per_iteration": 3.8267152309417725
    },
    {
      "auxiliary_loss_clip": 0.01110977,
      "auxiliary_loss_mlp": 0.00755242,
      "balance_loss_clip": 1.03085256,
      "balance_loss_mlp": 1.00013697,
      "epoch": 0.3109481151926892,
      "flos": 62445928723200.0,
      "grad_norm": 0.8115967776018719,
      "language_loss": 0.59379828,
      "learning_rate": 3.227745886327266e-06,
      "loss": 0.61246049,
      "num_input_tokens_seen": 55580665,
      "step": 2586,
      "time_per_iteration": 3.1432158946990967
    },
    {
      "auxiliary_loss_clip": 0.01111219,
      "auxiliary_loss_mlp": 0.01007313,
      "balance_loss_clip": 1.03095448,
      "balance_loss_mlp": 1.00562024,
      "epoch": 0.31106835808332833,
      "flos": 44746744723200.0,
      "grad_norm": 0.8258452167089132,
      "language_loss": 0.55849344,
      "learning_rate": 3.227130869226105e-06,
      "loss": 0.57967877,
      "num_input_tokens_seen": 55637825,
      "step": 2587,
      "time_per_iteration": 3.260242462158203
    },
    {
      "auxiliary_loss_clip": 0.01187964,
      "auxiliary_loss_mlp": 0.01033791,
      "balance_loss_clip": 1.05566764,
      "balance_loss_mlp": 1.02496934,
      "epoch": 0.3111886009739674,
      "flos": 23403056100480.0,
      "grad_norm": 6.370900320680225,
      "language_loss": 0.82553178,
      "learning_rate": 3.226515665970167e-06,
      "loss": 0.84774935,
      "num_input_tokens_seen": 55655365,
      "step": 2588,
      "time_per_iteration": 3.6194047927856445
    },
    {
      "auxiliary_loss_clip": 0.01185723,
      "auxiliary_loss_mlp": 0.01029111,
      "balance_loss_clip": 1.05610681,
      "balance_loss_mlp": 1.01943672,
      "epoch": 0.3113088438646065,
      "flos": 17530728192000.0,
      "grad_norm": 2.4252852462791457,
      "language_loss": 0.86650389,
      "learning_rate": 3.225900276652777e-06,
      "loss": 0.88865227,
      "num_input_tokens_seen": 55672140,
      "step": 2589,
      "time_per_iteration": 2.686906337738037
    },
    {
      "auxiliary_loss_clip": 0.0117771,
      "auxiliary_loss_mlp": 0.0102862,
      "balance_loss_clip": 1.05460358,
      "balance_loss_mlp": 1.01999474,
      "epoch": 0.3114290867552456,
      "flos": 28365802882560.0,
      "grad_norm": 1.736884287791215,
      "language_loss": 0.76096851,
      "learning_rate": 3.2252847013672906e-06,
      "loss": 0.78303182,
      "num_input_tokens_seen": 55694800,
      "step": 2590,
      "time_per_iteration": 2.7721645832061768
    },
    {
      "auxiliary_loss_clip": 0.01150733,
      "auxiliary_loss_mlp": 0.01029742,
      "balance_loss_clip": 1.0504967,
      "balance_loss_mlp": 1.02073538,
      "epoch": 0.31154932964588467,
      "flos": 27379157126400.0,
      "grad_norm": 2.086670221297666,
      "language_loss": 0.76929253,
      "learning_rate": 3.224668940207089e-06,
      "loss": 0.79109728,
      "num_input_tokens_seen": 55713785,
      "step": 2591,
      "time_per_iteration": 2.733731269836426
    },
    {
      "auxiliary_loss_clip": 0.0113611,
      "auxiliary_loss_mlp": 0.01034048,
      "balance_loss_clip": 1.04721594,
      "balance_loss_mlp": 1.02505994,
      "epoch": 0.3116695725365238,
      "flos": 26541864120960.0,
      "grad_norm": 2.4930027815179763,
      "language_loss": 0.87789726,
      "learning_rate": 3.2240529932655828e-06,
      "loss": 0.89959884,
      "num_input_tokens_seen": 55733050,
      "step": 2592,
      "time_per_iteration": 3.8972930908203125
    },
    {
      "auxiliary_loss_clip": 0.01171551,
      "auxiliary_loss_mlp": 0.01035058,
      "balance_loss_clip": 1.05526781,
      "balance_loss_mlp": 1.02626622,
      "epoch": 0.3117898154271629,
      "flos": 21177600134400.0,
      "grad_norm": 3.39127244373679,
      "language_loss": 0.88870031,
      "learning_rate": 3.223436860636211e-06,
      "loss": 0.91076636,
      "num_input_tokens_seen": 55748685,
      "step": 2593,
      "time_per_iteration": 2.6810288429260254
    },
    {
      "auxiliary_loss_clip": 0.01204228,
      "auxiliary_loss_mlp": 0.01034554,
      "balance_loss_clip": 1.0593338,
      "balance_loss_mlp": 1.02570868,
      "epoch": 0.31191005831780194,
      "flos": 27272430840960.0,
      "grad_norm": 1.734949811409212,
      "language_loss": 0.74262643,
      "learning_rate": 3.2228205424124403e-06,
      "loss": 0.76501429,
      "num_input_tokens_seen": 55771840,
      "step": 2594,
      "time_per_iteration": 3.597313165664673
    },
    {
      "auxiliary_loss_clip": 0.01157482,
      "auxiliary_loss_mlp": 0.01029059,
      "balance_loss_clip": 1.04964185,
      "balance_loss_mlp": 1.02071476,
      "epoch": 0.31203030120844105,
      "flos": 12963501043200.0,
      "grad_norm": 2.747775808366881,
      "language_loss": 0.74428678,
      "learning_rate": 3.222204038687765e-06,
      "loss": 0.7661522,
      "num_input_tokens_seen": 55784975,
      "step": 2595,
      "time_per_iteration": 2.689415454864502
    },
    {
      "auxiliary_loss_clip": 0.01182848,
      "auxiliary_loss_mlp": 0.01030733,
      "balance_loss_clip": 1.05498409,
      "balance_loss_mlp": 1.02143502,
      "epoch": 0.31215054409908016,
      "flos": 27562014288000.0,
      "grad_norm": 1.612437533210896,
      "language_loss": 0.88068271,
      "learning_rate": 3.221587349555709e-06,
      "loss": 0.9028185,
      "num_input_tokens_seen": 55805235,
      "step": 2596,
      "time_per_iteration": 2.6912341117858887
    },
    {
      "auxiliary_loss_clip": 0.01177662,
      "auxiliary_loss_mlp": 0.01029995,
      "balance_loss_clip": 1.05553377,
      "balance_loss_mlp": 1.02142978,
      "epoch": 0.3122707869897192,
      "flos": 21506326427520.0,
      "grad_norm": 1.8405239547389505,
      "language_loss": 0.69638354,
      "learning_rate": 3.2209704751098236e-06,
      "loss": 0.7184602,
      "num_input_tokens_seen": 55824265,
      "step": 2597,
      "time_per_iteration": 2.8742594718933105
    },
    {
      "auxiliary_loss_clip": 0.01175576,
      "auxiliary_loss_mlp": 0.01033424,
      "balance_loss_clip": 1.05699563,
      "balance_loss_mlp": 1.02375627,
      "epoch": 0.31239102988035833,
      "flos": 15187017674880.0,
      "grad_norm": 2.4360313229463686,
      "language_loss": 0.82853889,
      "learning_rate": 3.2203534154436875e-06,
      "loss": 0.85062897,
      "num_input_tokens_seen": 55838620,
      "step": 2598,
      "time_per_iteration": 2.731823682785034
    },
    {
      "auxiliary_loss_clip": 0.01127833,
      "auxiliary_loss_mlp": 0.01034909,
      "balance_loss_clip": 1.04914618,
      "balance_loss_mlp": 1.02638578,
      "epoch": 0.31251127277099744,
      "flos": 22053712763520.0,
      "grad_norm": 2.3233275236733264,
      "language_loss": 0.75887567,
      "learning_rate": 3.2197361706509084e-06,
      "loss": 0.78050315,
      "num_input_tokens_seen": 55859375,
      "step": 2599,
      "time_per_iteration": 2.8066349029541016
    },
    {
      "auxiliary_loss_clip": 0.01206307,
      "auxiliary_loss_mlp": 0.01029394,
      "balance_loss_clip": 1.05828881,
      "balance_loss_mlp": 1.01966023,
      "epoch": 0.3126315156616365,
      "flos": 15193984913280.0,
      "grad_norm": 2.7058289841667293,
      "language_loss": 0.83328021,
      "learning_rate": 3.2191187408251228e-06,
      "loss": 0.85563719,
      "num_input_tokens_seen": 55876535,
      "step": 2600,
      "time_per_iteration": 2.6307055950164795
    },
    {
      "auxiliary_loss_clip": 0.01191587,
      "auxiliary_loss_mlp": 0.01034022,
      "balance_loss_clip": 1.05425096,
      "balance_loss_mlp": 1.0241034,
      "epoch": 0.3127517585522756,
      "flos": 18145338831360.0,
      "grad_norm": 2.0859993487783512,
      "language_loss": 0.79221201,
      "learning_rate": 3.218501126059993e-06,
      "loss": 0.81446809,
      "num_input_tokens_seen": 55891930,
      "step": 2601,
      "time_per_iteration": 2.6731486320495605
    },
    {
      "auxiliary_loss_clip": 0.01189033,
      "auxiliary_loss_mlp": 0.01035055,
      "balance_loss_clip": 1.05473423,
      "balance_loss_mlp": 1.02631116,
      "epoch": 0.31287200144291466,
      "flos": 21908633731200.0,
      "grad_norm": 2.794102552641243,
      "language_loss": 0.81344163,
      "learning_rate": 3.2178833264492116e-06,
      "loss": 0.83568257,
      "num_input_tokens_seen": 55910635,
      "step": 2602,
      "time_per_iteration": 2.6586289405822754
    },
    {
      "auxiliary_loss_clip": 0.01194569,
      "auxiliary_loss_mlp": 0.01031131,
      "balance_loss_clip": 1.05801857,
      "balance_loss_mlp": 1.02139711,
      "epoch": 0.31299224433355377,
      "flos": 29896997800320.0,
      "grad_norm": 1.8485177835533062,
      "language_loss": 0.76011968,
      "learning_rate": 3.217265342086498e-06,
      "loss": 0.78237665,
      "num_input_tokens_seen": 55931125,
      "step": 2603,
      "time_per_iteration": 2.7473864555358887
    },
    {
      "auxiliary_loss_clip": 0.0116726,
      "auxiliary_loss_mlp": 0.00765199,
      "balance_loss_clip": 1.05670714,
      "balance_loss_mlp": 1.00078487,
      "epoch": 0.3131124872241929,
      "flos": 11655886331520.0,
      "grad_norm": 3.1814177350985555,
      "language_loss": 0.73139071,
      "learning_rate": 3.216647173065599e-06,
      "loss": 0.75071526,
      "num_input_tokens_seen": 55946590,
      "step": 2604,
      "time_per_iteration": 2.744267225265503
    },
    {
      "auxiliary_loss_clip": 0.01173404,
      "auxiliary_loss_mlp": 0.01028487,
      "balance_loss_clip": 1.05746365,
      "balance_loss_mlp": 1.01982093,
      "epoch": 0.31323273011483194,
      "flos": 49848785470080.0,
      "grad_norm": 2.421024362308577,
      "language_loss": 0.73993158,
      "learning_rate": 3.216028819480292e-06,
      "loss": 0.76195049,
      "num_input_tokens_seen": 55967930,
      "step": 2605,
      "time_per_iteration": 2.9302990436553955
    },
    {
      "auxiliary_loss_clip": 0.01162256,
      "auxiliary_loss_mlp": 0.01031832,
      "balance_loss_clip": 1.05264711,
      "balance_loss_mlp": 1.02152085,
      "epoch": 0.31335297300547105,
      "flos": 22601278667520.0,
      "grad_norm": 4.519481272155611,
      "language_loss": 0.75618207,
      "learning_rate": 3.2154102814243793e-06,
      "loss": 0.7781229,
      "num_input_tokens_seen": 55987070,
      "step": 2606,
      "time_per_iteration": 2.7568044662475586
    },
    {
      "auxiliary_loss_clip": 0.01167848,
      "auxiliary_loss_mlp": 0.0103244,
      "balance_loss_clip": 1.05643177,
      "balance_loss_mlp": 1.02314734,
      "epoch": 0.31347321589611016,
      "flos": 34710858708480.0,
      "grad_norm": 2.102517068844022,
      "language_loss": 0.67177868,
      "learning_rate": 3.2147915589916937e-06,
      "loss": 0.6937815,
      "num_input_tokens_seen": 56008630,
      "step": 2607,
      "time_per_iteration": 2.8514387607574463
    },
    {
      "auxiliary_loss_clip": 0.01165002,
      "auxiliary_loss_mlp": 0.01038883,
      "balance_loss_clip": 1.05089855,
      "balance_loss_mlp": 1.02938247,
      "epoch": 0.3135934587867492,
      "flos": 19755789108480.0,
      "grad_norm": 2.778112753052168,
      "language_loss": 0.8272658,
      "learning_rate": 3.2141726522760938e-06,
      "loss": 0.84930468,
      "num_input_tokens_seen": 56026690,
      "step": 2608,
      "time_per_iteration": 2.725917339324951
    },
    {
      "auxiliary_loss_clip": 0.01100471,
      "auxiliary_loss_mlp": 0.01005125,
      "balance_loss_clip": 1.03345561,
      "balance_loss_mlp": 1.00333643,
      "epoch": 0.3137137016773883,
      "flos": 65815535583360.0,
      "grad_norm": 0.7023265582996232,
      "language_loss": 0.52581531,
      "learning_rate": 3.213553561371469e-06,
      "loss": 0.54687124,
      "num_input_tokens_seen": 56090425,
      "step": 2609,
      "time_per_iteration": 3.306851863861084
    },
    {
      "auxiliary_loss_clip": 0.01148221,
      "auxiliary_loss_mlp": 0.0102752,
      "balance_loss_clip": 1.05588531,
      "balance_loss_mlp": 1.01841831,
      "epoch": 0.31383394456802743,
      "flos": 16252739222400.0,
      "grad_norm": 2.2322890902633183,
      "language_loss": 0.96001619,
      "learning_rate": 3.212934286371733e-06,
      "loss": 0.98177356,
      "num_input_tokens_seen": 56107135,
      "step": 2610,
      "time_per_iteration": 2.734653949737549
    },
    {
      "auxiliary_loss_clip": 0.01192152,
      "auxiliary_loss_mlp": 0.01039808,
      "balance_loss_clip": 1.06242371,
      "balance_loss_mlp": 1.03041482,
      "epoch": 0.3139541874586665,
      "flos": 38795517613440.0,
      "grad_norm": 13.816818044994886,
      "language_loss": 0.83839154,
      "learning_rate": 3.2123148273708304e-06,
      "loss": 0.8607111,
      "num_input_tokens_seen": 56127325,
      "step": 2611,
      "time_per_iteration": 3.7003185749053955
    },
    {
      "auxiliary_loss_clip": 0.01202129,
      "auxiliary_loss_mlp": 0.01031255,
      "balance_loss_clip": 1.05816519,
      "balance_loss_mlp": 1.02286267,
      "epoch": 0.3140744303493056,
      "flos": 25046328430080.0,
      "grad_norm": 2.025157068147185,
      "language_loss": 0.7687043,
      "learning_rate": 3.211695184462733e-06,
      "loss": 0.79103816,
      "num_input_tokens_seen": 56148500,
      "step": 2612,
      "time_per_iteration": 2.6558837890625
    },
    {
      "auxiliary_loss_clip": 0.01074829,
      "auxiliary_loss_mlp": 0.01001964,
      "balance_loss_clip": 1.02780521,
      "balance_loss_mlp": 1.00015187,
      "epoch": 0.3141946732399447,
      "flos": 72504254782080.0,
      "grad_norm": 0.8932223328667137,
      "language_loss": 0.60517657,
      "learning_rate": 3.2110753577414383e-06,
      "loss": 0.62594444,
      "num_input_tokens_seen": 56210080,
      "step": 2613,
      "time_per_iteration": 3.2563836574554443
    },
    {
      "auxiliary_loss_clip": 0.01177826,
      "auxiliary_loss_mlp": 0.01035769,
      "balance_loss_clip": 1.05482149,
      "balance_loss_mlp": 1.02657771,
      "epoch": 0.31431491613058377,
      "flos": 19239788280960.0,
      "grad_norm": 3.3333816410693298,
      "language_loss": 0.78780884,
      "learning_rate": 3.2104553473009757e-06,
      "loss": 0.80994475,
      "num_input_tokens_seen": 56228200,
      "step": 2614,
      "time_per_iteration": 3.575906991958618
    },
    {
      "auxiliary_loss_clip": 0.01139851,
      "auxiliary_loss_mlp": 0.01028918,
      "balance_loss_clip": 1.05139959,
      "balance_loss_mlp": 1.01966715,
      "epoch": 0.3144351590212229,
      "flos": 36210596290560.0,
      "grad_norm": 1.8917408290515596,
      "language_loss": 0.67458177,
      "learning_rate": 3.209835153235399e-06,
      "loss": 0.69626939,
      "num_input_tokens_seen": 56249755,
      "step": 2615,
      "time_per_iteration": 2.828031539916992
    },
    {
      "auxiliary_loss_clip": 0.01149615,
      "auxiliary_loss_mlp": 0.01034519,
      "balance_loss_clip": 1.0514437,
      "balance_loss_mlp": 1.02536988,
      "epoch": 0.314555401911862,
      "flos": 18551740285440.0,
      "grad_norm": 2.8531178427800357,
      "language_loss": 0.677369,
      "learning_rate": 3.2092147756387916e-06,
      "loss": 0.69921035,
      "num_input_tokens_seen": 56270080,
      "step": 2616,
      "time_per_iteration": 2.7846598625183105
    },
    {
      "auxiliary_loss_clip": 0.01164865,
      "auxiliary_loss_mlp": 0.0103692,
      "balance_loss_clip": 1.05082631,
      "balance_loss_mlp": 1.02774715,
      "epoch": 0.31467564480250104,
      "flos": 16362877299840.0,
      "grad_norm": 2.410927905975255,
      "language_loss": 0.83783311,
      "learning_rate": 3.208594214605264e-06,
      "loss": 0.85985094,
      "num_input_tokens_seen": 56288625,
      "step": 2617,
      "time_per_iteration": 2.683873176574707
    },
    {
      "auxiliary_loss_clip": 0.01164584,
      "auxiliary_loss_mlp": 0.01037123,
      "balance_loss_clip": 1.05425215,
      "balance_loss_mlp": 1.02831352,
      "epoch": 0.31479588769314015,
      "flos": 21652375127040.0,
      "grad_norm": 3.0099375339724115,
      "language_loss": 0.77368975,
      "learning_rate": 3.2079734702289553e-06,
      "loss": 0.79570675,
      "num_input_tokens_seen": 56307520,
      "step": 2618,
      "time_per_iteration": 2.646979570388794
    },
    {
      "auxiliary_loss_clip": 0.01091526,
      "auxiliary_loss_mlp": 0.00755501,
      "balance_loss_clip": 1.02663028,
      "balance_loss_mlp": 1.0001235,
      "epoch": 0.3149161305837792,
      "flos": 66051072040320.0,
      "grad_norm": 0.8264986375275456,
      "language_loss": 0.60435736,
      "learning_rate": 3.207352542604031e-06,
      "loss": 0.62282765,
      "num_input_tokens_seen": 56369855,
      "step": 2619,
      "time_per_iteration": 5.120925426483154
    },
    {
      "auxiliary_loss_clip": 0.01148998,
      "auxiliary_loss_mlp": 0.01030642,
      "balance_loss_clip": 1.05236936,
      "balance_loss_mlp": 1.02213001,
      "epoch": 0.3150363734744183,
      "flos": 28987201192320.0,
      "grad_norm": 1.6767756855932965,
      "language_loss": 0.78367007,
      "learning_rate": 3.2067314318246864e-06,
      "loss": 0.80546647,
      "num_input_tokens_seen": 56390570,
      "step": 2620,
      "time_per_iteration": 2.766129493713379
    },
    {
      "auxiliary_loss_clip": 0.01161504,
      "auxiliary_loss_mlp": 0.01034291,
      "balance_loss_clip": 1.05668855,
      "balance_loss_mlp": 1.02538013,
      "epoch": 0.31515661636505743,
      "flos": 27636600879360.0,
      "grad_norm": 2.3608924098108264,
      "language_loss": 0.78109753,
      "learning_rate": 3.206110137985143e-06,
      "loss": 0.80305552,
      "num_input_tokens_seen": 56410775,
      "step": 2621,
      "time_per_iteration": 2.7752673625946045
    },
    {
      "auxiliary_loss_clip": 0.01147499,
      "auxiliary_loss_mlp": 0.0103533,
      "balance_loss_clip": 1.05190694,
      "balance_loss_mlp": 1.0265677,
      "epoch": 0.3152768592556965,
      "flos": 24605632465920.0,
      "grad_norm": 1.8410629335559188,
      "language_loss": 0.92106497,
      "learning_rate": 3.2054886611796505e-06,
      "loss": 0.94289333,
      "num_input_tokens_seen": 56429770,
      "step": 2622,
      "time_per_iteration": 2.77343487739563
    },
    {
      "auxiliary_loss_clip": 0.0111333,
      "auxiliary_loss_mlp": 0.01002795,
      "balance_loss_clip": 1.03328633,
      "balance_loss_mlp": 1.00105429,
      "epoch": 0.3153971021463356,
      "flos": 68476908026880.0,
      "grad_norm": 0.9415554576475984,
      "language_loss": 0.63508356,
      "learning_rate": 3.204867001502487e-06,
      "loss": 0.65624481,
      "num_input_tokens_seen": 56488425,
      "step": 2623,
      "time_per_iteration": 3.2479238510131836
    },
    {
      "auxiliary_loss_clip": 0.01205936,
      "auxiliary_loss_mlp": 0.01030265,
      "balance_loss_clip": 1.0611248,
      "balance_loss_mlp": 1.02133584,
      "epoch": 0.3155173450369747,
      "flos": 25593714766080.0,
      "grad_norm": 2.7221207354673918,
      "language_loss": 0.81095088,
      "learning_rate": 3.2042451590479567e-06,
      "loss": 0.83331287,
      "num_input_tokens_seen": 56508940,
      "step": 2624,
      "time_per_iteration": 2.768083333969116
    },
    {
      "auxiliary_loss_clip": 0.01201034,
      "auxiliary_loss_mlp": 0.0102741,
      "balance_loss_clip": 1.06065083,
      "balance_loss_mlp": 1.01891613,
      "epoch": 0.31563758792761376,
      "flos": 24309333175680.0,
      "grad_norm": 1.8028327724919309,
      "language_loss": 0.86670387,
      "learning_rate": 3.203623133910394e-06,
      "loss": 0.88898838,
      "num_input_tokens_seen": 56527245,
      "step": 2625,
      "time_per_iteration": 2.678718328475952
    },
    {
      "auxiliary_loss_clip": 0.01131388,
      "auxiliary_loss_mlp": 0.01031093,
      "balance_loss_clip": 1.05080783,
      "balance_loss_mlp": 1.02159786,
      "epoch": 0.31575783081825287,
      "flos": 31903865550720.0,
      "grad_norm": 2.58081583704734,
      "language_loss": 0.77533019,
      "learning_rate": 3.203000926184158e-06,
      "loss": 0.79695499,
      "num_input_tokens_seen": 56546170,
      "step": 2626,
      "time_per_iteration": 2.862945079803467
    },
    {
      "auxiliary_loss_clip": 0.01202313,
      "auxiliary_loss_mlp": 0.01028562,
      "balance_loss_clip": 1.05982423,
      "balance_loss_mlp": 1.0199728,
      "epoch": 0.315878073708892,
      "flos": 30810960385920.0,
      "grad_norm": 1.8393919812365884,
      "language_loss": 0.77714896,
      "learning_rate": 3.202378535963639e-06,
      "loss": 0.79945773,
      "num_input_tokens_seen": 56567085,
      "step": 2627,
      "time_per_iteration": 2.776393175125122
    },
    {
      "auxiliary_loss_clip": 0.01165065,
      "auxiliary_loss_mlp": 0.00764602,
      "balance_loss_clip": 1.05245519,
      "balance_loss_mlp": 1.00065517,
      "epoch": 0.31599831659953104,
      "flos": 22200264253440.0,
      "grad_norm": 1.6259465237682558,
      "language_loss": 0.84043336,
      "learning_rate": 3.2017559633432516e-06,
      "loss": 0.85973001,
      "num_input_tokens_seen": 56586715,
      "step": 2628,
      "time_per_iteration": 2.685262680053711
    },
    {
      "auxiliary_loss_clip": 0.01181081,
      "auxiliary_loss_mlp": 0.01033742,
      "balance_loss_clip": 1.05447388,
      "balance_loss_mlp": 1.02475965,
      "epoch": 0.31611855949017015,
      "flos": 25593463370880.0,
      "grad_norm": 2.0238212245823064,
      "language_loss": 0.66577661,
      "learning_rate": 3.2011332084174398e-06,
      "loss": 0.68792486,
      "num_input_tokens_seen": 56607585,
      "step": 2629,
      "time_per_iteration": 2.7650938034057617
    },
    {
      "auxiliary_loss_clip": 0.01186739,
      "auxiliary_loss_mlp": 0.010297,
      "balance_loss_clip": 1.05790079,
      "balance_loss_mlp": 1.02034199,
      "epoch": 0.31623880238080926,
      "flos": 20594087694720.0,
      "grad_norm": 1.613461353861769,
      "language_loss": 0.89079487,
      "learning_rate": 3.2005102712806756e-06,
      "loss": 0.91295922,
      "num_input_tokens_seen": 56626415,
      "step": 2630,
      "time_per_iteration": 2.7306313514709473
    },
    {
      "auxiliary_loss_clip": 0.01193524,
      "auxiliary_loss_mlp": 0.0103111,
      "balance_loss_clip": 1.05725121,
      "balance_loss_mlp": 1.02213991,
      "epoch": 0.3163590452714483,
      "flos": 12784917600000.0,
      "grad_norm": 2.2683170909575585,
      "language_loss": 0.72567487,
      "learning_rate": 3.1998871520274575e-06,
      "loss": 0.74792123,
      "num_input_tokens_seen": 56641750,
      "step": 2631,
      "time_per_iteration": 2.715928792953491
    },
    {
      "auxiliary_loss_clip": 0.01172694,
      "auxiliary_loss_mlp": 0.01029167,
      "balance_loss_clip": 1.0519619,
      "balance_loss_mlp": 1.0199821,
      "epoch": 0.3164792881620874,
      "flos": 23041292273280.0,
      "grad_norm": 2.001691742140243,
      "language_loss": 0.84929496,
      "learning_rate": 3.199263850752312e-06,
      "loss": 0.87131357,
      "num_input_tokens_seen": 56662585,
      "step": 2632,
      "time_per_iteration": 2.6578855514526367
    },
    {
      "auxiliary_loss_clip": 0.01193622,
      "auxiliary_loss_mlp": 0.0102657,
      "balance_loss_clip": 1.05903339,
      "balance_loss_mlp": 1.01808214,
      "epoch": 0.31659953105272653,
      "flos": 18296271780480.0,
      "grad_norm": 2.433490123139996,
      "language_loss": 0.85471421,
      "learning_rate": 3.198640367549795e-06,
      "loss": 0.87691617,
      "num_input_tokens_seen": 56681480,
      "step": 2633,
      "time_per_iteration": 2.713207244873047
    },
    {
      "auxiliary_loss_clip": 0.01190114,
      "auxiliary_loss_mlp": 0.00764106,
      "balance_loss_clip": 1.05664539,
      "balance_loss_mlp": 1.00042272,
      "epoch": 0.3167197739433656,
      "flos": 25703421880320.0,
      "grad_norm": 1.8657111697015272,
      "language_loss": 0.86002499,
      "learning_rate": 3.198016702514487e-06,
      "loss": 0.87956715,
      "num_input_tokens_seen": 56701760,
      "step": 2634,
      "time_per_iteration": 2.7980594635009766
    },
    {
      "auxiliary_loss_clip": 0.01200557,
      "auxiliary_loss_mlp": 0.010322,
      "balance_loss_clip": 1.05760634,
      "balance_loss_mlp": 1.02331328,
      "epoch": 0.3168400168340047,
      "flos": 23546016230400.0,
      "grad_norm": 1.67890672035078,
      "language_loss": 0.84526592,
      "learning_rate": 3.1973928557409972e-06,
      "loss": 0.86759347,
      "num_input_tokens_seen": 56719800,
      "step": 2635,
      "time_per_iteration": 2.7512800693511963
    },
    {
      "auxiliary_loss_clip": 0.01199868,
      "auxiliary_loss_mlp": 0.01037612,
      "balance_loss_clip": 1.05910039,
      "balance_loss_mlp": 1.02904677,
      "epoch": 0.31696025972464376,
      "flos": 28366449327360.0,
      "grad_norm": 2.1437159437840925,
      "language_loss": 0.71276867,
      "learning_rate": 3.1967688273239636e-06,
      "loss": 0.73514354,
      "num_input_tokens_seen": 56739605,
      "step": 2636,
      "time_per_iteration": 2.6691768169403076
    },
    {
      "auxiliary_loss_clip": 0.01159727,
      "auxiliary_loss_mlp": 0.01034068,
      "balance_loss_clip": 1.05498743,
      "balance_loss_mlp": 1.02459121,
      "epoch": 0.31708050261528287,
      "flos": 16399111144320.0,
      "grad_norm": 1.8378356169694257,
      "language_loss": 0.82050645,
      "learning_rate": 3.1961446173580503e-06,
      "loss": 0.84244442,
      "num_input_tokens_seen": 56756545,
      "step": 2637,
      "time_per_iteration": 3.593311309814453
    },
    {
      "auxiliary_loss_clip": 0.0117176,
      "auxiliary_loss_mlp": 0.01036931,
      "balance_loss_clip": 1.05584383,
      "balance_loss_mlp": 1.0280385,
      "epoch": 0.317200745505922,
      "flos": 26212347728640.0,
      "grad_norm": 2.062776611139064,
      "language_loss": 0.77575248,
      "learning_rate": 3.1955202259379502e-06,
      "loss": 0.79783946,
      "num_input_tokens_seen": 56778275,
      "step": 2638,
      "time_per_iteration": 2.723254919052124
    },
    {
      "auxiliary_loss_clip": 0.01182808,
      "auxiliary_loss_mlp": 0.01034042,
      "balance_loss_clip": 1.05293953,
      "balance_loss_mlp": 1.02448726,
      "epoch": 0.31732098839656103,
      "flos": 31350876693120.0,
      "grad_norm": 1.891802392218053,
      "language_loss": 0.83003306,
      "learning_rate": 3.194895653158381e-06,
      "loss": 0.85220158,
      "num_input_tokens_seen": 56797215,
      "step": 2639,
      "time_per_iteration": 2.6900076866149902
    },
    {
      "auxiliary_loss_clip": 0.0111398,
      "auxiliary_loss_mlp": 0.01002136,
      "balance_loss_clip": 1.03354549,
      "balance_loss_mlp": 1.00027585,
      "epoch": 0.31744123128720014,
      "flos": 58989024835200.0,
      "grad_norm": 0.8622434202157263,
      "language_loss": 0.55475438,
      "learning_rate": 3.194270899114093e-06,
      "loss": 0.57591552,
      "num_input_tokens_seen": 56863010,
      "step": 2640,
      "time_per_iteration": 3.2613706588745117
    },
    {
      "auxiliary_loss_clip": 0.011959,
      "auxiliary_loss_mlp": 0.01031353,
      "balance_loss_clip": 1.0584954,
      "balance_loss_mlp": 1.02252555,
      "epoch": 0.31756147417783925,
      "flos": 17417573372160.0,
      "grad_norm": 1.797841984987475,
      "language_loss": 0.82110029,
      "learning_rate": 3.193645963899858e-06,
      "loss": 0.84337282,
      "num_input_tokens_seen": 56880625,
      "step": 2641,
      "time_per_iteration": 3.5762710571289062
    },
    {
      "auxiliary_loss_clip": 0.01168057,
      "auxiliary_loss_mlp": 0.01034129,
      "balance_loss_clip": 1.05321193,
      "balance_loss_mlp": 1.02446091,
      "epoch": 0.3176817170684783,
      "flos": 25481673267840.0,
      "grad_norm": 1.827788289540545,
      "language_loss": 0.84107763,
      "learning_rate": 3.193020847610479e-06,
      "loss": 0.86309946,
      "num_input_tokens_seen": 56900945,
      "step": 2642,
      "time_per_iteration": 2.7009689807891846
    },
    {
      "auxiliary_loss_clip": 0.01169682,
      "auxiliary_loss_mlp": 0.01035692,
      "balance_loss_clip": 1.05655122,
      "balance_loss_mlp": 1.02607799,
      "epoch": 0.3178019599591174,
      "flos": 24972603765120.0,
      "grad_norm": 2.245734990135557,
      "language_loss": 0.71723342,
      "learning_rate": 3.192395550340787e-06,
      "loss": 0.73928714,
      "num_input_tokens_seen": 56918895,
      "step": 2643,
      "time_per_iteration": 2.790253162384033
    },
    {
      "auxiliary_loss_clip": 0.01187082,
      "auxiliary_loss_mlp": 0.01036156,
      "balance_loss_clip": 1.05635858,
      "balance_loss_mlp": 1.02673841,
      "epoch": 0.31792220284975653,
      "flos": 12422220019200.0,
      "grad_norm": 3.155109229749805,
      "language_loss": 0.76569188,
      "learning_rate": 3.191770072185638e-06,
      "loss": 0.78792423,
      "num_input_tokens_seen": 56935890,
      "step": 2644,
      "time_per_iteration": 3.598496675491333
    },
    {
      "auxiliary_loss_clip": 0.01188082,
      "auxiliary_loss_mlp": 0.01036123,
      "balance_loss_clip": 1.05918217,
      "balance_loss_mlp": 1.02711725,
      "epoch": 0.3180424457403956,
      "flos": 15485759089920.0,
      "grad_norm": 2.560303057579813,
      "language_loss": 0.72711205,
      "learning_rate": 3.191144413239916e-06,
      "loss": 0.74935412,
      "num_input_tokens_seen": 56952460,
      "step": 2645,
      "time_per_iteration": 3.5060436725616455
    },
    {
      "auxiliary_loss_clip": 0.0117469,
      "auxiliary_loss_mlp": 0.01024358,
      "balance_loss_clip": 1.05627441,
      "balance_loss_mlp": 1.01501167,
      "epoch": 0.3181626886310347,
      "flos": 26174964648960.0,
      "grad_norm": 2.062135016092122,
      "language_loss": 0.88290763,
      "learning_rate": 3.190518573598534e-06,
      "loss": 0.90489805,
      "num_input_tokens_seen": 56969065,
      "step": 2646,
      "time_per_iteration": 2.693160057067871
    },
    {
      "auxiliary_loss_clip": 0.01164797,
      "auxiliary_loss_mlp": 0.01034565,
      "balance_loss_clip": 1.05076027,
      "balance_loss_mlp": 1.02532065,
      "epoch": 0.3182829315216738,
      "flos": 25483109811840.0,
      "grad_norm": 1.5098846619109556,
      "language_loss": 0.77240378,
      "learning_rate": 3.1898925533564308e-06,
      "loss": 0.79439741,
      "num_input_tokens_seen": 56990535,
      "step": 2647,
      "time_per_iteration": 2.6794095039367676
    },
    {
      "auxiliary_loss_clip": 0.01150265,
      "auxiliary_loss_mlp": 0.01038679,
      "balance_loss_clip": 1.05268478,
      "balance_loss_mlp": 1.02982235,
      "epoch": 0.31840317441231286,
      "flos": 18113701927680.0,
      "grad_norm": 2.040530845000522,
      "language_loss": 0.64331335,
      "learning_rate": 3.1892663526085733e-06,
      "loss": 0.6652028,
      "num_input_tokens_seen": 57008910,
      "step": 2648,
      "time_per_iteration": 2.716061592102051
    },
    {
      "auxiliary_loss_clip": 0.01111013,
      "auxiliary_loss_mlp": 0.01000461,
      "balance_loss_clip": 1.03096509,
      "balance_loss_mlp": 0.99873257,
      "epoch": 0.31852341730295197,
      "flos": 64741948957440.0,
      "grad_norm": 0.7554043591838755,
      "language_loss": 0.56937969,
      "learning_rate": 3.188639971449956e-06,
      "loss": 0.59049439,
      "num_input_tokens_seen": 57074960,
      "step": 2649,
      "time_per_iteration": 3.1449673175811768
    },
    {
      "auxiliary_loss_clip": 0.01206892,
      "auxiliary_loss_mlp": 0.01027615,
      "balance_loss_clip": 1.06129241,
      "balance_loss_mlp": 1.01834083,
      "epoch": 0.318643660193591,
      "flos": 20668135582080.0,
      "grad_norm": 2.160410633016965,
      "language_loss": 0.7220965,
      "learning_rate": 3.1880134099756e-06,
      "loss": 0.74444163,
      "num_input_tokens_seen": 57094595,
      "step": 2650,
      "time_per_iteration": 2.586362838745117
    },
    {
      "auxiliary_loss_clip": 0.01186174,
      "auxiliary_loss_mlp": 0.0103224,
      "balance_loss_clip": 1.05524802,
      "balance_loss_mlp": 1.0226382,
      "epoch": 0.31876390308423014,
      "flos": 26943345411840.0,
      "grad_norm": 2.059900866819686,
      "language_loss": 0.69800842,
      "learning_rate": 3.1873866682805535e-06,
      "loss": 0.72019261,
      "num_input_tokens_seen": 57115290,
      "step": 2651,
      "time_per_iteration": 2.6323041915893555
    },
    {
      "auxiliary_loss_clip": 0.01179122,
      "auxiliary_loss_mlp": 0.01027151,
      "balance_loss_clip": 1.05728364,
      "balance_loss_mlp": 1.01874721,
      "epoch": 0.31888414597486925,
      "flos": 18041916597120.0,
      "grad_norm": 1.8292060605653502,
      "language_loss": 0.88797295,
      "learning_rate": 3.186759746459894e-06,
      "loss": 0.91003573,
      "num_input_tokens_seen": 57134400,
      "step": 2652,
      "time_per_iteration": 2.58427095413208
    },
    {
      "auxiliary_loss_clip": 0.01174249,
      "auxiliary_loss_mlp": 0.01029201,
      "balance_loss_clip": 1.05728281,
      "balance_loss_mlp": 1.02024209,
      "epoch": 0.3190043888655083,
      "flos": 25149319701120.0,
      "grad_norm": 1.9541746741175319,
      "language_loss": 0.79754037,
      "learning_rate": 3.1861326446087246e-06,
      "loss": 0.81957483,
      "num_input_tokens_seen": 57153140,
      "step": 2653,
      "time_per_iteration": 2.6651721000671387
    },
    {
      "auxiliary_loss_clip": 0.01189141,
      "auxiliary_loss_mlp": 0.01030091,
      "balance_loss_clip": 1.05745327,
      "balance_loss_mlp": 1.02125728,
      "epoch": 0.3191246317561474,
      "flos": 22053892331520.0,
      "grad_norm": 2.147429252579355,
      "language_loss": 0.71481299,
      "learning_rate": 3.1855053628221763e-06,
      "loss": 0.73700529,
      "num_input_tokens_seen": 57172395,
      "step": 2654,
      "time_per_iteration": 2.5365352630615234
    },
    {
      "auxiliary_loss_clip": 0.01148972,
      "auxiliary_loss_mlp": 0.01035172,
      "balance_loss_clip": 1.04869509,
      "balance_loss_mlp": 1.02509904,
      "epoch": 0.3192448746467865,
      "flos": 14901815687040.0,
      "grad_norm": 2.717073045207291,
      "language_loss": 0.89930892,
      "learning_rate": 3.184877901195407e-06,
      "loss": 0.92115033,
      "num_input_tokens_seen": 57189090,
      "step": 2655,
      "time_per_iteration": 2.6384241580963135
    },
    {
      "auxiliary_loss_clip": 0.0108329,
      "auxiliary_loss_mlp": 0.01002807,
      "balance_loss_clip": 1.0296762,
      "balance_loss_mlp": 1.00073266,
      "epoch": 0.3193651175374256,
      "flos": 67234832657280.0,
      "grad_norm": 0.7932010337600385,
      "language_loss": 0.62812161,
      "learning_rate": 3.184250259823602e-06,
      "loss": 0.64898258,
      "num_input_tokens_seen": 57251620,
      "step": 2656,
      "time_per_iteration": 3.2411487102508545
    },
    {
      "auxiliary_loss_clip": 0.01157957,
      "auxiliary_loss_mlp": 0.01039284,
      "balance_loss_clip": 1.05211139,
      "balance_loss_mlp": 1.02968168,
      "epoch": 0.3194853604280647,
      "flos": 12233077977600.0,
      "grad_norm": 2.2777250357320233,
      "language_loss": 0.8170957,
      "learning_rate": 3.183622438801974e-06,
      "loss": 0.83906806,
      "num_input_tokens_seen": 57266910,
      "step": 2657,
      "time_per_iteration": 2.7138192653656006
    },
    {
      "auxiliary_loss_clip": 0.01203314,
      "auxiliary_loss_mlp": 0.01028863,
      "balance_loss_clip": 1.05961597,
      "balance_loss_mlp": 1.02042294,
      "epoch": 0.3196056033187038,
      "flos": 14939917038720.0,
      "grad_norm": 2.0808455768533207,
      "language_loss": 0.75539839,
      "learning_rate": 3.1829944382257637e-06,
      "loss": 0.77772015,
      "num_input_tokens_seen": 57285040,
      "step": 2658,
      "time_per_iteration": 2.5761444568634033
    },
    {
      "auxiliary_loss_clip": 0.01184822,
      "auxiliary_loss_mlp": 0.01029782,
      "balance_loss_clip": 1.05807185,
      "balance_loss_mlp": 1.02127647,
      "epoch": 0.31972584620934286,
      "flos": 23768878164480.0,
      "grad_norm": 2.5102832990760833,
      "language_loss": 0.8172366,
      "learning_rate": 3.1823662581902373e-06,
      "loss": 0.83938265,
      "num_input_tokens_seen": 57302725,
      "step": 2659,
      "time_per_iteration": 2.646191120147705
    },
    {
      "auxiliary_loss_clip": 0.01142415,
      "auxiliary_loss_mlp": 0.01037386,
      "balance_loss_clip": 1.04534554,
      "balance_loss_mlp": 1.02790844,
      "epoch": 0.31984608909998197,
      "flos": 21251540280960.0,
      "grad_norm": 2.315722707100912,
      "language_loss": 0.7472682,
      "learning_rate": 3.1817378987906896e-06,
      "loss": 0.76906621,
      "num_input_tokens_seen": 57322230,
      "step": 2660,
      "time_per_iteration": 2.7173209190368652
    },
    {
      "auxiliary_loss_clip": 0.01139208,
      "auxiliary_loss_mlp": 0.01041791,
      "balance_loss_clip": 1.05236459,
      "balance_loss_mlp": 1.03256464,
      "epoch": 0.3199663319906211,
      "flos": 18296235866880.0,
      "grad_norm": 1.856731989896956,
      "language_loss": 0.80013651,
      "learning_rate": 3.181109360122442e-06,
      "loss": 0.8219465,
      "num_input_tokens_seen": 57339820,
      "step": 2661,
      "time_per_iteration": 2.7141568660736084
    },
    {
      "auxiliary_loss_clip": 0.0115575,
      "auxiliary_loss_mlp": 0.01028795,
      "balance_loss_clip": 1.05226803,
      "balance_loss_mlp": 1.01925194,
      "epoch": 0.32008657488126013,
      "flos": 18733627779840.0,
      "grad_norm": 12.717548356599197,
      "language_loss": 0.7830075,
      "learning_rate": 3.1804806422808445e-06,
      "loss": 0.80485296,
      "num_input_tokens_seen": 57356955,
      "step": 2662,
      "time_per_iteration": 2.6475565433502197
    },
    {
      "auxiliary_loss_clip": 0.01160835,
      "auxiliary_loss_mlp": 0.0102929,
      "balance_loss_clip": 1.05161011,
      "balance_loss_mlp": 1.0200932,
      "epoch": 0.32020681777189924,
      "flos": 20595344670720.0,
      "grad_norm": 2.019564904205496,
      "language_loss": 0.72773623,
      "learning_rate": 3.1798517453612714e-06,
      "loss": 0.74963742,
      "num_input_tokens_seen": 57376760,
      "step": 2663,
      "time_per_iteration": 3.678034782409668
    },
    {
      "auxiliary_loss_clip": 0.01188805,
      "auxiliary_loss_mlp": 0.01032099,
      "balance_loss_clip": 1.06200576,
      "balance_loss_mlp": 1.02325964,
      "epoch": 0.32032706066253835,
      "flos": 35261692750080.0,
      "grad_norm": 1.8536858884831489,
      "language_loss": 0.75676847,
      "learning_rate": 3.1792226694591265e-06,
      "loss": 0.77897745,
      "num_input_tokens_seen": 57398145,
      "step": 2664,
      "time_per_iteration": 2.797323703765869
    },
    {
      "auxiliary_loss_clip": 0.01159655,
      "auxiliary_loss_mlp": 0.01035582,
      "balance_loss_clip": 1.05540061,
      "balance_loss_mlp": 1.02728486,
      "epoch": 0.3204473035531774,
      "flos": 15304230731520.0,
      "grad_norm": 2.1999846292850664,
      "language_loss": 0.80655587,
      "learning_rate": 3.178593414669841e-06,
      "loss": 0.82850826,
      "num_input_tokens_seen": 57416730,
      "step": 2665,
      "time_per_iteration": 2.6691126823425293
    },
    {
      "auxiliary_loss_clip": 0.01192261,
      "auxiliary_loss_mlp": 0.01033548,
      "balance_loss_clip": 1.05801356,
      "balance_loss_mlp": 1.02371943,
      "epoch": 0.3205675464438165,
      "flos": 24462564595200.0,
      "grad_norm": 3.6194245220349774,
      "language_loss": 0.70929843,
      "learning_rate": 3.1779639810888707e-06,
      "loss": 0.73155659,
      "num_input_tokens_seen": 57436325,
      "step": 2666,
      "time_per_iteration": 2.6896119117736816
    },
    {
      "auxiliary_loss_clip": 0.0118585,
      "auxiliary_loss_mlp": 0.01035621,
      "balance_loss_clip": 1.05905342,
      "balance_loss_mlp": 1.0267514,
      "epoch": 0.3206877893344556,
      "flos": 22456235548800.0,
      "grad_norm": 4.444255100585475,
      "language_loss": 0.75947434,
      "learning_rate": 3.1773343688117013e-06,
      "loss": 0.78168905,
      "num_input_tokens_seen": 57457235,
      "step": 2667,
      "time_per_iteration": 3.671142816543579
    },
    {
      "auxiliary_loss_clip": 0.01175563,
      "auxiliary_loss_mlp": 0.00764016,
      "balance_loss_clip": 1.05527055,
      "balance_loss_mlp": 1.00067687,
      "epoch": 0.3208080322250947,
      "flos": 20412236113920.0,
      "grad_norm": 2.4619134730490573,
      "language_loss": 0.84092939,
      "learning_rate": 3.1767045779338445e-06,
      "loss": 0.86032522,
      "num_input_tokens_seen": 57474895,
      "step": 2668,
      "time_per_iteration": 2.654910087585449
    },
    {
      "auxiliary_loss_clip": 0.01185924,
      "auxiliary_loss_mlp": 0.01032518,
      "balance_loss_clip": 1.05610991,
      "balance_loss_mlp": 1.02367282,
      "epoch": 0.3209282751157338,
      "flos": 21762118154880.0,
      "grad_norm": 2.6101778224097516,
      "language_loss": 0.92002618,
      "learning_rate": 3.176074608550839e-06,
      "loss": 0.94221056,
      "num_input_tokens_seen": 57490715,
      "step": 2669,
      "time_per_iteration": 2.6926400661468506
    },
    {
      "auxiliary_loss_clip": 0.01127689,
      "auxiliary_loss_mlp": 0.01028566,
      "balance_loss_clip": 1.04982281,
      "balance_loss_mlp": 1.01944637,
      "epoch": 0.32104851800637285,
      "flos": 22055041566720.0,
      "grad_norm": 4.526927705808214,
      "language_loss": 0.82451153,
      "learning_rate": 3.17544446075825e-06,
      "loss": 0.8460741,
      "num_input_tokens_seen": 57509880,
      "step": 2670,
      "time_per_iteration": 3.7198309898376465
    },
    {
      "auxiliary_loss_clip": 0.01178198,
      "auxiliary_loss_mlp": 0.01040229,
      "balance_loss_clip": 1.05589473,
      "balance_loss_mlp": 1.03189659,
      "epoch": 0.32116876089701196,
      "flos": 37012301896320.0,
      "grad_norm": 1.6072743879347087,
      "language_loss": 0.71155024,
      "learning_rate": 3.174814134651671e-06,
      "loss": 0.73373449,
      "num_input_tokens_seen": 57532430,
      "step": 2671,
      "time_per_iteration": 3.7673261165618896
    },
    {
      "auxiliary_loss_clip": 0.01200183,
      "auxiliary_loss_mlp": 0.01032068,
      "balance_loss_clip": 1.05831993,
      "balance_loss_mlp": 1.02352631,
      "epoch": 0.3212890037876511,
      "flos": 21979233912960.0,
      "grad_norm": 2.00341738107503,
      "language_loss": 0.80526698,
      "learning_rate": 3.1741836303267215e-06,
      "loss": 0.82758951,
      "num_input_tokens_seen": 57551965,
      "step": 2672,
      "time_per_iteration": 2.668379068374634
    },
    {
      "auxiliary_loss_clip": 0.01200784,
      "auxiliary_loss_mlp": 0.01026278,
      "balance_loss_clip": 1.05927181,
      "balance_loss_mlp": 1.01821971,
      "epoch": 0.32140924667829013,
      "flos": 10342345875840.0,
      "grad_norm": 2.2027818663589813,
      "language_loss": 0.75143903,
      "learning_rate": 3.1735529478790496e-06,
      "loss": 0.77370965,
      "num_input_tokens_seen": 57569955,
      "step": 2673,
      "time_per_iteration": 2.5593314170837402
    },
    {
      "auxiliary_loss_clip": 0.01186477,
      "auxiliary_loss_mlp": 0.01034853,
      "balance_loss_clip": 1.05760396,
      "balance_loss_mlp": 1.02564371,
      "epoch": 0.32152948956892924,
      "flos": 50798910072960.0,
      "grad_norm": 2.6038315450773735,
      "language_loss": 0.79391623,
      "learning_rate": 3.172922087404328e-06,
      "loss": 0.81612945,
      "num_input_tokens_seen": 57592215,
      "step": 2674,
      "time_per_iteration": 3.006711959838867
    },
    {
      "auxiliary_loss_clip": 0.01113378,
      "auxiliary_loss_mlp": 0.01001224,
      "balance_loss_clip": 1.03344941,
      "balance_loss_mlp": 0.9994117,
      "epoch": 0.32164973245956835,
      "flos": 63863250549120.0,
      "grad_norm": 0.7787441855430769,
      "language_loss": 0.55247223,
      "learning_rate": 3.1722910489982586e-06,
      "loss": 0.57361817,
      "num_input_tokens_seen": 57652575,
      "step": 2675,
      "time_per_iteration": 3.2773170471191406
    },
    {
      "auxiliary_loss_clip": 0.01168376,
      "auxiliary_loss_mlp": 0.01027415,
      "balance_loss_clip": 1.0542587,
      "balance_loss_mlp": 1.01905847,
      "epoch": 0.3217699753502074,
      "flos": 23513948363520.0,
      "grad_norm": 1.4558518399401064,
      "language_loss": 0.80182612,
      "learning_rate": 3.1716598327565694e-06,
      "loss": 0.82378399,
      "num_input_tokens_seen": 57672215,
      "step": 2676,
      "time_per_iteration": 2.6673173904418945
    },
    {
      "auxiliary_loss_clip": 0.01200121,
      "auxiliary_loss_mlp": 0.01028462,
      "balance_loss_clip": 1.05835462,
      "balance_loss_mlp": 1.01995599,
      "epoch": 0.3218902182408465,
      "flos": 19062533640960.0,
      "grad_norm": 1.818222936595592,
      "language_loss": 0.84403127,
      "learning_rate": 3.171028438775015e-06,
      "loss": 0.86631709,
      "num_input_tokens_seen": 57691410,
      "step": 2677,
      "time_per_iteration": 2.6804986000061035
    },
    {
      "auxiliary_loss_clip": 0.0119887,
      "auxiliary_loss_mlp": 0.01027673,
      "balance_loss_clip": 1.05770195,
      "balance_loss_mlp": 1.01882827,
      "epoch": 0.3220104611314856,
      "flos": 20375571306240.0,
      "grad_norm": 2.239854653174669,
      "language_loss": 0.84183598,
      "learning_rate": 3.170396867149377e-06,
      "loss": 0.86410141,
      "num_input_tokens_seen": 57709415,
      "step": 2678,
      "time_per_iteration": 2.5963778495788574
    },
    {
      "auxiliary_loss_clip": 0.01137047,
      "auxiliary_loss_mlp": 0.010264,
      "balance_loss_clip": 1.05215406,
      "balance_loss_mlp": 1.01785827,
      "epoch": 0.3221307040221247,
      "flos": 20117014231680.0,
      "grad_norm": 1.8077321639356727,
      "language_loss": 0.86147922,
      "learning_rate": 3.1697651179754653e-06,
      "loss": 0.88311368,
      "num_input_tokens_seen": 57728075,
      "step": 2679,
      "time_per_iteration": 2.7663826942443848
    },
    {
      "auxiliary_loss_clip": 0.01162432,
      "auxiliary_loss_mlp": 0.01034202,
      "balance_loss_clip": 1.05871558,
      "balance_loss_mlp": 1.02604866,
      "epoch": 0.3222509469127638,
      "flos": 23987789602560.0,
      "grad_norm": 2.0939866428061498,
      "language_loss": 0.73339653,
      "learning_rate": 3.1691331913491153e-06,
      "loss": 0.75536287,
      "num_input_tokens_seen": 57750645,
      "step": 2680,
      "time_per_iteration": 2.8074047565460205
    },
    {
      "auxiliary_loss_clip": 0.01199465,
      "auxiliary_loss_mlp": 0.01031313,
      "balance_loss_clip": 1.05671239,
      "balance_loss_mlp": 1.02313566,
      "epoch": 0.32237118980340285,
      "flos": 17675735397120.0,
      "grad_norm": 2.303828002156036,
      "language_loss": 0.85336363,
      "learning_rate": 3.1685010873661898e-06,
      "loss": 0.87567151,
      "num_input_tokens_seen": 57769820,
      "step": 2681,
      "time_per_iteration": 2.620922803878784
    },
    {
      "auxiliary_loss_clip": 0.01181703,
      "auxiliary_loss_mlp": 0.01029544,
      "balance_loss_clip": 1.05534959,
      "balance_loss_mlp": 1.02030575,
      "epoch": 0.32249143269404196,
      "flos": 23147982645120.0,
      "grad_norm": 2.841733961832477,
      "language_loss": 0.7957238,
      "learning_rate": 3.167868806122578e-06,
      "loss": 0.81783628,
      "num_input_tokens_seen": 57788870,
      "step": 2682,
      "time_per_iteration": 2.6248533725738525
    },
    {
      "auxiliary_loss_clip": 0.01175875,
      "auxiliary_loss_mlp": 0.0103261,
      "balance_loss_clip": 1.05551839,
      "balance_loss_mlp": 1.02386642,
      "epoch": 0.32261167558468107,
      "flos": 24422308427520.0,
      "grad_norm": 1.8500413183996878,
      "language_loss": 0.6588155,
      "learning_rate": 3.1672363477141968e-06,
      "loss": 0.68090034,
      "num_input_tokens_seen": 57808165,
      "step": 2683,
      "time_per_iteration": 2.6878409385681152
    },
    {
      "auxiliary_loss_clip": 0.01174659,
      "auxiliary_loss_mlp": 0.01033253,
      "balance_loss_clip": 1.05270362,
      "balance_loss_mlp": 1.02413988,
      "epoch": 0.3227319184753201,
      "flos": 30367175852160.0,
      "grad_norm": 2.385675418284763,
      "language_loss": 0.85025442,
      "learning_rate": 3.1666037122369903e-06,
      "loss": 0.87233359,
      "num_input_tokens_seen": 57828825,
      "step": 2684,
      "time_per_iteration": 2.816071033477783
    },
    {
      "auxiliary_loss_clip": 0.01180412,
      "auxiliary_loss_mlp": 0.01029844,
      "balance_loss_clip": 1.05033648,
      "balance_loss_mlp": 1.02032471,
      "epoch": 0.32285216136595923,
      "flos": 16946174257920.0,
      "grad_norm": 7.804299814259131,
      "language_loss": 0.86498702,
      "learning_rate": 3.165970899786928e-06,
      "loss": 0.88708961,
      "num_input_tokens_seen": 57846740,
      "step": 2685,
      "time_per_iteration": 2.7324421405792236
    },
    {
      "auxiliary_loss_clip": 0.01161657,
      "auxiliary_loss_mlp": 0.01032866,
      "balance_loss_clip": 1.05338311,
      "balance_loss_mlp": 1.02371097,
      "epoch": 0.32297240425659834,
      "flos": 21981532383360.0,
      "grad_norm": 1.9519212969107351,
      "language_loss": 0.75558466,
      "learning_rate": 3.1653379104600067e-06,
      "loss": 0.77752984,
      "num_input_tokens_seen": 57866885,
      "step": 2686,
      "time_per_iteration": 2.6811389923095703
    },
    {
      "auxiliary_loss_clip": 0.01184418,
      "auxiliary_loss_mlp": 0.01035187,
      "balance_loss_clip": 1.05381632,
      "balance_loss_mlp": 1.02635944,
      "epoch": 0.3230926471472374,
      "flos": 22748045639040.0,
      "grad_norm": 1.4794399716933917,
      "language_loss": 0.69333667,
      "learning_rate": 3.164704744352251e-06,
      "loss": 0.71553266,
      "num_input_tokens_seen": 57887690,
      "step": 2687,
      "time_per_iteration": 2.641662836074829
    },
    {
      "auxiliary_loss_clip": 0.01180484,
      "auxiliary_loss_mlp": 0.01028419,
      "balance_loss_clip": 1.05222178,
      "balance_loss_mlp": 1.01950192,
      "epoch": 0.3232128900378765,
      "flos": 16942977947520.0,
      "grad_norm": 2.126783457420904,
      "language_loss": 0.80950356,
      "learning_rate": 3.164071401559713e-06,
      "loss": 0.83159256,
      "num_input_tokens_seen": 57905090,
      "step": 2688,
      "time_per_iteration": 2.581743001937866
    },
    {
      "auxiliary_loss_clip": 0.01172693,
      "auxiliary_loss_mlp": 0.01035582,
      "balance_loss_clip": 1.0538435,
      "balance_loss_mlp": 1.02692151,
      "epoch": 0.3233331329285156,
      "flos": 24023736138240.0,
      "grad_norm": 1.9411633349304152,
      "language_loss": 0.71291256,
      "learning_rate": 3.1634378821784674e-06,
      "loss": 0.73499537,
      "num_input_tokens_seen": 57925305,
      "step": 2689,
      "time_per_iteration": 3.6882474422454834
    },
    {
      "auxiliary_loss_clip": 0.01161486,
      "auxiliary_loss_mlp": 0.01029921,
      "balance_loss_clip": 1.05653632,
      "balance_loss_mlp": 1.02119446,
      "epoch": 0.3234533758191547,
      "flos": 18113845582080.0,
      "grad_norm": 2.181804237186721,
      "language_loss": 0.74275684,
      "learning_rate": 3.1628041863046208e-06,
      "loss": 0.76467097,
      "num_input_tokens_seen": 57942720,
      "step": 2690,
      "time_per_iteration": 2.717665910720825
    },
    {
      "auxiliary_loss_clip": 0.01202664,
      "auxiliary_loss_mlp": 0.01031514,
      "balance_loss_clip": 1.05548096,
      "balance_loss_mlp": 1.02238226,
      "epoch": 0.3235736187097938,
      "flos": 16946138344320.0,
      "grad_norm": 2.1688446784817,
      "language_loss": 0.91246855,
      "learning_rate": 3.162170314034304e-06,
      "loss": 0.93481028,
      "num_input_tokens_seen": 57960135,
      "step": 2691,
      "time_per_iteration": 2.5122272968292236
    },
    {
      "auxiliary_loss_clip": 0.01204652,
      "auxiliary_loss_mlp": 0.01034011,
      "balance_loss_clip": 1.05794799,
      "balance_loss_mlp": 1.02479577,
      "epoch": 0.3236938616004329,
      "flos": 22127150119680.0,
      "grad_norm": 1.7283905481137893,
      "language_loss": 0.809744,
      "learning_rate": 3.1615362654636738e-06,
      "loss": 0.83213067,
      "num_input_tokens_seen": 57980875,
      "step": 2692,
      "time_per_iteration": 2.5773417949676514
    },
    {
      "auxiliary_loss_clip": 0.01152783,
      "auxiliary_loss_mlp": 0.01033291,
      "balance_loss_clip": 1.0547297,
      "balance_loss_mlp": 1.02517891,
      "epoch": 0.32381410449107195,
      "flos": 17164618819200.0,
      "grad_norm": 1.697457137586407,
      "language_loss": 0.86964941,
      "learning_rate": 3.1609020406889163e-06,
      "loss": 0.89151013,
      "num_input_tokens_seen": 57998310,
      "step": 2693,
      "time_per_iteration": 3.5669150352478027
    },
    {
      "auxiliary_loss_clip": 0.01170754,
      "auxiliary_loss_mlp": 0.01034687,
      "balance_loss_clip": 1.0510149,
      "balance_loss_mlp": 1.02587748,
      "epoch": 0.32393434738171106,
      "flos": 16578125550720.0,
      "grad_norm": 1.6582915558559088,
      "language_loss": 0.85000008,
      "learning_rate": 3.1602676398062416e-06,
      "loss": 0.87205446,
      "num_input_tokens_seen": 58017220,
      "step": 2694,
      "time_per_iteration": 2.688220977783203
    },
    {
      "auxiliary_loss_clip": 0.01181408,
      "auxiliary_loss_mlp": 0.01030072,
      "balance_loss_clip": 1.05519795,
      "balance_loss_mlp": 1.02068436,
      "epoch": 0.3240545902723502,
      "flos": 25483612602240.0,
      "grad_norm": 3.607763711615817,
      "language_loss": 0.61276221,
      "learning_rate": 3.1596330629118886e-06,
      "loss": 0.63487703,
      "num_input_tokens_seen": 58037190,
      "step": 2695,
      "time_per_iteration": 2.675760269165039
    },
    {
      "auxiliary_loss_clip": 0.01132187,
      "auxiliary_loss_mlp": 0.01030921,
      "balance_loss_clip": 1.04867327,
      "balance_loss_mlp": 1.0221833,
      "epoch": 0.32417483316298923,
      "flos": 35845851634560.0,
      "grad_norm": 2.080011649110052,
      "language_loss": 0.73202497,
      "learning_rate": 3.1589983101021223e-06,
      "loss": 0.75365609,
      "num_input_tokens_seen": 58055820,
      "step": 2696,
      "time_per_iteration": 3.8391034603118896
    },
    {
      "auxiliary_loss_clip": 0.0117194,
      "auxiliary_loss_mlp": 0.01030873,
      "balance_loss_clip": 1.05369258,
      "balance_loss_mlp": 1.02228403,
      "epoch": 0.32429507605362834,
      "flos": 30080501406720.0,
      "grad_norm": 2.095173365583587,
      "language_loss": 0.84784007,
      "learning_rate": 3.1583633814732337e-06,
      "loss": 0.86986822,
      "num_input_tokens_seen": 58075340,
      "step": 2697,
      "time_per_iteration": 3.6581292152404785
    },
    {
      "auxiliary_loss_clip": 0.01195993,
      "auxiliary_loss_mlp": 0.0103467,
      "balance_loss_clip": 1.0535357,
      "balance_loss_mlp": 1.02596784,
      "epoch": 0.3244153189442674,
      "flos": 18223265387520.0,
      "grad_norm": 2.4280442470545025,
      "language_loss": 0.71742296,
      "learning_rate": 3.157728277121541e-06,
      "loss": 0.73972964,
      "num_input_tokens_seen": 58093515,
      "step": 2698,
      "time_per_iteration": 2.5995421409606934
    },
    {
      "auxiliary_loss_clip": 0.01198776,
      "auxiliary_loss_mlp": 0.0103385,
      "balance_loss_clip": 1.05444634,
      "balance_loss_mlp": 1.02536237,
      "epoch": 0.3245355618349065,
      "flos": 17710317216000.0,
      "grad_norm": 3.4922552122903214,
      "language_loss": 0.78478909,
      "learning_rate": 3.1570929971433897e-06,
      "loss": 0.80711532,
      "num_input_tokens_seen": 58109300,
      "step": 2699,
      "time_per_iteration": 2.5398757457733154
    },
    {
      "auxiliary_loss_clip": 0.0118703,
      "auxiliary_loss_mlp": 0.01035243,
      "balance_loss_clip": 1.05788898,
      "balance_loss_mlp": 1.02569437,
      "epoch": 0.3246558047255456,
      "flos": 23440798316160.0,
      "grad_norm": 2.013188420593872,
      "language_loss": 0.83747292,
      "learning_rate": 3.1564575416351504e-06,
      "loss": 0.85969567,
      "num_input_tokens_seen": 58128000,
      "step": 2700,
      "time_per_iteration": 2.707736015319824
    },
    {
      "auxiliary_loss_clip": 0.01201643,
      "auxiliary_loss_mlp": 0.01028354,
      "balance_loss_clip": 1.05733097,
      "balance_loss_mlp": 1.01959777,
      "epoch": 0.32477604761618467,
      "flos": 21760861178880.0,
      "grad_norm": 2.2926862266450945,
      "language_loss": 0.74221408,
      "learning_rate": 3.155821910693221e-06,
      "loss": 0.76451403,
      "num_input_tokens_seen": 58147415,
      "step": 2701,
      "time_per_iteration": 2.598891258239746
    },
    {
      "auxiliary_loss_clip": 0.01167121,
      "auxiliary_loss_mlp": 0.01031058,
      "balance_loss_clip": 1.05099344,
      "balance_loss_mlp": 1.02143741,
      "epoch": 0.3248962905068238,
      "flos": 19828328624640.0,
      "grad_norm": 1.6585726538962215,
      "language_loss": 0.86138117,
      "learning_rate": 3.1551861044140275e-06,
      "loss": 0.88336301,
      "num_input_tokens_seen": 58167050,
      "step": 2702,
      "time_per_iteration": 2.681370258331299
    },
    {
      "auxiliary_loss_clip": 0.01134887,
      "auxiliary_loss_mlp": 0.01030648,
      "balance_loss_clip": 1.05053306,
      "balance_loss_mlp": 1.02134943,
      "epoch": 0.3250165333974629,
      "flos": 23948215793280.0,
      "grad_norm": 1.7630983921575254,
      "language_loss": 0.77691948,
      "learning_rate": 3.15455012289402e-06,
      "loss": 0.79857481,
      "num_input_tokens_seen": 58186695,
      "step": 2703,
      "time_per_iteration": 2.756605625152588
    },
    {
      "auxiliary_loss_clip": 0.01186021,
      "auxiliary_loss_mlp": 0.01030883,
      "balance_loss_clip": 1.05699563,
      "balance_loss_mlp": 1.02161479,
      "epoch": 0.32513677628810195,
      "flos": 23989333887360.0,
      "grad_norm": 2.384380470252949,
      "language_loss": 0.84684229,
      "learning_rate": 3.153913966229677e-06,
      "loss": 0.86901134,
      "num_input_tokens_seen": 58205815,
      "step": 2704,
      "time_per_iteration": 2.6943352222442627
    },
    {
      "auxiliary_loss_clip": 0.01098558,
      "auxiliary_loss_mlp": 0.01008128,
      "balance_loss_clip": 1.02944577,
      "balance_loss_mlp": 1.00648332,
      "epoch": 0.32525701917874106,
      "flos": 70655790009600.0,
      "grad_norm": 0.644508628970415,
      "language_loss": 0.50211573,
      "learning_rate": 3.1532776345175027e-06,
      "loss": 0.52318257,
      "num_input_tokens_seen": 58270960,
      "step": 2705,
      "time_per_iteration": 3.2086570262908936
    },
    {
      "auxiliary_loss_clip": 0.01196053,
      "auxiliary_loss_mlp": 0.01028983,
      "balance_loss_clip": 1.05486655,
      "balance_loss_mlp": 1.01998878,
      "epoch": 0.32537726206938017,
      "flos": 19682639061120.0,
      "grad_norm": 1.7446376379257678,
      "language_loss": 0.78467262,
      "learning_rate": 3.1526411278540285e-06,
      "loss": 0.80692291,
      "num_input_tokens_seen": 58289390,
      "step": 2706,
      "time_per_iteration": 2.5861656665802
    },
    {
      "auxiliary_loss_clip": 0.01172528,
      "auxiliary_loss_mlp": 0.01030973,
      "balance_loss_clip": 1.04970145,
      "balance_loss_mlp": 1.02164507,
      "epoch": 0.3254975049600192,
      "flos": 28760999293440.0,
      "grad_norm": 2.4325227784662395,
      "language_loss": 0.81630635,
      "learning_rate": 3.1520044463358116e-06,
      "loss": 0.83834136,
      "num_input_tokens_seen": 58306120,
      "step": 2707,
      "time_per_iteration": 2.69480562210083
    },
    {
      "auxiliary_loss_clip": 0.01181407,
      "auxiliary_loss_mlp": 0.01037312,
      "balance_loss_clip": 1.05369127,
      "balance_loss_mlp": 1.0284493,
      "epoch": 0.32561774785065833,
      "flos": 18877378008960.0,
      "grad_norm": 2.6494665956555314,
      "language_loss": 0.80337524,
      "learning_rate": 3.151367590059436e-06,
      "loss": 0.82556248,
      "num_input_tokens_seen": 58324545,
      "step": 2708,
      "time_per_iteration": 2.6525285243988037
    },
    {
      "auxiliary_loss_clip": 0.01201753,
      "auxiliary_loss_mlp": 0.00764278,
      "balance_loss_clip": 1.05856109,
      "balance_loss_mlp": 1.00063896,
      "epoch": 0.32573799074129745,
      "flos": 23112107936640.0,
      "grad_norm": 2.003324126507848,
      "language_loss": 0.86920458,
      "learning_rate": 3.1507305591215117e-06,
      "loss": 0.88886493,
      "num_input_tokens_seen": 58342455,
      "step": 2709,
      "time_per_iteration": 2.647879123687744
    },
    {
      "auxiliary_loss_clip": 0.01095487,
      "auxiliary_loss_mlp": 0.01003741,
      "balance_loss_clip": 1.02736974,
      "balance_loss_mlp": 1.00204849,
      "epoch": 0.3258582336319365,
      "flos": 71237650423680.0,
      "grad_norm": 0.6793714373587663,
      "language_loss": 0.55718565,
      "learning_rate": 3.150093353618677e-06,
      "loss": 0.57817793,
      "num_input_tokens_seen": 58407185,
      "step": 2710,
      "time_per_iteration": 3.2688558101654053
    },
    {
      "auxiliary_loss_clip": 0.01190397,
      "auxiliary_loss_mlp": 0.01032038,
      "balance_loss_clip": 1.05490875,
      "balance_loss_mlp": 1.02254903,
      "epoch": 0.3259784765225756,
      "flos": 22456020067200.0,
      "grad_norm": 2.203156727383354,
      "language_loss": 0.88424277,
      "learning_rate": 3.149455973647596e-06,
      "loss": 0.90646708,
      "num_input_tokens_seen": 58425245,
      "step": 2711,
      "time_per_iteration": 2.727320671081543
    },
    {
      "auxiliary_loss_clip": 0.01145603,
      "auxiliary_loss_mlp": 0.01037936,
      "balance_loss_clip": 1.04603148,
      "balance_loss_mlp": 1.02862036,
      "epoch": 0.32609871941321467,
      "flos": 20484811543680.0,
      "grad_norm": 2.0436688276475294,
      "language_loss": 0.77381182,
      "learning_rate": 3.1488184193049563e-06,
      "loss": 0.79564726,
      "num_input_tokens_seen": 58444780,
      "step": 2712,
      "time_per_iteration": 2.6932926177978516
    },
    {
      "auxiliary_loss_clip": 0.01195645,
      "auxiliary_loss_mlp": 0.01032021,
      "balance_loss_clip": 1.05589318,
      "balance_loss_mlp": 1.02297878,
      "epoch": 0.3262189623038538,
      "flos": 22416805393920.0,
      "grad_norm": 1.927821861631247,
      "language_loss": 0.72152805,
      "learning_rate": 3.1481806906874767e-06,
      "loss": 0.74380469,
      "num_input_tokens_seen": 58466090,
      "step": 2713,
      "time_per_iteration": 2.6816134452819824
    },
    {
      "auxiliary_loss_clip": 0.01199366,
      "auxiliary_loss_mlp": 0.01030634,
      "balance_loss_clip": 1.05724692,
      "balance_loss_mlp": 1.02201486,
      "epoch": 0.3263392051944929,
      "flos": 20923496346240.0,
      "grad_norm": 1.688457753955921,
      "language_loss": 0.87984264,
      "learning_rate": 3.147542787891899e-06,
      "loss": 0.90214264,
      "num_input_tokens_seen": 58485435,
      "step": 2714,
      "time_per_iteration": 2.5901939868927
    },
    {
      "auxiliary_loss_clip": 0.01175617,
      "auxiliary_loss_mlp": 0.01040731,
      "balance_loss_clip": 1.05757809,
      "balance_loss_mlp": 1.03105128,
      "epoch": 0.32645944808513194,
      "flos": 24025172682240.0,
      "grad_norm": 1.7642589594246745,
      "language_loss": 0.75608838,
      "learning_rate": 3.1469047110149926e-06,
      "loss": 0.77825189,
      "num_input_tokens_seen": 58504175,
      "step": 2715,
      "time_per_iteration": 2.7627196311950684
    },
    {
      "auxiliary_loss_clip": 0.01131864,
      "auxiliary_loss_mlp": 0.01040227,
      "balance_loss_clip": 1.04791284,
      "balance_loss_mlp": 1.03082681,
      "epoch": 0.32657969097577105,
      "flos": 21032413361280.0,
      "grad_norm": 2.1669625716632903,
      "language_loss": 0.85252929,
      "learning_rate": 3.146266460153554e-06,
      "loss": 0.87425017,
      "num_input_tokens_seen": 58523885,
      "step": 2716,
      "time_per_iteration": 3.6698122024536133
    },
    {
      "auxiliary_loss_clip": 0.01166124,
      "auxiliary_loss_mlp": 0.00764439,
      "balance_loss_clip": 1.0531404,
      "balance_loss_mlp": 1.00055552,
      "epoch": 0.32669993386641016,
      "flos": 22710267509760.0,
      "grad_norm": 1.6507118938602057,
      "language_loss": 0.80202168,
      "learning_rate": 3.145628035404404e-06,
      "loss": 0.82132733,
      "num_input_tokens_seen": 58543085,
      "step": 2717,
      "time_per_iteration": 2.696194887161255
    },
    {
      "auxiliary_loss_clip": 0.01089147,
      "auxiliary_loss_mlp": 0.0100364,
      "balance_loss_clip": 1.02462018,
      "balance_loss_mlp": 1.00189996,
      "epoch": 0.3268201767570492,
      "flos": 72105718406400.0,
      "grad_norm": 0.8844296341462872,
      "language_loss": 0.57475275,
      "learning_rate": 3.1449894368643922e-06,
      "loss": 0.59568065,
      "num_input_tokens_seen": 58605400,
      "step": 2718,
      "time_per_iteration": 3.3167662620544434
    },
    {
      "auxiliary_loss_clip": 0.01156548,
      "auxiliary_loss_mlp": 0.0103612,
      "balance_loss_clip": 1.05682421,
      "balance_loss_mlp": 1.02787721,
      "epoch": 0.32694041964768833,
      "flos": 24535175938560.0,
      "grad_norm": 1.495595141410126,
      "language_loss": 0.71276188,
      "learning_rate": 3.1443506646303934e-06,
      "loss": 0.73468852,
      "num_input_tokens_seen": 58626700,
      "step": 2719,
      "time_per_iteration": 3.6778078079223633
    },
    {
      "auxiliary_loss_clip": 0.01187519,
      "auxiliary_loss_mlp": 0.01033361,
      "balance_loss_clip": 1.05527353,
      "balance_loss_mlp": 1.02390218,
      "epoch": 0.32706066253832744,
      "flos": 33183003755520.0,
      "grad_norm": 1.8922990074290353,
      "language_loss": 0.66900104,
      "learning_rate": 3.1437117187993086e-06,
      "loss": 0.69120979,
      "num_input_tokens_seen": 58649020,
      "step": 2720,
      "time_per_iteration": 2.716674327850342
    },
    {
      "auxiliary_loss_clip": 0.01146954,
      "auxiliary_loss_mlp": 0.01036626,
      "balance_loss_clip": 1.04799747,
      "balance_loss_mlp": 1.02787006,
      "epoch": 0.3271809054289665,
      "flos": 24061622008320.0,
      "grad_norm": 1.52626288594651,
      "language_loss": 0.80026329,
      "learning_rate": 3.143072599468065e-06,
      "loss": 0.82209909,
      "num_input_tokens_seen": 58668845,
      "step": 2721,
      "time_per_iteration": 2.7395308017730713
    },
    {
      "auxiliary_loss_clip": 0.01169925,
      "auxiliary_loss_mlp": 0.01033294,
      "balance_loss_clip": 1.05724251,
      "balance_loss_mlp": 1.02516353,
      "epoch": 0.3273011483196056,
      "flos": 38253769712640.0,
      "grad_norm": 2.0901700707802555,
      "language_loss": 0.755234,
      "learning_rate": 3.1424333067336174e-06,
      "loss": 0.7772662,
      "num_input_tokens_seen": 58691610,
      "step": 2722,
      "time_per_iteration": 2.767987012863159
    },
    {
      "auxiliary_loss_clip": 0.01188489,
      "auxiliary_loss_mlp": 0.0102626,
      "balance_loss_clip": 1.05452013,
      "balance_loss_mlp": 1.01652098,
      "epoch": 0.3274213912102447,
      "flos": 29054389582080.0,
      "grad_norm": 1.9110965813729102,
      "language_loss": 0.78544033,
      "learning_rate": 3.141793840692945e-06,
      "loss": 0.8075878,
      "num_input_tokens_seen": 58712360,
      "step": 2723,
      "time_per_iteration": 4.443854093551636
    },
    {
      "auxiliary_loss_clip": 0.01159917,
      "auxiliary_loss_mlp": 0.0103846,
      "balance_loss_clip": 1.05030847,
      "balance_loss_mlp": 1.02907276,
      "epoch": 0.32754163410088377,
      "flos": 29133249891840.0,
      "grad_norm": 4.375530108118556,
      "language_loss": 0.61836398,
      "learning_rate": 3.1411542014430553e-06,
      "loss": 0.64034784,
      "num_input_tokens_seen": 58733440,
      "step": 2724,
      "time_per_iteration": 2.7210636138916016
    },
    {
      "auxiliary_loss_clip": 0.01154698,
      "auxiliary_loss_mlp": 0.01030267,
      "balance_loss_clip": 1.04926348,
      "balance_loss_mlp": 1.02236366,
      "epoch": 0.3276618769915229,
      "flos": 20631075724800.0,
      "grad_norm": 1.675444218359621,
      "language_loss": 0.82086873,
      "learning_rate": 3.1405143890809804e-06,
      "loss": 0.84271836,
      "num_input_tokens_seen": 58752735,
      "step": 2725,
      "time_per_iteration": 2.692605972290039
    },
    {
      "auxiliary_loss_clip": 0.01165002,
      "auxiliary_loss_mlp": 0.0102873,
      "balance_loss_clip": 1.05159974,
      "balance_loss_mlp": 1.02024841,
      "epoch": 0.327782119882162,
      "flos": 18657425076480.0,
      "grad_norm": 1.7586110953937648,
      "language_loss": 0.7004565,
      "learning_rate": 3.1398744037037796e-06,
      "loss": 0.72239381,
      "num_input_tokens_seen": 58772070,
      "step": 2726,
      "time_per_iteration": 2.655078411102295
    },
    {
      "auxiliary_loss_clip": 0.01168012,
      "auxiliary_loss_mlp": 0.0103649,
      "balance_loss_clip": 1.05345798,
      "balance_loss_mlp": 1.02818704,
      "epoch": 0.32790236277280105,
      "flos": 21795802133760.0,
      "grad_norm": 2.432855516289774,
      "language_loss": 0.84516215,
      "learning_rate": 3.139234245408538e-06,
      "loss": 0.86720717,
      "num_input_tokens_seen": 58790950,
      "step": 2727,
      "time_per_iteration": 2.662137269973755
    },
    {
      "auxiliary_loss_clip": 0.01153588,
      "auxiliary_loss_mlp": 0.00763653,
      "balance_loss_clip": 1.05129695,
      "balance_loss_mlp": 1.00061703,
      "epoch": 0.32802260566344016,
      "flos": 23331414424320.0,
      "grad_norm": 1.8418717717494817,
      "language_loss": 0.7591095,
      "learning_rate": 3.1385939142923666e-06,
      "loss": 0.77828193,
      "num_input_tokens_seen": 58813340,
      "step": 2728,
      "time_per_iteration": 2.773582696914673
    },
    {
      "auxiliary_loss_clip": 0.01169,
      "auxiliary_loss_mlp": 0.01034719,
      "balance_loss_clip": 1.05198228,
      "balance_loss_mlp": 1.02545059,
      "epoch": 0.3281428485540792,
      "flos": 24206988349440.0,
      "grad_norm": 2.3015369566123205,
      "language_loss": 0.7830205,
      "learning_rate": 3.137953410452405e-06,
      "loss": 0.80505764,
      "num_input_tokens_seen": 58833610,
      "step": 2729,
      "time_per_iteration": 2.651838779449463
    },
    {
      "auxiliary_loss_clip": 0.01166122,
      "auxiliary_loss_mlp": 0.01026874,
      "balance_loss_clip": 1.05188966,
      "balance_loss_mlp": 1.01823723,
      "epoch": 0.3282630914447183,
      "flos": 34128962380800.0,
      "grad_norm": 1.6071313154171916,
      "language_loss": 0.7462821,
      "learning_rate": 3.1373127339858146e-06,
      "loss": 0.76821208,
      "num_input_tokens_seen": 58856210,
      "step": 2730,
      "time_per_iteration": 2.838601589202881
    },
    {
      "auxiliary_loss_clip": 0.01148767,
      "auxiliary_loss_mlp": 0.01024677,
      "balance_loss_clip": 1.05100334,
      "balance_loss_mlp": 1.01673782,
      "epoch": 0.32838333433535744,
      "flos": 27600726170880.0,
      "grad_norm": 1.8124446777107657,
      "language_loss": 0.74855691,
      "learning_rate": 3.136671884989787e-06,
      "loss": 0.77029139,
      "num_input_tokens_seen": 58876120,
      "step": 2731,
      "time_per_iteration": 2.731778621673584
    },
    {
      "auxiliary_loss_clip": 0.01127252,
      "auxiliary_loss_mlp": 0.01034107,
      "balance_loss_clip": 1.04796124,
      "balance_loss_mlp": 1.02526712,
      "epoch": 0.3285035772259965,
      "flos": 12349500935040.0,
      "grad_norm": 3.6125201833346297,
      "language_loss": 0.87894964,
      "learning_rate": 3.1360308635615383e-06,
      "loss": 0.90056324,
      "num_input_tokens_seen": 58894660,
      "step": 2732,
      "time_per_iteration": 2.7847747802734375
    },
    {
      "auxiliary_loss_clip": 0.01177303,
      "auxiliary_loss_mlp": 0.01027616,
      "balance_loss_clip": 1.05559576,
      "balance_loss_mlp": 1.01837111,
      "epoch": 0.3286238201166356,
      "flos": 24316084932480.0,
      "grad_norm": 2.5663987903993926,
      "language_loss": 0.78839916,
      "learning_rate": 3.135389669798311e-06,
      "loss": 0.81044841,
      "num_input_tokens_seen": 58912720,
      "step": 2733,
      "time_per_iteration": 2.718369245529175
    },
    {
      "auxiliary_loss_clip": 0.01180951,
      "auxiliary_loss_mlp": 0.00763938,
      "balance_loss_clip": 1.05279458,
      "balance_loss_mlp": 1.00051045,
      "epoch": 0.3287440630072747,
      "flos": 21392812471680.0,
      "grad_norm": 2.1203648417051997,
      "language_loss": 0.7999928,
      "learning_rate": 3.134748303797373e-06,
      "loss": 0.81944174,
      "num_input_tokens_seen": 58930090,
      "step": 2734,
      "time_per_iteration": 2.622119665145874
    },
    {
      "auxiliary_loss_clip": 0.01137341,
      "auxiliary_loss_mlp": 0.01033878,
      "balance_loss_clip": 1.04753184,
      "balance_loss_mlp": 1.02419257,
      "epoch": 0.32886430589791377,
      "flos": 23732536579200.0,
      "grad_norm": 2.3062577776158757,
      "language_loss": 0.8126601,
      "learning_rate": 3.1341067656560203e-06,
      "loss": 0.83437228,
      "num_input_tokens_seen": 58947935,
      "step": 2735,
      "time_per_iteration": 2.83725643157959
    },
    {
      "auxiliary_loss_clip": 0.011772,
      "auxiliary_loss_mlp": 0.01025924,
      "balance_loss_clip": 1.05431223,
      "balance_loss_mlp": 1.01684642,
      "epoch": 0.3289845487885529,
      "flos": 22418708814720.0,
      "grad_norm": 1.9174676875169832,
      "language_loss": 0.8644948,
      "learning_rate": 3.133465055471572e-06,
      "loss": 0.88652605,
      "num_input_tokens_seen": 58967720,
      "step": 2736,
      "time_per_iteration": 2.7002885341644287
    },
    {
      "auxiliary_loss_clip": 0.01145983,
      "auxiliary_loss_mlp": 0.01034743,
      "balance_loss_clip": 1.05050159,
      "balance_loss_mlp": 1.02574217,
      "epoch": 0.329104791679192,
      "flos": 19682603147520.0,
      "grad_norm": 2.4803616734978022,
      "language_loss": 0.66790497,
      "learning_rate": 3.1328231733413767e-06,
      "loss": 0.68971217,
      "num_input_tokens_seen": 58984360,
      "step": 2737,
      "time_per_iteration": 2.7366583347320557
    },
    {
      "auxiliary_loss_clip": 0.01177782,
      "auxiliary_loss_mlp": 0.01029465,
      "balance_loss_clip": 1.05436945,
      "balance_loss_mlp": 1.02074456,
      "epoch": 0.32922503456983104,
      "flos": 15997234803840.0,
      "grad_norm": 2.0592519836628282,
      "language_loss": 0.91126168,
      "learning_rate": 3.1321811193628067e-06,
      "loss": 0.93333417,
      "num_input_tokens_seen": 59002505,
      "step": 2738,
      "time_per_iteration": 2.6712002754211426
    },
    {
      "auxiliary_loss_clip": 0.0118352,
      "auxiliary_loss_mlp": 0.00764426,
      "balance_loss_clip": 1.05516744,
      "balance_loss_mlp": 1.00059998,
      "epoch": 0.32934527746047015,
      "flos": 26834069260800.0,
      "grad_norm": 1.8715024411965862,
      "language_loss": 0.69649929,
      "learning_rate": 3.131538893633261e-06,
      "loss": 0.71597874,
      "num_input_tokens_seen": 59022065,
      "step": 2739,
      "time_per_iteration": 2.8487672805786133
    },
    {
      "auxiliary_loss_clip": 0.01199617,
      "auxiliary_loss_mlp": 0.01027976,
      "balance_loss_clip": 1.05857217,
      "balance_loss_mlp": 1.01952386,
      "epoch": 0.32946552035110926,
      "flos": 23403774372480.0,
      "grad_norm": 2.310796069720394,
      "language_loss": 0.77843308,
      "learning_rate": 3.130896496250165e-06,
      "loss": 0.80070901,
      "num_input_tokens_seen": 59041890,
      "step": 2740,
      "time_per_iteration": 2.709935426712036
    },
    {
      "auxiliary_loss_clip": 0.01196597,
      "auxiliary_loss_mlp": 0.0102801,
      "balance_loss_clip": 1.05300164,
      "balance_loss_mlp": 1.019135,
      "epoch": 0.3295857632417483,
      "flos": 14172470029440.0,
      "grad_norm": 1.9443878705393807,
      "language_loss": 0.87073892,
      "learning_rate": 3.1302539273109693e-06,
      "loss": 0.89298493,
      "num_input_tokens_seen": 59058715,
      "step": 2741,
      "time_per_iteration": 3.503798007965088
    },
    {
      "auxiliary_loss_clip": 0.01163464,
      "auxiliary_loss_mlp": 0.01032671,
      "balance_loss_clip": 1.05428672,
      "balance_loss_mlp": 1.02360559,
      "epoch": 0.32970600613238743,
      "flos": 22196708807040.0,
      "grad_norm": 1.6440839861621774,
      "language_loss": 0.80638808,
      "learning_rate": 3.1296111869131513e-06,
      "loss": 0.82834941,
      "num_input_tokens_seen": 59076140,
      "step": 2742,
      "time_per_iteration": 2.716278553009033
    },
    {
      "auxiliary_loss_clip": 0.01195244,
      "auxiliary_loss_mlp": 0.01034529,
      "balance_loss_clip": 1.05435705,
      "balance_loss_mlp": 1.02601159,
      "epoch": 0.32982624902302654,
      "flos": 22053784590720.0,
      "grad_norm": 1.924245613651228,
      "language_loss": 0.86040592,
      "learning_rate": 3.1289682751542153e-06,
      "loss": 0.88270366,
      "num_input_tokens_seen": 59095700,
      "step": 2743,
      "time_per_iteration": 2.594625234603882
    },
    {
      "auxiliary_loss_clip": 0.01177995,
      "auxiliary_loss_mlp": 0.01028528,
      "balance_loss_clip": 1.05233264,
      "balance_loss_mlp": 1.019665,
      "epoch": 0.3299464919136656,
      "flos": 18661626967680.0,
      "grad_norm": 1.9313328465259096,
      "language_loss": 0.71468514,
      "learning_rate": 3.1283251921316883e-06,
      "loss": 0.73675036,
      "num_input_tokens_seen": 59113445,
      "step": 2744,
      "time_per_iteration": 2.6850879192352295
    },
    {
      "auxiliary_loss_clip": 0.01138701,
      "auxiliary_loss_mlp": 0.01030974,
      "balance_loss_clip": 1.05246329,
      "balance_loss_mlp": 1.02320099,
      "epoch": 0.3300667348043047,
      "flos": 13407357404160.0,
      "grad_norm": 3.517469270049778,
      "language_loss": 0.80781817,
      "learning_rate": 3.1276819379431277e-06,
      "loss": 0.82951498,
      "num_input_tokens_seen": 59131535,
      "step": 2745,
      "time_per_iteration": 3.630502223968506
    },
    {
      "auxiliary_loss_clip": 0.01175063,
      "auxiliary_loss_mlp": 0.00764098,
      "balance_loss_clip": 1.0537591,
      "balance_loss_mlp": 1.00050354,
      "epoch": 0.33018697769494376,
      "flos": 15742556398080.0,
      "grad_norm": 2.0751916934723162,
      "language_loss": 0.75446367,
      "learning_rate": 3.1270385126861134e-06,
      "loss": 0.77385527,
      "num_input_tokens_seen": 59149520,
      "step": 2746,
      "time_per_iteration": 2.673215389251709
    },
    {
      "auxiliary_loss_clip": 0.01195992,
      "auxiliary_loss_mlp": 0.01029203,
      "balance_loss_clip": 1.0543673,
      "balance_loss_mlp": 1.02067971,
      "epoch": 0.3303072205855829,
      "flos": 18258601392000.0,
      "grad_norm": 1.8595928738847969,
      "language_loss": 0.81959462,
      "learning_rate": 3.1263949164582533e-06,
      "loss": 0.84184659,
      "num_input_tokens_seen": 59169170,
      "step": 2747,
      "time_per_iteration": 2.5732316970825195
    },
    {
      "auxiliary_loss_clip": 0.01197703,
      "auxiliary_loss_mlp": 0.01028638,
      "balance_loss_clip": 1.05394101,
      "balance_loss_mlp": 1.02072811,
      "epoch": 0.330427463476222,
      "flos": 17749424148480.0,
      "grad_norm": 3.980768436195556,
      "language_loss": 0.78436404,
      "learning_rate": 3.1257511493571797e-06,
      "loss": 0.80662745,
      "num_input_tokens_seen": 59187675,
      "step": 2748,
      "time_per_iteration": 3.5830931663513184
    },
    {
      "auxiliary_loss_clip": 0.01156475,
      "auxiliary_loss_mlp": 0.01031397,
      "balance_loss_clip": 1.05486333,
      "balance_loss_mlp": 1.02307653,
      "epoch": 0.33054770636686104,
      "flos": 27162580072320.0,
      "grad_norm": 2.4615579324648094,
      "language_loss": 0.78758657,
      "learning_rate": 3.125107211480552e-06,
      "loss": 0.80946529,
      "num_input_tokens_seen": 59207610,
      "step": 2749,
      "time_per_iteration": 2.7574968338012695
    },
    {
      "auxiliary_loss_clip": 0.01122088,
      "auxiliary_loss_mlp": 0.01030103,
      "balance_loss_clip": 1.04756391,
      "balance_loss_mlp": 1.02170157,
      "epoch": 0.33066794925750015,
      "flos": 20117193799680.0,
      "grad_norm": 1.660941808985637,
      "language_loss": 0.7993902,
      "learning_rate": 3.124463102926054e-06,
      "loss": 0.82091212,
      "num_input_tokens_seen": 59226945,
      "step": 2750,
      "time_per_iteration": 3.693782329559326
    },
    {
      "auxiliary_loss_clip": 0.0108432,
      "auxiliary_loss_mlp": 0.01003989,
      "balance_loss_clip": 1.02398694,
      "balance_loss_mlp": 1.0021174,
      "epoch": 0.33078819214813926,
      "flos": 70642609718400.0,
      "grad_norm": 0.8006265573960069,
      "language_loss": 0.61651969,
      "learning_rate": 3.1238188237913984e-06,
      "loss": 0.63740277,
      "num_input_tokens_seen": 59291485,
      "step": 2751,
      "time_per_iteration": 3.2915022373199463
    },
    {
      "auxiliary_loss_clip": 0.01205073,
      "auxiliary_loss_mlp": 0.01035457,
      "balance_loss_clip": 1.059147,
      "balance_loss_mlp": 1.0262605,
      "epoch": 0.3309084350387783,
      "flos": 21141940907520.0,
      "grad_norm": 3.467555231418141,
      "language_loss": 0.76673847,
      "learning_rate": 3.1231743741743202e-06,
      "loss": 0.7891438,
      "num_input_tokens_seen": 59310990,
      "step": 2752,
      "time_per_iteration": 2.6316752433776855
    },
    {
      "auxiliary_loss_clip": 0.01177779,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 1.05091429,
      "balance_loss_mlp": 1.01798749,
      "epoch": 0.3310286779294174,
      "flos": 14209350318720.0,
      "grad_norm": 3.5509819221546177,
      "language_loss": 0.84183729,
      "learning_rate": 3.122529754172582e-06,
      "loss": 0.86388612,
      "num_input_tokens_seen": 59327875,
      "step": 2753,
      "time_per_iteration": 2.5843453407287598
    },
    {
      "auxiliary_loss_clip": 0.01183911,
      "auxiliary_loss_mlp": 0.01027278,
      "balance_loss_clip": 1.05586922,
      "balance_loss_mlp": 1.01858187,
      "epoch": 0.33114892082005654,
      "flos": 20778130005120.0,
      "grad_norm": 1.8537745980482834,
      "language_loss": 0.72648108,
      "learning_rate": 3.1218849638839736e-06,
      "loss": 0.74859297,
      "num_input_tokens_seen": 59347135,
      "step": 2754,
      "time_per_iteration": 2.6683666706085205
    },
    {
      "auxiliary_loss_clip": 0.01140969,
      "auxiliary_loss_mlp": 0.01035068,
      "balance_loss_clip": 1.04611528,
      "balance_loss_mlp": 1.02560902,
      "epoch": 0.3312691637106956,
      "flos": 17090750499840.0,
      "grad_norm": 2.4195691490461813,
      "language_loss": 0.78467965,
      "learning_rate": 3.121240003406307e-06,
      "loss": 0.80644006,
      "num_input_tokens_seen": 59365985,
      "step": 2755,
      "time_per_iteration": 2.6800169944763184
    },
    {
      "auxiliary_loss_clip": 0.01160748,
      "auxiliary_loss_mlp": 0.01040033,
      "balance_loss_clip": 1.05449092,
      "balance_loss_mlp": 1.03066301,
      "epoch": 0.3313894066013347,
      "flos": 29456230008960.0,
      "grad_norm": 2.187078467345408,
      "language_loss": 0.72534245,
      "learning_rate": 3.120594872837425e-06,
      "loss": 0.74735034,
      "num_input_tokens_seen": 59384655,
      "step": 2756,
      "time_per_iteration": 2.8027520179748535
    },
    {
      "auxiliary_loss_clip": 0.01088092,
      "auxiliary_loss_mlp": 0.00755158,
      "balance_loss_clip": 1.02406764,
      "balance_loss_mlp": 1.00024736,
      "epoch": 0.3315096494919738,
      "flos": 61419242280960.0,
      "grad_norm": 0.8292190739309188,
      "language_loss": 0.62370116,
      "learning_rate": 3.1199495722751906e-06,
      "loss": 0.64213377,
      "num_input_tokens_seen": 59444185,
      "step": 2757,
      "time_per_iteration": 3.2961905002593994
    },
    {
      "auxiliary_loss_clip": 0.01145187,
      "auxiliary_loss_mlp": 0.01032607,
      "balance_loss_clip": 1.05067515,
      "balance_loss_mlp": 1.02375603,
      "epoch": 0.33162989238261287,
      "flos": 21653057485440.0,
      "grad_norm": 2.175579658264054,
      "language_loss": 0.84069932,
      "learning_rate": 3.1193041018174972e-06,
      "loss": 0.86247718,
      "num_input_tokens_seen": 59464900,
      "step": 2758,
      "time_per_iteration": 2.7551767826080322
    },
    {
      "auxiliary_loss_clip": 0.01189524,
      "auxiliary_loss_mlp": 0.01029348,
      "balance_loss_clip": 1.05880404,
      "balance_loss_mlp": 1.0201565,
      "epoch": 0.331750135273252,
      "flos": 22674787850880.0,
      "grad_norm": 3.8892618632284224,
      "language_loss": 0.94621348,
      "learning_rate": 3.118658461562261e-06,
      "loss": 0.96840221,
      "num_input_tokens_seen": 59481000,
      "step": 2759,
      "time_per_iteration": 2.6369640827178955
    },
    {
      "auxiliary_loss_clip": 0.0116993,
      "auxiliary_loss_mlp": 0.01033705,
      "balance_loss_clip": 1.05734086,
      "balance_loss_mlp": 1.0245738,
      "epoch": 0.33187037816389103,
      "flos": 22746896403840.0,
      "grad_norm": 1.4501376393381444,
      "language_loss": 0.84905493,
      "learning_rate": 3.118012651607426e-06,
      "loss": 0.87109131,
      "num_input_tokens_seen": 59502605,
      "step": 2760,
      "time_per_iteration": 2.7259433269500732
    },
    {
      "auxiliary_loss_clip": 0.01201352,
      "auxiliary_loss_mlp": 0.01040284,
      "balance_loss_clip": 1.05892813,
      "balance_loss_mlp": 1.03139138,
      "epoch": 0.33199062105453014,
      "flos": 19203769918080.0,
      "grad_norm": 2.1611612267010822,
      "language_loss": 0.83954895,
      "learning_rate": 3.1173666720509603e-06,
      "loss": 0.86196536,
      "num_input_tokens_seen": 59519540,
      "step": 2761,
      "time_per_iteration": 2.7091403007507324
    },
    {
      "auxiliary_loss_clip": 0.01174641,
      "auxiliary_loss_mlp": 0.01028537,
      "balance_loss_clip": 1.05453396,
      "balance_loss_mlp": 1.02000809,
      "epoch": 0.33211086394516925,
      "flos": 31577006764800.0,
      "grad_norm": 1.6927443908533388,
      "language_loss": 0.68455118,
      "learning_rate": 3.116720522990859e-06,
      "loss": 0.7065829,
      "num_input_tokens_seen": 59540415,
      "step": 2762,
      "time_per_iteration": 2.7502403259277344
    },
    {
      "auxiliary_loss_clip": 0.01127735,
      "auxiliary_loss_mlp": 0.01039812,
      "balance_loss_clip": 1.04798675,
      "balance_loss_mlp": 1.03077543,
      "epoch": 0.3322311068358083,
      "flos": 17932496791680.0,
      "grad_norm": 2.871445007361222,
      "language_loss": 0.61708784,
      "learning_rate": 3.116074204525142e-06,
      "loss": 0.63876331,
      "num_input_tokens_seen": 59558590,
      "step": 2763,
      "time_per_iteration": 2.807497501373291
    },
    {
      "auxiliary_loss_clip": 0.01175602,
      "auxiliary_loss_mlp": 0.01032169,
      "balance_loss_clip": 1.05485713,
      "balance_loss_mlp": 1.02225637,
      "epoch": 0.3323513497264474,
      "flos": 32269831269120.0,
      "grad_norm": 1.5450952812937975,
      "language_loss": 0.83677655,
      "learning_rate": 3.1154277167518553e-06,
      "loss": 0.85885429,
      "num_input_tokens_seen": 59580205,
      "step": 2764,
      "time_per_iteration": 2.7140774726867676
    },
    {
      "auxiliary_loss_clip": 0.01069754,
      "auxiliary_loss_mlp": 0.0100307,
      "balance_loss_clip": 1.02085829,
      "balance_loss_mlp": 1.00135314,
      "epoch": 0.33247159261708653,
      "flos": 52668674588160.0,
      "grad_norm": 0.7875501090503614,
      "language_loss": 0.59536761,
      "learning_rate": 3.114781059769072e-06,
      "loss": 0.61609584,
      "num_input_tokens_seen": 59631530,
      "step": 2765,
      "time_per_iteration": 3.17437481880188
    },
    {
      "auxiliary_loss_clip": 0.01168423,
      "auxiliary_loss_mlp": 0.01031339,
      "balance_loss_clip": 1.05361676,
      "balance_loss_mlp": 1.02169538,
      "epoch": 0.3325918355077256,
      "flos": 27125232906240.0,
      "grad_norm": 2.705560606974291,
      "language_loss": 0.67095315,
      "learning_rate": 3.1141342336748874e-06,
      "loss": 0.69295079,
      "num_input_tokens_seen": 59651090,
      "step": 2766,
      "time_per_iteration": 2.717733144760132
    },
    {
      "auxiliary_loss_clip": 0.01180643,
      "auxiliary_loss_mlp": 0.01029118,
      "balance_loss_clip": 1.05492651,
      "balance_loss_mlp": 1.01990902,
      "epoch": 0.3327120783983647,
      "flos": 23664414435840.0,
      "grad_norm": 1.7772123307067857,
      "language_loss": 0.82220244,
      "learning_rate": 3.1134872385674253e-06,
      "loss": 0.84430003,
      "num_input_tokens_seen": 59675245,
      "step": 2767,
      "time_per_iteration": 3.589491605758667
    },
    {
      "auxiliary_loss_clip": 0.01169875,
      "auxiliary_loss_mlp": 0.01033348,
      "balance_loss_clip": 1.05099547,
      "balance_loss_mlp": 1.02409136,
      "epoch": 0.3328323212890038,
      "flos": 19171378828800.0,
      "grad_norm": 1.7526796924187293,
      "language_loss": 0.85739338,
      "learning_rate": 3.1128400745448353e-06,
      "loss": 0.87942559,
      "num_input_tokens_seen": 59694625,
      "step": 2768,
      "time_per_iteration": 2.6876275539398193
    },
    {
      "auxiliary_loss_clip": 0.01188025,
      "auxiliary_loss_mlp": 0.0103445,
      "balance_loss_clip": 1.05657089,
      "balance_loss_mlp": 1.02541947,
      "epoch": 0.33295256417964286,
      "flos": 37706347463040.0,
      "grad_norm": 2.052329576057974,
      "language_loss": 0.62879753,
      "learning_rate": 3.11219274170529e-06,
      "loss": 0.65102232,
      "num_input_tokens_seen": 59716435,
      "step": 2769,
      "time_per_iteration": 2.8205909729003906
    },
    {
      "auxiliary_loss_clip": 0.0116519,
      "auxiliary_loss_mlp": 0.01036768,
      "balance_loss_clip": 1.05313504,
      "balance_loss_mlp": 1.02814937,
      "epoch": 0.333072807070282,
      "flos": 26505989412480.0,
      "grad_norm": 2.02557401494615,
      "language_loss": 0.81650162,
      "learning_rate": 3.1115452401469903e-06,
      "loss": 0.83852124,
      "num_input_tokens_seen": 59736835,
      "step": 2770,
      "time_per_iteration": 2.7227792739868164
    },
    {
      "auxiliary_loss_clip": 0.01131542,
      "auxiliary_loss_mlp": 0.0103758,
      "balance_loss_clip": 1.0474422,
      "balance_loss_mlp": 1.02844262,
      "epoch": 0.3331930499609211,
      "flos": 21430913823360.0,
      "grad_norm": 1.9302578542474742,
      "language_loss": 0.86741459,
      "learning_rate": 3.1108975699681613e-06,
      "loss": 0.8891058,
      "num_input_tokens_seen": 59754230,
      "step": 2771,
      "time_per_iteration": 3.7085483074188232
    },
    {
      "auxiliary_loss_clip": 0.01153837,
      "auxiliary_loss_mlp": 0.01028927,
      "balance_loss_clip": 1.05171514,
      "balance_loss_mlp": 1.02066028,
      "epoch": 0.33331329285156014,
      "flos": 20659947281280.0,
      "grad_norm": 2.1672825587528064,
      "language_loss": 0.71801674,
      "learning_rate": 3.1102497312670542e-06,
      "loss": 0.73984438,
      "num_input_tokens_seen": 59772235,
      "step": 2772,
      "time_per_iteration": 2.729571580886841
    },
    {
      "auxiliary_loss_clip": 0.0115754,
      "auxiliary_loss_mlp": 0.01029777,
      "balance_loss_clip": 1.05178428,
      "balance_loss_mlp": 1.02131295,
      "epoch": 0.33343353574219925,
      "flos": 28001596930560.0,
      "grad_norm": 1.9510537949633502,
      "language_loss": 0.80806005,
      "learning_rate": 3.109601724141946e-06,
      "loss": 0.82993329,
      "num_input_tokens_seen": 59791230,
      "step": 2773,
      "time_per_iteration": 2.782825469970703
    },
    {
      "auxiliary_loss_clip": 0.01167232,
      "auxiliary_loss_mlp": 0.0104062,
      "balance_loss_clip": 1.0544939,
      "balance_loss_mlp": 1.0312264,
      "epoch": 0.33355377863283836,
      "flos": 23764963582080.0,
      "grad_norm": 1.6834253650595827,
      "language_loss": 0.68229687,
      "learning_rate": 3.108953548691138e-06,
      "loss": 0.70437539,
      "num_input_tokens_seen": 59811315,
      "step": 2774,
      "time_per_iteration": 3.6489365100860596
    },
    {
      "auxiliary_loss_clip": 0.01199203,
      "auxiliary_loss_mlp": 0.01034402,
      "balance_loss_clip": 1.05658519,
      "balance_loss_mlp": 1.02480602,
      "epoch": 0.3336740215234774,
      "flos": 37779677078400.0,
      "grad_norm": 2.4904774893621227,
      "language_loss": 0.72381419,
      "learning_rate": 3.108305205012959e-06,
      "loss": 0.74615026,
      "num_input_tokens_seen": 59832010,
      "step": 2775,
      "time_per_iteration": 3.71881103515625
    },
    {
      "auxiliary_loss_clip": 0.01172031,
      "auxiliary_loss_mlp": 0.01028271,
      "balance_loss_clip": 1.05593038,
      "balance_loss_mlp": 1.0196166,
      "epoch": 0.3337942644141165,
      "flos": 25519056347520.0,
      "grad_norm": 2.065119224142056,
      "language_loss": 0.87441444,
      "learning_rate": 3.107656693205761e-06,
      "loss": 0.89641738,
      "num_input_tokens_seen": 59851450,
      "step": 2776,
      "time_per_iteration": 2.7264277935028076
    },
    {
      "auxiliary_loss_clip": 0.01204706,
      "auxiliary_loss_mlp": 0.01035209,
      "balance_loss_clip": 1.0582993,
      "balance_loss_mlp": 1.0246892,
      "epoch": 0.3339145073047556,
      "flos": 25989844930560.0,
      "grad_norm": 2.7657690928032226,
      "language_loss": 0.70830971,
      "learning_rate": 3.107008013367924e-06,
      "loss": 0.7307089,
      "num_input_tokens_seen": 59870245,
      "step": 2777,
      "time_per_iteration": 2.5891430377960205
    },
    {
      "auxiliary_loss_clip": 0.01154088,
      "auxiliary_loss_mlp": 0.01038114,
      "balance_loss_clip": 1.05188537,
      "balance_loss_mlp": 1.02866042,
      "epoch": 0.3340347501953947,
      "flos": 19062569554560.0,
      "grad_norm": 4.63623091206627,
      "language_loss": 0.86902034,
      "learning_rate": 3.1063591655978507e-06,
      "loss": 0.89094239,
      "num_input_tokens_seen": 59886195,
      "step": 2778,
      "time_per_iteration": 2.652740716934204
    },
    {
      "auxiliary_loss_clip": 0.01129116,
      "auxiliary_loss_mlp": 0.01035561,
      "balance_loss_clip": 1.0468179,
      "balance_loss_mlp": 1.02694845,
      "epoch": 0.3341549930860338,
      "flos": 18109715518080.0,
      "grad_norm": 1.7381894403158913,
      "language_loss": 0.79720557,
      "learning_rate": 3.105710149993972e-06,
      "loss": 0.81885231,
      "num_input_tokens_seen": 59905525,
      "step": 2779,
      "time_per_iteration": 2.808835506439209
    },
    {
      "auxiliary_loss_clip": 0.01203642,
      "auxiliary_loss_mlp": 0.01033911,
      "balance_loss_clip": 1.06002271,
      "balance_loss_mlp": 1.02495885,
      "epoch": 0.33427523597667286,
      "flos": 22674967418880.0,
      "grad_norm": 1.941720006706963,
      "language_loss": 0.8541255,
      "learning_rate": 3.1050609666547427e-06,
      "loss": 0.87650096,
      "num_input_tokens_seen": 59925085,
      "step": 2780,
      "time_per_iteration": 2.6581199169158936
    },
    {
      "auxiliary_loss_clip": 0.01161051,
      "auxiliary_loss_mlp": 0.01034235,
      "balance_loss_clip": 1.0537504,
      "balance_loss_mlp": 1.02484727,
      "epoch": 0.33439547886731197,
      "flos": 22638338524800.0,
      "grad_norm": 5.283062925689047,
      "language_loss": 0.77756256,
      "learning_rate": 3.104411615678644e-06,
      "loss": 0.79951537,
      "num_input_tokens_seen": 59943935,
      "step": 2781,
      "time_per_iteration": 2.7014265060424805
    },
    {
      "auxiliary_loss_clip": 0.01169294,
      "auxiliary_loss_mlp": 0.01034572,
      "balance_loss_clip": 1.05559826,
      "balance_loss_mlp": 1.02478504,
      "epoch": 0.3345157217579511,
      "flos": 24096383395200.0,
      "grad_norm": 2.3102890478465645,
      "language_loss": 0.73055589,
      "learning_rate": 3.1037620971641803e-06,
      "loss": 0.75259459,
      "num_input_tokens_seen": 59963725,
      "step": 2782,
      "time_per_iteration": 2.751354932785034
    },
    {
      "auxiliary_loss_clip": 0.01200128,
      "auxiliary_loss_mlp": 0.0103606,
      "balance_loss_clip": 1.05684745,
      "balance_loss_mlp": 1.02649939,
      "epoch": 0.33463596464859013,
      "flos": 18989491334400.0,
      "grad_norm": 3.1767676337035335,
      "language_loss": 0.64895451,
      "learning_rate": 3.1031124112098844e-06,
      "loss": 0.67131639,
      "num_input_tokens_seen": 59981935,
      "step": 2783,
      "time_per_iteration": 2.656076431274414
    },
    {
      "auxiliary_loss_clip": 0.01175575,
      "auxiliary_loss_mlp": 0.01032925,
      "balance_loss_clip": 1.05662513,
      "balance_loss_mlp": 1.02416897,
      "epoch": 0.33475620753922924,
      "flos": 20375607219840.0,
      "grad_norm": 2.0334236063998334,
      "language_loss": 0.72513419,
      "learning_rate": 3.1024625579143127e-06,
      "loss": 0.7472192,
      "num_input_tokens_seen": 59999455,
      "step": 2784,
      "time_per_iteration": 2.6542627811431885
    },
    {
      "auxiliary_loss_clip": 0.01198904,
      "auxiliary_loss_mlp": 0.01029854,
      "balance_loss_clip": 1.05575025,
      "balance_loss_mlp": 1.02143741,
      "epoch": 0.33487645042986836,
      "flos": 18182578256640.0,
      "grad_norm": 1.8852435844452802,
      "language_loss": 0.73233676,
      "learning_rate": 3.101812537376048e-06,
      "loss": 0.75462437,
      "num_input_tokens_seen": 60018475,
      "step": 2785,
      "time_per_iteration": 2.5882985591888428
    },
    {
      "auxiliary_loss_clip": 0.01161797,
      "auxiliary_loss_mlp": 0.00764082,
      "balance_loss_clip": 1.05156612,
      "balance_loss_mlp": 1.00050426,
      "epoch": 0.3349966933205074,
      "flos": 25848824135040.0,
      "grad_norm": 2.8201991452249797,
      "language_loss": 0.84739023,
      "learning_rate": 3.1011623496936973e-06,
      "loss": 0.86664903,
      "num_input_tokens_seen": 60036770,
      "step": 2786,
      "time_per_iteration": 2.7940471172332764
    },
    {
      "auxiliary_loss_clip": 0.01200298,
      "auxiliary_loss_mlp": 0.01028357,
      "balance_loss_clip": 1.05871451,
      "balance_loss_mlp": 1.02017367,
      "epoch": 0.3351169362111465,
      "flos": 28111447699200.0,
      "grad_norm": 2.6818820332907642,
      "language_loss": 0.69569206,
      "learning_rate": 3.100511994965893e-06,
      "loss": 0.7179786,
      "num_input_tokens_seen": 60056725,
      "step": 2787,
      "time_per_iteration": 2.6679646968841553
    },
    {
      "auxiliary_loss_clip": 0.01179438,
      "auxiliary_loss_mlp": 0.0103053,
      "balance_loss_clip": 1.05507398,
      "balance_loss_mlp": 1.02158928,
      "epoch": 0.33523717910178563,
      "flos": 22673315393280.0,
      "grad_norm": 1.7229060595479473,
      "language_loss": 0.84943473,
      "learning_rate": 3.0998614732912947e-06,
      "loss": 0.87153435,
      "num_input_tokens_seen": 60076100,
      "step": 2788,
      "time_per_iteration": 2.615370035171509
    },
    {
      "auxiliary_loss_clip": 0.01183583,
      "auxiliary_loss_mlp": 0.01032914,
      "balance_loss_clip": 1.05715358,
      "balance_loss_mlp": 1.02472472,
      "epoch": 0.3353574219924247,
      "flos": 15669801400320.0,
      "grad_norm": 10.149686133009592,
      "language_loss": 0.68092316,
      "learning_rate": 3.0992107847685855e-06,
      "loss": 0.70308816,
      "num_input_tokens_seen": 60093815,
      "step": 2789,
      "time_per_iteration": 2.6418635845184326
    },
    {
      "auxiliary_loss_clip": 0.01168481,
      "auxiliary_loss_mlp": 0.01032548,
      "balance_loss_clip": 1.05576241,
      "balance_loss_mlp": 1.02333903,
      "epoch": 0.3354776648830638,
      "flos": 24790644443520.0,
      "grad_norm": 1.6665968633384247,
      "language_loss": 0.79348218,
      "learning_rate": 3.0985599294964736e-06,
      "loss": 0.81549251,
      "num_input_tokens_seen": 60113370,
      "step": 2790,
      "time_per_iteration": 2.6592540740966797
    },
    {
      "auxiliary_loss_clip": 0.01167466,
      "auxiliary_loss_mlp": 0.01033749,
      "balance_loss_clip": 1.05383074,
      "balance_loss_mlp": 1.02484417,
      "epoch": 0.33559790777370285,
      "flos": 28694852398080.0,
      "grad_norm": 1.8818440899557454,
      "language_loss": 0.69890761,
      "learning_rate": 3.097908907573695e-06,
      "loss": 0.72091979,
      "num_input_tokens_seen": 60131350,
      "step": 2791,
      "time_per_iteration": 2.7778708934783936
    },
    {
      "auxiliary_loss_clip": 0.01125732,
      "auxiliary_loss_mlp": 0.01034308,
      "balance_loss_clip": 1.05034316,
      "balance_loss_mlp": 1.0258621,
      "epoch": 0.33571815066434196,
      "flos": 22235779825920.0,
      "grad_norm": 2.018703092970654,
      "language_loss": 0.89691931,
      "learning_rate": 3.0972577190990067e-06,
      "loss": 0.91851968,
      "num_input_tokens_seen": 60149830,
      "step": 2792,
      "time_per_iteration": 2.7718048095703125
    },
    {
      "auxiliary_loss_clip": 0.01160297,
      "auxiliary_loss_mlp": 0.0103237,
      "balance_loss_clip": 1.05341375,
      "balance_loss_mlp": 1.02325034,
      "epoch": 0.3358383935549811,
      "flos": 23842279607040.0,
      "grad_norm": 1.8826570461259735,
      "language_loss": 0.79732108,
      "learning_rate": 3.096606364171196e-06,
      "loss": 0.81924772,
      "num_input_tokens_seen": 60169620,
      "step": 2793,
      "time_per_iteration": 3.680549144744873
    },
    {
      "auxiliary_loss_clip": 0.01141044,
      "auxiliary_loss_mlp": 0.01033499,
      "balance_loss_clip": 1.05004573,
      "balance_loss_mlp": 1.02421284,
      "epoch": 0.33595863644562013,
      "flos": 22267308988800.0,
      "grad_norm": 3.3402076041702533,
      "language_loss": 0.85195374,
      "learning_rate": 3.0959548428890703e-06,
      "loss": 0.87369919,
      "num_input_tokens_seen": 60188490,
      "step": 2794,
      "time_per_iteration": 2.7962357997894287
    },
    {
      "auxiliary_loss_clip": 0.01180993,
      "auxiliary_loss_mlp": 0.01033383,
      "balance_loss_clip": 1.05635762,
      "balance_loss_mlp": 1.02410913,
      "epoch": 0.33607887933625924,
      "flos": 20119779578880.0,
      "grad_norm": 1.7203195887057026,
      "language_loss": 0.84222066,
      "learning_rate": 3.095303155351468e-06,
      "loss": 0.86436445,
      "num_input_tokens_seen": 60208695,
      "step": 2795,
      "time_per_iteration": 2.669407844543457
    },
    {
      "auxiliary_loss_clip": 0.01126729,
      "auxiliary_loss_mlp": 0.01029857,
      "balance_loss_clip": 1.04736042,
      "balance_loss_mlp": 1.0203141,
      "epoch": 0.33619912222689835,
      "flos": 19318109886720.0,
      "grad_norm": 2.1545803532969856,
      "language_loss": 0.79919744,
      "learning_rate": 3.0946513016572464e-06,
      "loss": 0.82076329,
      "num_input_tokens_seen": 60227600,
      "step": 2796,
      "time_per_iteration": 2.760286808013916
    },
    {
      "auxiliary_loss_clip": 0.01186132,
      "auxiliary_loss_mlp": 0.01030419,
      "balance_loss_clip": 1.05476964,
      "balance_loss_mlp": 1.02086473,
      "epoch": 0.3363193651175374,
      "flos": 16800664262400.0,
      "grad_norm": 2.131567966234348,
      "language_loss": 0.76982653,
      "learning_rate": 3.0939992819052938e-06,
      "loss": 0.79199207,
      "num_input_tokens_seen": 60245110,
      "step": 2797,
      "time_per_iteration": 3.5698530673980713
    },
    {
      "auxiliary_loss_clip": 0.01168158,
      "auxiliary_loss_mlp": 0.01029347,
      "balance_loss_clip": 1.05310917,
      "balance_loss_mlp": 1.02114892,
      "epoch": 0.3364396080081765,
      "flos": 23550289948800.0,
      "grad_norm": 2.2927965669825863,
      "language_loss": 0.81525993,
      "learning_rate": 3.0933470961945193e-06,
      "loss": 0.83723497,
      "num_input_tokens_seen": 60263405,
      "step": 2798,
      "time_per_iteration": 2.7090811729431152
    },
    {
      "auxiliary_loss_clip": 0.01163864,
      "auxiliary_loss_mlp": 0.01035787,
      "balance_loss_clip": 1.05369282,
      "balance_loss_mlp": 1.02712035,
      "epoch": 0.3365598508988156,
      "flos": 28037902602240.0,
      "grad_norm": 1.7733210904358991,
      "language_loss": 0.68200386,
      "learning_rate": 3.0926947446238597e-06,
      "loss": 0.70400029,
      "num_input_tokens_seen": 60282975,
      "step": 2799,
      "time_per_iteration": 2.7108585834503174
    },
    {
      "auxiliary_loss_clip": 0.01185605,
      "auxiliary_loss_mlp": 0.01031278,
      "balance_loss_clip": 1.05149758,
      "balance_loss_mlp": 1.02121675,
      "epoch": 0.3366800937894547,
      "flos": 16982767238400.0,
      "grad_norm": 2.9296072510530333,
      "language_loss": 0.83034962,
      "learning_rate": 3.092042227292276e-06,
      "loss": 0.85251844,
      "num_input_tokens_seen": 60299810,
      "step": 2800,
      "time_per_iteration": 2.6267004013061523
    },
    {
      "auxiliary_loss_clip": 0.01196862,
      "auxiliary_loss_mlp": 0.01036726,
      "balance_loss_clip": 1.05706692,
      "balance_loss_mlp": 1.02839363,
      "epoch": 0.3368003366800938,
      "flos": 23915321913600.0,
      "grad_norm": 1.632235398624199,
      "language_loss": 0.87999731,
      "learning_rate": 3.0913895442987557e-06,
      "loss": 0.90233314,
      "num_input_tokens_seen": 60320775,
      "step": 2801,
      "time_per_iteration": 3.571031332015991
    },
    {
      "auxiliary_loss_clip": 0.01152625,
      "auxiliary_loss_mlp": 0.00764079,
      "balance_loss_clip": 1.05284381,
      "balance_loss_mlp": 1.00045705,
      "epoch": 0.3369205795707329,
      "flos": 24791219061120.0,
      "grad_norm": 1.5711717708205655,
      "language_loss": 0.86081803,
      "learning_rate": 3.090736695742308e-06,
      "loss": 0.87998509,
      "num_input_tokens_seen": 60341905,
      "step": 2802,
      "time_per_iteration": 2.742711305618286
    },
    {
      "auxiliary_loss_clip": 0.01132032,
      "auxiliary_loss_mlp": 0.01032355,
      "balance_loss_clip": 1.04797518,
      "balance_loss_mlp": 1.02370656,
      "epoch": 0.33704082246137196,
      "flos": 17931096161280.0,
      "grad_norm": 2.801381425337414,
      "language_loss": 0.52326983,
      "learning_rate": 3.0900836817219713e-06,
      "loss": 0.54491365,
      "num_input_tokens_seen": 60358335,
      "step": 2803,
      "time_per_iteration": 2.674513578414917
    },
    {
      "auxiliary_loss_clip": 0.01196115,
      "auxiliary_loss_mlp": 0.01033845,
      "balance_loss_clip": 1.05483282,
      "balance_loss_mlp": 1.02525616,
      "epoch": 0.33716106535201107,
      "flos": 21286517149440.0,
      "grad_norm": 1.7423530096992423,
      "language_loss": 0.84192306,
      "learning_rate": 3.089430502336807e-06,
      "loss": 0.86422265,
      "num_input_tokens_seen": 60378305,
      "step": 2804,
      "time_per_iteration": 2.6149492263793945
    },
    {
      "auxiliary_loss_clip": 0.01186031,
      "auxiliary_loss_mlp": 0.01034604,
      "balance_loss_clip": 1.05531621,
      "balance_loss_mlp": 1.02553844,
      "epoch": 0.3372813082426502,
      "flos": 18402962152320.0,
      "grad_norm": 2.5983167148425697,
      "language_loss": 0.90893143,
      "learning_rate": 3.088777157685902e-06,
      "loss": 0.9311378,
      "num_input_tokens_seen": 60393895,
      "step": 2805,
      "time_per_iteration": 2.672654390335083
    },
    {
      "auxiliary_loss_clip": 0.01166044,
      "auxiliary_loss_mlp": 0.01032113,
      "balance_loss_clip": 1.05452132,
      "balance_loss_mlp": 1.02346396,
      "epoch": 0.33740155113328923,
      "flos": 17201391367680.0,
      "grad_norm": 2.335731490446569,
      "language_loss": 0.85586655,
      "learning_rate": 3.088123647868367e-06,
      "loss": 0.87784815,
      "num_input_tokens_seen": 60410445,
      "step": 2806,
      "time_per_iteration": 2.664973020553589
    },
    {
      "auxiliary_loss_clip": 0.01184481,
      "auxiliary_loss_mlp": 0.01036169,
      "balance_loss_clip": 1.05087185,
      "balance_loss_mlp": 1.0270853,
      "epoch": 0.33752179402392835,
      "flos": 29058950609280.0,
      "grad_norm": 2.258607861398835,
      "language_loss": 0.81345415,
      "learning_rate": 3.0874699729833405e-06,
      "loss": 0.8356607,
      "num_input_tokens_seen": 60431815,
      "step": 2807,
      "time_per_iteration": 2.749363660812378
    },
    {
      "auxiliary_loss_clip": 0.01168162,
      "auxiliary_loss_mlp": 0.01032276,
      "balance_loss_clip": 1.05549347,
      "balance_loss_mlp": 1.02338922,
      "epoch": 0.3376420369145674,
      "flos": 25080730680960.0,
      "grad_norm": 2.125469663323752,
      "language_loss": 0.80090201,
      "learning_rate": 3.086816133129983e-06,
      "loss": 0.82290637,
      "num_input_tokens_seen": 60452075,
      "step": 2808,
      "time_per_iteration": 2.6814825534820557
    },
    {
      "auxiliary_loss_clip": 0.01200755,
      "auxiliary_loss_mlp": 0.01030892,
      "balance_loss_clip": 1.0599879,
      "balance_loss_mlp": 1.02217793,
      "epoch": 0.3377622798052065,
      "flos": 27490624007040.0,
      "grad_norm": 2.401757926194325,
      "language_loss": 0.76078475,
      "learning_rate": 3.0861621284074826e-06,
      "loss": 0.7831012,
      "num_input_tokens_seen": 60472600,
      "step": 2809,
      "time_per_iteration": 2.7200918197631836
    },
    {
      "auxiliary_loss_clip": 0.01177321,
      "auxiliary_loss_mlp": 0.01035206,
      "balance_loss_clip": 1.05557168,
      "balance_loss_mlp": 1.0262537,
      "epoch": 0.3378825226958456,
      "flos": 21975211589760.0,
      "grad_norm": 1.6809913030309178,
      "language_loss": 0.73073161,
      "learning_rate": 3.085507958915051e-06,
      "loss": 0.75285697,
      "num_input_tokens_seen": 60491030,
      "step": 2810,
      "time_per_iteration": 2.654655694961548
    },
    {
      "auxiliary_loss_clip": 0.01167784,
      "auxiliary_loss_mlp": 0.01036366,
      "balance_loss_clip": 1.05541229,
      "balance_loss_mlp": 1.0270555,
      "epoch": 0.3380027655864847,
      "flos": 42523189200000.0,
      "grad_norm": 2.8119749370231264,
      "language_loss": 0.71059239,
      "learning_rate": 3.084853624751925e-06,
      "loss": 0.73263383,
      "num_input_tokens_seen": 60512615,
      "step": 2811,
      "time_per_iteration": 2.8271470069885254
    },
    {
      "auxiliary_loss_clip": 0.01158712,
      "auxiliary_loss_mlp": 0.01030993,
      "balance_loss_clip": 1.05478907,
      "balance_loss_mlp": 1.02218342,
      "epoch": 0.3381230084771238,
      "flos": 26725080418560.0,
      "grad_norm": 1.7703771495871892,
      "language_loss": 0.85612559,
      "learning_rate": 3.0841991260173668e-06,
      "loss": 0.87802255,
      "num_input_tokens_seen": 60532520,
      "step": 2812,
      "time_per_iteration": 2.786619186401367
    },
    {
      "auxiliary_loss_clip": 0.01202808,
      "auxiliary_loss_mlp": 0.01030176,
      "balance_loss_clip": 1.05889559,
      "balance_loss_mlp": 1.0209254,
      "epoch": 0.3382432513677629,
      "flos": 22710375250560.0,
      "grad_norm": 1.9078769962642366,
      "language_loss": 0.80593419,
      "learning_rate": 3.0835444628106634e-06,
      "loss": 0.82826406,
      "num_input_tokens_seen": 60551500,
      "step": 2813,
      "time_per_iteration": 2.5884058475494385
    },
    {
      "auxiliary_loss_clip": 0.01201542,
      "auxiliary_loss_mlp": 0.00765014,
      "balance_loss_clip": 1.05907917,
      "balance_loss_mlp": 1.00044179,
      "epoch": 0.33836349425840195,
      "flos": 22122409524480.0,
      "grad_norm": 1.9742119693899398,
      "language_loss": 0.83493108,
      "learning_rate": 3.082889635231126e-06,
      "loss": 0.85459661,
      "num_input_tokens_seen": 60570160,
      "step": 2814,
      "time_per_iteration": 2.631774425506592
    },
    {
      "auxiliary_loss_clip": 0.01171936,
      "auxiliary_loss_mlp": 0.01025619,
      "balance_loss_clip": 1.0526545,
      "balance_loss_mlp": 1.01632071,
      "epoch": 0.33848373714904106,
      "flos": 27308090067840.0,
      "grad_norm": 2.3043014587678794,
      "language_loss": 0.76489413,
      "learning_rate": 3.0822346433780925e-06,
      "loss": 0.78686965,
      "num_input_tokens_seen": 60590885,
      "step": 2815,
      "time_per_iteration": 2.713453531265259
    },
    {
      "auxiliary_loss_clip": 0.01184235,
      "auxiliary_loss_mlp": 0.01035135,
      "balance_loss_clip": 1.0524199,
      "balance_loss_mlp": 1.0252763,
      "epoch": 0.3386039800396802,
      "flos": 25848716394240.0,
      "grad_norm": 1.9372993705655075,
      "language_loss": 0.87278295,
      "learning_rate": 3.0815794873509237e-06,
      "loss": 0.89497662,
      "num_input_tokens_seen": 60609170,
      "step": 2816,
      "time_per_iteration": 2.6761701107025146
    },
    {
      "auxiliary_loss_clip": 0.01195595,
      "auxiliary_loss_mlp": 0.01026055,
      "balance_loss_clip": 1.05471861,
      "balance_loss_mlp": 1.01703095,
      "epoch": 0.33872422293031923,
      "flos": 18880646146560.0,
      "grad_norm": 1.7315712451747685,
      "language_loss": 0.72755194,
      "learning_rate": 3.0809241672490066e-06,
      "loss": 0.7497685,
      "num_input_tokens_seen": 60627340,
      "step": 2817,
      "time_per_iteration": 2.617640495300293
    },
    {
      "auxiliary_loss_clip": 0.01170895,
      "auxiliary_loss_mlp": 0.01033799,
      "balance_loss_clip": 1.05466354,
      "balance_loss_mlp": 1.02583599,
      "epoch": 0.33884446582095834,
      "flos": 23146977064320.0,
      "grad_norm": 4.629190821051471,
      "language_loss": 0.85146689,
      "learning_rate": 3.080268683171753e-06,
      "loss": 0.87351382,
      "num_input_tokens_seen": 60647630,
      "step": 2818,
      "time_per_iteration": 2.7083539962768555
    },
    {
      "auxiliary_loss_clip": 0.01182077,
      "auxiliary_loss_mlp": 0.01028056,
      "balance_loss_clip": 1.05195808,
      "balance_loss_mlp": 1.01958668,
      "epoch": 0.33896470871159745,
      "flos": 15997342544640.0,
      "grad_norm": 2.115155520875265,
      "language_loss": 0.89091986,
      "learning_rate": 3.0796130352185985e-06,
      "loss": 0.91302121,
      "num_input_tokens_seen": 60664485,
      "step": 2819,
      "time_per_iteration": 3.6213529109954834
    },
    {
      "auxiliary_loss_clip": 0.0115844,
      "auxiliary_loss_mlp": 0.00764947,
      "balance_loss_clip": 1.05053878,
      "balance_loss_mlp": 1.00054336,
      "epoch": 0.3390849516022365,
      "flos": 34495754112000.0,
      "grad_norm": 1.9484982247758387,
      "language_loss": 0.66536212,
      "learning_rate": 3.0789572234890057e-06,
      "loss": 0.68459594,
      "num_input_tokens_seen": 60686125,
      "step": 2820,
      "time_per_iteration": 2.7846906185150146
    },
    {
      "auxiliary_loss_clip": 0.01170457,
      "auxiliary_loss_mlp": 0.0103507,
      "balance_loss_clip": 1.05545449,
      "balance_loss_mlp": 1.02509248,
      "epoch": 0.3392051944928756,
      "flos": 16180307447040.0,
      "grad_norm": 1.7263497554181821,
      "language_loss": 0.77486765,
      "learning_rate": 3.0783012480824596e-06,
      "loss": 0.79692292,
      "num_input_tokens_seen": 60705270,
      "step": 2821,
      "time_per_iteration": 2.6532399654388428
    },
    {
      "auxiliary_loss_clip": 0.01198138,
      "auxiliary_loss_mlp": 0.01032282,
      "balance_loss_clip": 1.05590832,
      "balance_loss_mlp": 1.0224297,
      "epoch": 0.33932543738351467,
      "flos": 17086656349440.0,
      "grad_norm": 3.2473647271259716,
      "language_loss": 0.74694902,
      "learning_rate": 3.077645109098471e-06,
      "loss": 0.76925325,
      "num_input_tokens_seen": 60721540,
      "step": 2822,
      "time_per_iteration": 2.54506254196167
    },
    {
      "auxiliary_loss_clip": 0.01142773,
      "auxiliary_loss_mlp": 0.01036044,
      "balance_loss_clip": 1.05041027,
      "balance_loss_mlp": 1.02708602,
      "epoch": 0.3394456802741538,
      "flos": 22126970551680.0,
      "grad_norm": 1.629071663640302,
      "language_loss": 0.72238272,
      "learning_rate": 3.076988806636577e-06,
      "loss": 0.7441709,
      "num_input_tokens_seen": 60739300,
      "step": 2823,
      "time_per_iteration": 3.6988980770111084
    },
    {
      "auxiliary_loss_clip": 0.01173932,
      "auxiliary_loss_mlp": 0.00764985,
      "balance_loss_clip": 1.05492973,
      "balance_loss_mlp": 1.00060225,
      "epoch": 0.3395659231647929,
      "flos": 25226887121280.0,
      "grad_norm": 2.040491346928349,
      "language_loss": 0.8885498,
      "learning_rate": 3.0763323407963377e-06,
      "loss": 0.90793896,
      "num_input_tokens_seen": 60758910,
      "step": 2824,
      "time_per_iteration": 2.7320497035980225
    },
    {
      "auxiliary_loss_clip": 0.01183586,
      "auxiliary_loss_mlp": 0.01033474,
      "balance_loss_clip": 1.05301356,
      "balance_loss_mlp": 1.02467072,
      "epoch": 0.33968616605543195,
      "flos": 29096477343360.0,
      "grad_norm": 1.7187708445914205,
      "language_loss": 0.80011594,
      "learning_rate": 3.075675711677337e-06,
      "loss": 0.82228655,
      "num_input_tokens_seen": 60779005,
      "step": 2825,
      "time_per_iteration": 2.6903128623962402
    },
    {
      "auxiliary_loss_clip": 0.01164287,
      "auxiliary_loss_mlp": 0.01035507,
      "balance_loss_clip": 1.05413461,
      "balance_loss_mlp": 1.02660775,
      "epoch": 0.33980640894607106,
      "flos": 21433966479360.0,
      "grad_norm": 2.0786850703833397,
      "language_loss": 0.78006089,
      "learning_rate": 3.0750189193791865e-06,
      "loss": 0.80205882,
      "num_input_tokens_seen": 60798590,
      "step": 2826,
      "time_per_iteration": 2.6671502590179443
    },
    {
      "auxiliary_loss_clip": 0.01178917,
      "auxiliary_loss_mlp": 0.01037518,
      "balance_loss_clip": 1.05187857,
      "balance_loss_mlp": 1.02681327,
      "epoch": 0.33992665183671017,
      "flos": 32490035596800.0,
      "grad_norm": 2.0608531301958184,
      "language_loss": 0.69875908,
      "learning_rate": 3.0743619640015203e-06,
      "loss": 0.72092342,
      "num_input_tokens_seen": 60818840,
      "step": 2827,
      "time_per_iteration": 3.6292309761047363
    },
    {
      "auxiliary_loss_clip": 0.01174791,
      "auxiliary_loss_mlp": 0.01030764,
      "balance_loss_clip": 1.05249858,
      "balance_loss_mlp": 1.02119148,
      "epoch": 0.3400468947273492,
      "flos": 17055414495360.0,
      "grad_norm": 1.998968339583174,
      "language_loss": 0.92469299,
      "learning_rate": 3.073704845643999e-06,
      "loss": 0.94674861,
      "num_input_tokens_seen": 60835965,
      "step": 2828,
      "time_per_iteration": 2.684772253036499
    },
    {
      "auxiliary_loss_clip": 0.01186107,
      "auxiliary_loss_mlp": 0.01036345,
      "balance_loss_clip": 1.05260658,
      "balance_loss_mlp": 1.02655816,
      "epoch": 0.34016713761798834,
      "flos": 16872988296960.0,
      "grad_norm": 3.4057709300365318,
      "language_loss": 0.7771582,
      "learning_rate": 3.0730475644063063e-06,
      "loss": 0.79938275,
      "num_input_tokens_seen": 60851065,
      "step": 2829,
      "time_per_iteration": 2.627323627471924
    },
    {
      "auxiliary_loss_clip": 0.01161378,
      "auxiliary_loss_mlp": 0.00764218,
      "balance_loss_clip": 1.04953694,
      "balance_loss_mlp": 1.00042844,
      "epoch": 0.34028738050862745,
      "flos": 21907161273600.0,
      "grad_norm": 1.6844471749524403,
      "language_loss": 0.65132397,
      "learning_rate": 3.072390120388151e-06,
      "loss": 0.67058003,
      "num_input_tokens_seen": 60869390,
      "step": 2830,
      "time_per_iteration": 2.691819906234741
    },
    {
      "auxiliary_loss_clip": 0.01182068,
      "auxiliary_loss_mlp": 0.01033641,
      "balance_loss_clip": 1.05340576,
      "balance_loss_mlp": 1.02391386,
      "epoch": 0.3404076233992665,
      "flos": 22746034477440.0,
      "grad_norm": 4.72046042540528,
      "language_loss": 0.71333981,
      "learning_rate": 3.071732513689267e-06,
      "loss": 0.73549688,
      "num_input_tokens_seen": 60887925,
      "step": 2831,
      "time_per_iteration": 2.643235921859741
    },
    {
      "auxiliary_loss_clip": 0.01190595,
      "auxiliary_loss_mlp": 0.01036378,
      "balance_loss_clip": 1.05995131,
      "balance_loss_mlp": 1.02662122,
      "epoch": 0.3405278662899056,
      "flos": 17052361839360.0,
      "grad_norm": 2.2830905733589963,
      "language_loss": 0.67349213,
      "learning_rate": 3.0710747444094134e-06,
      "loss": 0.6957618,
      "num_input_tokens_seen": 60905955,
      "step": 2832,
      "time_per_iteration": 2.6071627140045166
    },
    {
      "auxiliary_loss_clip": 0.01171576,
      "auxiliary_loss_mlp": 0.01032916,
      "balance_loss_clip": 1.0532763,
      "balance_loss_mlp": 1.02376676,
      "epoch": 0.3406481091805447,
      "flos": 42813131783040.0,
      "grad_norm": 2.0053007801790903,
      "language_loss": 0.65211052,
      "learning_rate": 3.070416812648372e-06,
      "loss": 0.67415541,
      "num_input_tokens_seen": 60929405,
      "step": 2833,
      "time_per_iteration": 2.84083890914917
    },
    {
      "auxiliary_loss_clip": 0.01152286,
      "auxiliary_loss_mlp": 0.01026554,
      "balance_loss_clip": 1.0495863,
      "balance_loss_mlp": 1.01722622,
      "epoch": 0.3407683520711838,
      "flos": 26761457917440.0,
      "grad_norm": 2.4662652333914807,
      "language_loss": 0.65136611,
      "learning_rate": 3.069758718505951e-06,
      "loss": 0.67315447,
      "num_input_tokens_seen": 60951145,
      "step": 2834,
      "time_per_iteration": 2.74892258644104
    },
    {
      "auxiliary_loss_clip": 0.01197094,
      "auxiliary_loss_mlp": 0.01032694,
      "balance_loss_clip": 1.05724537,
      "balance_loss_mlp": 1.02356231,
      "epoch": 0.3408885949618229,
      "flos": 28767643309440.0,
      "grad_norm": 1.6564426154688001,
      "language_loss": 0.80295801,
      "learning_rate": 3.0691004620819836e-06,
      "loss": 0.82525593,
      "num_input_tokens_seen": 60971275,
      "step": 2835,
      "time_per_iteration": 2.6762404441833496
    },
    {
      "auxiliary_loss_clip": 0.0104673,
      "auxiliary_loss_mlp": 0.0100435,
      "balance_loss_clip": 1.01976395,
      "balance_loss_mlp": 1.00253797,
      "epoch": 0.341008837852462,
      "flos": 63576252881280.0,
      "grad_norm": 0.7995754838705618,
      "language_loss": 0.60143733,
      "learning_rate": 3.0684420434763254e-06,
      "loss": 0.62194812,
      "num_input_tokens_seen": 61037460,
      "step": 2836,
      "time_per_iteration": 3.322528600692749
    },
    {
      "auxiliary_loss_clip": 0.01148743,
      "auxiliary_loss_mlp": 0.01034412,
      "balance_loss_clip": 1.05397606,
      "balance_loss_mlp": 1.02551937,
      "epoch": 0.34112908074310105,
      "flos": 20812173120000.0,
      "grad_norm": 2.0513875816765386,
      "language_loss": 0.76779783,
      "learning_rate": 3.06778346278886e-06,
      "loss": 0.78962934,
      "num_input_tokens_seen": 61056295,
      "step": 2837,
      "time_per_iteration": 2.724052906036377
    },
    {
      "auxiliary_loss_clip": 0.01200124,
      "auxiliary_loss_mlp": 0.01033256,
      "balance_loss_clip": 1.05820274,
      "balance_loss_mlp": 1.02401757,
      "epoch": 0.34124932363374016,
      "flos": 24976446520320.0,
      "grad_norm": 2.8515673255095413,
      "language_loss": 0.79226649,
      "learning_rate": 3.0671247201194906e-06,
      "loss": 0.81460029,
      "num_input_tokens_seen": 61078430,
      "step": 2838,
      "time_per_iteration": 2.6360459327697754
    },
    {
      "auxiliary_loss_clip": 0.01155118,
      "auxiliary_loss_mlp": 0.01036935,
      "balance_loss_clip": 1.05024981,
      "balance_loss_mlp": 1.02754164,
      "epoch": 0.3413695665243792,
      "flos": 28402970480640.0,
      "grad_norm": 1.636593991037116,
      "language_loss": 0.75525796,
      "learning_rate": 3.066465815568151e-06,
      "loss": 0.77717853,
      "num_input_tokens_seen": 61099260,
      "step": 2839,
      "time_per_iteration": 2.809102773666382
    },
    {
      "auxiliary_loss_clip": 0.01183212,
      "auxiliary_loss_mlp": 0.01028983,
      "balance_loss_clip": 1.05224907,
      "balance_loss_mlp": 1.02005398,
      "epoch": 0.34148980941501833,
      "flos": 25302012416640.0,
      "grad_norm": 1.7736875327987929,
      "language_loss": 0.69107842,
      "learning_rate": 3.0658067492347947e-06,
      "loss": 0.71320033,
      "num_input_tokens_seen": 61121900,
      "step": 2840,
      "time_per_iteration": 2.6929428577423096
    },
    {
      "auxiliary_loss_clip": 0.01099438,
      "auxiliary_loss_mlp": 0.01039776,
      "balance_loss_clip": 1.04298997,
      "balance_loss_mlp": 1.03062701,
      "epoch": 0.34161005230565744,
      "flos": 17530081747200.0,
      "grad_norm": 1.9914166940659639,
      "language_loss": 0.66682637,
      "learning_rate": 3.065147521219402e-06,
      "loss": 0.68821847,
      "num_input_tokens_seen": 61141155,
      "step": 2841,
      "time_per_iteration": 2.821664571762085
    },
    {
      "auxiliary_loss_clip": 0.01161306,
      "auxiliary_loss_mlp": 0.01040333,
      "balance_loss_clip": 1.05410123,
      "balance_loss_mlp": 1.03210723,
      "epoch": 0.3417302951962965,
      "flos": 43650101566080.0,
      "grad_norm": 1.5251257817472503,
      "language_loss": 0.74676859,
      "learning_rate": 3.064488131621977e-06,
      "loss": 0.768785,
      "num_input_tokens_seen": 61164480,
      "step": 2842,
      "time_per_iteration": 2.894273281097412
    },
    {
      "auxiliary_loss_clip": 0.01172457,
      "auxiliary_loss_mlp": 0.0102866,
      "balance_loss_clip": 1.04879522,
      "balance_loss_mlp": 1.01925492,
      "epoch": 0.3418505380869356,
      "flos": 30882207012480.0,
      "grad_norm": 1.649158485868022,
      "language_loss": 0.73771644,
      "learning_rate": 3.063828580542549e-06,
      "loss": 0.7597276,
      "num_input_tokens_seen": 61185675,
      "step": 2843,
      "time_per_iteration": 2.721012830734253
    },
    {
      "auxiliary_loss_clip": 0.01168515,
      "auxiliary_loss_mlp": 0.01034242,
      "balance_loss_clip": 1.05252802,
      "balance_loss_mlp": 1.0261718,
      "epoch": 0.3419707809775747,
      "flos": 19463871277440.0,
      "grad_norm": 2.257092417795917,
      "language_loss": 0.73516053,
      "learning_rate": 3.0631688680811706e-06,
      "loss": 0.75718808,
      "num_input_tokens_seen": 61205300,
      "step": 2844,
      "time_per_iteration": 2.7157676219940186
    },
    {
      "auxiliary_loss_clip": 0.01196222,
      "auxiliary_loss_mlp": 0.01027087,
      "balance_loss_clip": 1.05376196,
      "balance_loss_mlp": 1.01873016,
      "epoch": 0.3420910238682138,
      "flos": 28727818104960.0,
      "grad_norm": 2.3137204545747396,
      "language_loss": 0.75612062,
      "learning_rate": 3.062508994337921e-06,
      "loss": 0.77835369,
      "num_input_tokens_seen": 61224905,
      "step": 2845,
      "time_per_iteration": 3.5770349502563477
    },
    {
      "auxiliary_loss_clip": 0.01178336,
      "auxiliary_loss_mlp": 0.01032252,
      "balance_loss_clip": 1.04969049,
      "balance_loss_mlp": 1.02259064,
      "epoch": 0.3422112667588529,
      "flos": 21397265758080.0,
      "grad_norm": 2.19985438351455,
      "language_loss": 0.79291874,
      "learning_rate": 3.0618489594129013e-06,
      "loss": 0.81502461,
      "num_input_tokens_seen": 61243045,
      "step": 2846,
      "time_per_iteration": 2.6371586322784424
    },
    {
      "auxiliary_loss_clip": 0.01162016,
      "auxiliary_loss_mlp": 0.0103485,
      "balance_loss_clip": 1.05326307,
      "balance_loss_mlp": 1.02555752,
      "epoch": 0.342331509649492,
      "flos": 13881450038400.0,
      "grad_norm": 2.926954843890118,
      "language_loss": 0.71478474,
      "learning_rate": 3.061188763406239e-06,
      "loss": 0.73675334,
      "num_input_tokens_seen": 61259190,
      "step": 2847,
      "time_per_iteration": 2.6804728507995605
    },
    {
      "auxiliary_loss_clip": 0.01162096,
      "auxiliary_loss_mlp": 0.01033662,
      "balance_loss_clip": 1.04872596,
      "balance_loss_mlp": 1.02479291,
      "epoch": 0.34245175254013105,
      "flos": 28621450955520.0,
      "grad_norm": 2.9979014380559716,
      "language_loss": 0.82621658,
      "learning_rate": 3.060528406418085e-06,
      "loss": 0.8481741,
      "num_input_tokens_seen": 61279040,
      "step": 2848,
      "time_per_iteration": 2.7577226161956787
    },
    {
      "auxiliary_loss_clip": 0.0115818,
      "auxiliary_loss_mlp": 0.01034581,
      "balance_loss_clip": 1.0486362,
      "balance_loss_mlp": 1.02596223,
      "epoch": 0.34257199543077016,
      "flos": 34127058960000.0,
      "grad_norm": 1.783017449588709,
      "language_loss": 0.62196153,
      "learning_rate": 3.0598678885486145e-06,
      "loss": 0.64388913,
      "num_input_tokens_seen": 61301580,
      "step": 2849,
      "time_per_iteration": 3.7052552700042725
    },
    {
      "auxiliary_loss_clip": 0.0115384,
      "auxiliary_loss_mlp": 0.00764538,
      "balance_loss_clip": 1.04710543,
      "balance_loss_mlp": 1.00041521,
      "epoch": 0.34269223832140927,
      "flos": 19974018188160.0,
      "grad_norm": 1.6228463838476326,
      "language_loss": 0.74397957,
      "learning_rate": 3.0592072098980282e-06,
      "loss": 0.76316339,
      "num_input_tokens_seen": 61321240,
      "step": 2850,
      "time_per_iteration": 2.747894048690796
    },
    {
      "auxiliary_loss_clip": 0.01167304,
      "auxiliary_loss_mlp": 0.01032218,
      "balance_loss_clip": 1.05446982,
      "balance_loss_mlp": 1.02279449,
      "epoch": 0.3428124812120483,
      "flos": 27235658292480.0,
      "grad_norm": 9.860825127107347,
      "language_loss": 0.73225224,
      "learning_rate": 3.0585463705665514e-06,
      "loss": 0.75424743,
      "num_input_tokens_seen": 61341615,
      "step": 2851,
      "time_per_iteration": 2.7802491188049316
    },
    {
      "auxiliary_loss_clip": 0.01150671,
      "auxiliary_loss_mlp": 0.01032792,
      "balance_loss_clip": 1.04866767,
      "balance_loss_mlp": 1.0237205,
      "epoch": 0.34293272410268744,
      "flos": 24570871079040.0,
      "grad_norm": 2.182687778186288,
      "language_loss": 0.70883656,
      "learning_rate": 3.0578853706544304e-06,
      "loss": 0.73067117,
      "num_input_tokens_seen": 61359005,
      "step": 2852,
      "time_per_iteration": 2.716604232788086
    },
    {
      "auxiliary_loss_clip": 0.01158729,
      "auxiliary_loss_mlp": 0.00765694,
      "balance_loss_clip": 1.05195796,
      "balance_loss_mlp": 1.00050068,
      "epoch": 0.34305296699332655,
      "flos": 21506865131520.0,
      "grad_norm": 2.0664333003523763,
      "language_loss": 0.65424538,
      "learning_rate": 3.0572242102619404e-06,
      "loss": 0.67348963,
      "num_input_tokens_seen": 61376160,
      "step": 2853,
      "time_per_iteration": 4.613559246063232
    },
    {
      "auxiliary_loss_clip": 0.01167819,
      "auxiliary_loss_mlp": 0.01034619,
      "balance_loss_clip": 1.05464435,
      "balance_loss_mlp": 1.02547526,
      "epoch": 0.3431732098839656,
      "flos": 24056665931520.0,
      "grad_norm": 2.266202526678044,
      "language_loss": 0.80314934,
      "learning_rate": 3.0565628894893784e-06,
      "loss": 0.82517374,
      "num_input_tokens_seen": 61396795,
      "step": 2854,
      "time_per_iteration": 2.7695865631103516
    },
    {
      "auxiliary_loss_clip": 0.01175277,
      "auxiliary_loss_mlp": 0.01032528,
      "balance_loss_clip": 1.0538274,
      "balance_loss_mlp": 1.02318788,
      "epoch": 0.3432934527746047,
      "flos": 16800879744000.0,
      "grad_norm": 1.7322494643114317,
      "language_loss": 0.74874997,
      "learning_rate": 3.0559014084370655e-06,
      "loss": 0.77082801,
      "num_input_tokens_seen": 61415320,
      "step": 2855,
      "time_per_iteration": 2.5943195819854736
    },
    {
      "auxiliary_loss_clip": 0.01174848,
      "auxiliary_loss_mlp": 0.01029988,
      "balance_loss_clip": 1.05183434,
      "balance_loss_mlp": 1.02130342,
      "epoch": 0.34341369566524377,
      "flos": 23439720908160.0,
      "grad_norm": 1.6783372036722697,
      "language_loss": 0.79107606,
      "learning_rate": 3.055239767205349e-06,
      "loss": 0.81312448,
      "num_input_tokens_seen": 61437070,
      "step": 2856,
      "time_per_iteration": 2.6969807147979736
    },
    {
      "auxiliary_loss_clip": 0.01184433,
      "auxiliary_loss_mlp": 0.01033702,
      "balance_loss_clip": 1.05910695,
      "balance_loss_mlp": 1.02448666,
      "epoch": 0.3435339385558829,
      "flos": 17267466435840.0,
      "grad_norm": 1.8244372709003325,
      "language_loss": 0.78408372,
      "learning_rate": 3.054577965894599e-06,
      "loss": 0.80626512,
      "num_input_tokens_seen": 61453215,
      "step": 2857,
      "time_per_iteration": 2.6055562496185303
    },
    {
      "auxiliary_loss_clip": 0.01178166,
      "auxiliary_loss_mlp": 0.01034437,
      "balance_loss_clip": 1.05764127,
      "balance_loss_mlp": 1.0246619,
      "epoch": 0.343654181446522,
      "flos": 22199366413440.0,
      "grad_norm": 1.6263068050894454,
      "language_loss": 0.70554394,
      "learning_rate": 3.0539160046052094e-06,
      "loss": 0.72766995,
      "num_input_tokens_seen": 61472915,
      "step": 2858,
      "time_per_iteration": 2.704587697982788
    },
    {
      "auxiliary_loss_clip": 0.0116071,
      "auxiliary_loss_mlp": 0.01029741,
      "balance_loss_clip": 1.04961073,
      "balance_loss_mlp": 1.01964378,
      "epoch": 0.34377442433716104,
      "flos": 19901801894400.0,
      "grad_norm": 2.4987884022229125,
      "language_loss": 0.70993578,
      "learning_rate": 3.0532538834376003e-06,
      "loss": 0.73184031,
      "num_input_tokens_seen": 61492475,
      "step": 2859,
      "time_per_iteration": 2.6559133529663086
    },
    {
      "auxiliary_loss_clip": 0.01188795,
      "auxiliary_loss_mlp": 0.01028133,
      "balance_loss_clip": 1.05535388,
      "balance_loss_mlp": 1.01860857,
      "epoch": 0.34389466722780015,
      "flos": 22197678474240.0,
      "grad_norm": 1.74606880100356,
      "language_loss": 0.78414196,
      "learning_rate": 3.0525916024922143e-06,
      "loss": 0.80631125,
      "num_input_tokens_seen": 61511660,
      "step": 2860,
      "time_per_iteration": 2.6374945640563965
    },
    {
      "auxiliary_loss_clip": 0.01165339,
      "auxiliary_loss_mlp": 0.01035252,
      "balance_loss_clip": 1.05136895,
      "balance_loss_mlp": 1.02608466,
      "epoch": 0.34401491011843927,
      "flos": 18624567110400.0,
      "grad_norm": 3.3647883122334425,
      "language_loss": 0.84345669,
      "learning_rate": 3.0519291618695193e-06,
      "loss": 0.8654626,
      "num_input_tokens_seen": 61529060,
      "step": 2861,
      "time_per_iteration": 2.7542717456817627
    },
    {
      "auxiliary_loss_clip": 0.01144258,
      "auxiliary_loss_mlp": 0.01033508,
      "balance_loss_clip": 1.04726577,
      "balance_loss_mlp": 1.02454352,
      "epoch": 0.3441351530090783,
      "flos": 17858197509120.0,
      "grad_norm": 1.7856859761648556,
      "language_loss": 0.75898218,
      "learning_rate": 3.0512665616700065e-06,
      "loss": 0.78075981,
      "num_input_tokens_seen": 61548125,
      "step": 2862,
      "time_per_iteration": 2.647963523864746
    },
    {
      "auxiliary_loss_clip": 0.01131008,
      "auxiliary_loss_mlp": 0.01031774,
      "balance_loss_clip": 1.04606116,
      "balance_loss_mlp": 1.02263081,
      "epoch": 0.34425539589971743,
      "flos": 23112754381440.0,
      "grad_norm": 2.2003947585094323,
      "language_loss": 0.8901161,
      "learning_rate": 3.0506038019941933e-06,
      "loss": 0.91174394,
      "num_input_tokens_seen": 61568135,
      "step": 2863,
      "time_per_iteration": 2.7640604972839355
    },
    {
      "auxiliary_loss_clip": 0.01155877,
      "auxiliary_loss_mlp": 0.01034949,
      "balance_loss_clip": 1.05197382,
      "balance_loss_mlp": 1.02573395,
      "epoch": 0.34437563879035654,
      "flos": 21907699977600.0,
      "grad_norm": 2.2976310968306812,
      "language_loss": 0.67866457,
      "learning_rate": 3.049940882942617e-06,
      "loss": 0.70057285,
      "num_input_tokens_seen": 61586920,
      "step": 2864,
      "time_per_iteration": 2.7059905529022217
    },
    {
      "auxiliary_loss_clip": 0.0119872,
      "auxiliary_loss_mlp": 0.01036276,
      "balance_loss_clip": 1.05596566,
      "balance_loss_mlp": 1.02729392,
      "epoch": 0.3444958816809956,
      "flos": 23076915586560.0,
      "grad_norm": 1.8978674968867242,
      "language_loss": 0.80128455,
      "learning_rate": 3.0492778046158448e-06,
      "loss": 0.82363451,
      "num_input_tokens_seen": 61608340,
      "step": 2865,
      "time_per_iteration": 2.6251463890075684
    },
    {
      "auxiliary_loss_clip": 0.01182291,
      "auxiliary_loss_mlp": 0.01043336,
      "balance_loss_clip": 1.05659151,
      "balance_loss_mlp": 1.03391838,
      "epoch": 0.3446161245716347,
      "flos": 21908633731200.0,
      "grad_norm": 2.467866111083122,
      "language_loss": 0.76898736,
      "learning_rate": 3.0486145671144633e-06,
      "loss": 0.79124361,
      "num_input_tokens_seen": 61628130,
      "step": 2866,
      "time_per_iteration": 2.649470806121826
    },
    {
      "auxiliary_loss_clip": 0.01109651,
      "auxiliary_loss_mlp": 0.01037497,
      "balance_loss_clip": 1.04620302,
      "balance_loss_mlp": 1.02816844,
      "epoch": 0.3447363674622738,
      "flos": 25112834461440.0,
      "grad_norm": 2.9902340593064585,
      "language_loss": 0.77028465,
      "learning_rate": 3.047951170539086e-06,
      "loss": 0.79175615,
      "num_input_tokens_seen": 61647755,
      "step": 2867,
      "time_per_iteration": 2.806272029876709
    },
    {
      "auxiliary_loss_clip": 0.01153494,
      "auxiliary_loss_mlp": 0.01029313,
      "balance_loss_clip": 1.05496538,
      "balance_loss_mlp": 1.02159405,
      "epoch": 0.3448566103529129,
      "flos": 11984684451840.0,
      "grad_norm": 2.389450639577141,
      "language_loss": 0.84310877,
      "learning_rate": 3.047287614990349e-06,
      "loss": 0.86493683,
      "num_input_tokens_seen": 61665675,
      "step": 2868,
      "time_per_iteration": 2.6753413677215576
    },
    {
      "auxiliary_loss_clip": 0.01160611,
      "auxiliary_loss_mlp": 0.01034555,
      "balance_loss_clip": 1.05167043,
      "balance_loss_mlp": 1.02512538,
      "epoch": 0.344976853243552,
      "flos": 40187882465280.0,
      "grad_norm": 4.5580492841690186,
      "language_loss": 0.62136453,
      "learning_rate": 3.046623900568914e-06,
      "loss": 0.64331615,
      "num_input_tokens_seen": 61688240,
      "step": 2869,
      "time_per_iteration": 2.8364362716674805
    },
    {
      "auxiliary_loss_clip": 0.0116668,
      "auxiliary_loss_mlp": 0.01031469,
      "balance_loss_clip": 1.05259931,
      "balance_loss_mlp": 1.02230215,
      "epoch": 0.34509709613419104,
      "flos": 28723652127360.0,
      "grad_norm": 2.669725601366213,
      "language_loss": 0.70567656,
      "learning_rate": 3.045960027375465e-06,
      "loss": 0.72765803,
      "num_input_tokens_seen": 61706075,
      "step": 2870,
      "time_per_iteration": 3.6382625102996826
    },
    {
      "auxiliary_loss_clip": 0.01183621,
      "auxiliary_loss_mlp": 0.01033302,
      "balance_loss_clip": 1.05145407,
      "balance_loss_mlp": 1.0233357,
      "epoch": 0.34521733902483015,
      "flos": 29967597982080.0,
      "grad_norm": 3.7615810647688424,
      "language_loss": 0.82995391,
      "learning_rate": 3.045295995510711e-06,
      "loss": 0.85212314,
      "num_input_tokens_seen": 61723045,
      "step": 2871,
      "time_per_iteration": 2.6747677326202393
    },
    {
      "auxiliary_loss_clip": 0.0116694,
      "auxiliary_loss_mlp": 0.01027287,
      "balance_loss_clip": 1.0542115,
      "balance_loss_mlp": 1.01899576,
      "epoch": 0.34533758191546926,
      "flos": 27923059843200.0,
      "grad_norm": 1.7900756850485666,
      "language_loss": 0.73687518,
      "learning_rate": 3.0446318050753865e-06,
      "loss": 0.75881743,
      "num_input_tokens_seen": 61743525,
      "step": 2872,
      "time_per_iteration": 2.749258041381836
    },
    {
      "auxiliary_loss_clip": 0.01173147,
      "auxiliary_loss_mlp": 0.01034379,
      "balance_loss_clip": 1.05095756,
      "balance_loss_mlp": 1.02606463,
      "epoch": 0.3454578248061083,
      "flos": 27125879351040.0,
      "grad_norm": 1.9999534637097485,
      "language_loss": 0.77585185,
      "learning_rate": 3.0439674561702474e-06,
      "loss": 0.79792708,
      "num_input_tokens_seen": 61763025,
      "step": 2873,
      "time_per_iteration": 2.619555711746216
    },
    {
      "auxiliary_loss_clip": 0.0117788,
      "auxiliary_loss_mlp": 0.01028336,
      "balance_loss_clip": 1.05203247,
      "balance_loss_mlp": 1.01926982,
      "epoch": 0.3455780676967474,
      "flos": 19024899166080.0,
      "grad_norm": 2.6805770412502206,
      "language_loss": 0.88129765,
      "learning_rate": 3.043302948896076e-06,
      "loss": 0.90335977,
      "num_input_tokens_seen": 61781630,
      "step": 2874,
      "time_per_iteration": 2.638460397720337
    },
    {
      "auxiliary_loss_clip": 0.01126874,
      "auxiliary_loss_mlp": 0.01037629,
      "balance_loss_clip": 1.04737663,
      "balance_loss_mlp": 1.02813435,
      "epoch": 0.34569831058738654,
      "flos": 34496005507200.0,
      "grad_norm": 1.7865243925462448,
      "language_loss": 0.60478449,
      "learning_rate": 3.0426382833536756e-06,
      "loss": 0.62642956,
      "num_input_tokens_seen": 61804985,
      "step": 2875,
      "time_per_iteration": 3.727134943008423
    },
    {
      "auxiliary_loss_clip": 0.01145448,
      "auxiliary_loss_mlp": 0.01030679,
      "balance_loss_clip": 1.04743695,
      "balance_loss_mlp": 1.02073693,
      "epoch": 0.3458185534780256,
      "flos": 31138681098240.0,
      "grad_norm": 2.8996408047444957,
      "language_loss": 0.78090417,
      "learning_rate": 3.041973459643877e-06,
      "loss": 0.80266547,
      "num_input_tokens_seen": 61824440,
      "step": 2876,
      "time_per_iteration": 2.7294657230377197
    },
    {
      "auxiliary_loss_clip": 0.01130143,
      "auxiliary_loss_mlp": 0.01029149,
      "balance_loss_clip": 1.04448164,
      "balance_loss_mlp": 1.01992214,
      "epoch": 0.3459387963686647,
      "flos": 32452508862720.0,
      "grad_norm": 3.164096910885201,
      "language_loss": 0.67439353,
      "learning_rate": 3.0413084778675334e-06,
      "loss": 0.69598639,
      "num_input_tokens_seen": 61845690,
      "step": 2877,
      "time_per_iteration": 2.8787052631378174
    },
    {
      "auxiliary_loss_clip": 0.01159682,
      "auxiliary_loss_mlp": 0.00763886,
      "balance_loss_clip": 1.04812133,
      "balance_loss_mlp": 1.00036764,
      "epoch": 0.3460590392593038,
      "flos": 24675658030080.0,
      "grad_norm": 2.108638406057002,
      "language_loss": 0.83878267,
      "learning_rate": 3.0406433381255214e-06,
      "loss": 0.85801828,
      "num_input_tokens_seen": 61863725,
      "step": 2878,
      "time_per_iteration": 2.692603349685669
    },
    {
      "auxiliary_loss_clip": 0.01182279,
      "auxiliary_loss_mlp": 0.01032062,
      "balance_loss_clip": 1.05661821,
      "balance_loss_mlp": 1.02353263,
      "epoch": 0.34617928214994287,
      "flos": 18807316531200.0,
      "grad_norm": 3.5299347199522035,
      "language_loss": 0.82588059,
      "learning_rate": 3.0399780405187425e-06,
      "loss": 0.84802401,
      "num_input_tokens_seen": 61882720,
      "step": 2879,
      "time_per_iteration": 4.46898078918457
    },
    {
      "auxiliary_loss_clip": 0.01175361,
      "auxiliary_loss_mlp": 0.01026183,
      "balance_loss_clip": 1.05087245,
      "balance_loss_mlp": 1.01736176,
      "epoch": 0.346299525040582,
      "flos": 24857653265280.0,
      "grad_norm": 1.7925401548908713,
      "language_loss": 0.78722072,
      "learning_rate": 3.0393125851481216e-06,
      "loss": 0.80923611,
      "num_input_tokens_seen": 61902595,
      "step": 2880,
      "time_per_iteration": 2.70742130279541
    },
    {
      "auxiliary_loss_clip": 0.01146425,
      "auxiliary_loss_mlp": 0.01025734,
      "balance_loss_clip": 1.04992115,
      "balance_loss_mlp": 1.01698995,
      "epoch": 0.3464197679312211,
      "flos": 16434914025600.0,
      "grad_norm": 2.1586192424280477,
      "language_loss": 0.86443168,
      "learning_rate": 3.038646972114608e-06,
      "loss": 0.88615328,
      "num_input_tokens_seen": 61918920,
      "step": 2881,
      "time_per_iteration": 2.6598329544067383
    },
    {
      "auxiliary_loss_clip": 0.01149637,
      "auxiliary_loss_mlp": 0.01031893,
      "balance_loss_clip": 1.05235338,
      "balance_loss_mlp": 1.02307153,
      "epoch": 0.34654001082186014,
      "flos": 22382474970240.0,
      "grad_norm": 1.7664041781111552,
      "language_loss": 0.67468321,
      "learning_rate": 3.037981201519174e-06,
      "loss": 0.69649851,
      "num_input_tokens_seen": 61939520,
      "step": 2882,
      "time_per_iteration": 2.791757822036743
    },
    {
      "auxiliary_loss_clip": 0.01181564,
      "auxiliary_loss_mlp": 0.01029375,
      "balance_loss_clip": 1.05540395,
      "balance_loss_mlp": 1.02082145,
      "epoch": 0.34666025371249926,
      "flos": 19573901614080.0,
      "grad_norm": 2.018592588478196,
      "language_loss": 0.71181327,
      "learning_rate": 3.0373152734628175e-06,
      "loss": 0.7339226,
      "num_input_tokens_seen": 61957800,
      "step": 2883,
      "time_per_iteration": 2.5992274284362793
    },
    {
      "auxiliary_loss_clip": 0.011751,
      "auxiliary_loss_mlp": 0.01032658,
      "balance_loss_clip": 1.05171108,
      "balance_loss_mlp": 1.0242002,
      "epoch": 0.34678049660313837,
      "flos": 15267637751040.0,
      "grad_norm": 2.357607523761916,
      "language_loss": 0.75991654,
      "learning_rate": 3.0366491880465584e-06,
      "loss": 0.7819941,
      "num_input_tokens_seen": 61975820,
      "step": 2884,
      "time_per_iteration": 2.6236507892608643
    },
    {
      "auxiliary_loss_clip": 0.01201355,
      "auxiliary_loss_mlp": 0.01034887,
      "balance_loss_clip": 1.05933952,
      "balance_loss_mlp": 1.02557063,
      "epoch": 0.3469007394937774,
      "flos": 21181550630400.0,
      "grad_norm": 1.642785465326633,
      "language_loss": 0.81999558,
      "learning_rate": 3.035982945371443e-06,
      "loss": 0.84235799,
      "num_input_tokens_seen": 61997515,
      "step": 2885,
      "time_per_iteration": 2.6460988521575928
    },
    {
      "auxiliary_loss_clip": 0.01175191,
      "auxiliary_loss_mlp": 0.01029775,
      "balance_loss_clip": 1.05365038,
      "balance_loss_mlp": 1.02114975,
      "epoch": 0.34702098238441653,
      "flos": 22375471818240.0,
      "grad_norm": 2.0345710573140416,
      "language_loss": 0.8548528,
      "learning_rate": 3.035316545538537e-06,
      "loss": 0.87690252,
      "num_input_tokens_seen": 62016310,
      "step": 2886,
      "time_per_iteration": 2.7000975608825684
    },
    {
      "auxiliary_loss_clip": 0.01168764,
      "auxiliary_loss_mlp": 0.01030408,
      "balance_loss_clip": 1.05856061,
      "balance_loss_mlp": 1.02202177,
      "epoch": 0.3471412252750556,
      "flos": 22929430343040.0,
      "grad_norm": 3.107626737647839,
      "language_loss": 0.79695755,
      "learning_rate": 3.034649988648935e-06,
      "loss": 0.81894928,
      "num_input_tokens_seen": 62036075,
      "step": 2887,
      "time_per_iteration": 2.664337158203125
    },
    {
      "auxiliary_loss_clip": 0.01167208,
      "auxiliary_loss_mlp": 0.01027824,
      "balance_loss_clip": 1.05183291,
      "balance_loss_mlp": 1.0189662,
      "epoch": 0.3472614681656947,
      "flos": 21324259365120.0,
      "grad_norm": 1.829341436275867,
      "language_loss": 0.80652487,
      "learning_rate": 3.033983274803752e-06,
      "loss": 0.82847512,
      "num_input_tokens_seen": 62055865,
      "step": 2888,
      "time_per_iteration": 2.7027506828308105
    },
    {
      "auxiliary_loss_clip": 0.01161904,
      "auxiliary_loss_mlp": 0.01033257,
      "balance_loss_clip": 1.05066836,
      "balance_loss_mlp": 1.02475178,
      "epoch": 0.3473817110563338,
      "flos": 23475739271040.0,
      "grad_norm": 2.0733122407669224,
      "language_loss": 0.72475326,
      "learning_rate": 3.0333164041041283e-06,
      "loss": 0.74670494,
      "num_input_tokens_seen": 62072180,
      "step": 2889,
      "time_per_iteration": 2.6812119483947754
    },
    {
      "auxiliary_loss_clip": 0.01126991,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 1.04945183,
      "balance_loss_mlp": 1.01970553,
      "epoch": 0.34750195394697286,
      "flos": 22346025644160.0,
      "grad_norm": 2.5021119396543354,
      "language_loss": 0.71676064,
      "learning_rate": 3.032649376651228e-06,
      "loss": 0.73831046,
      "num_input_tokens_seen": 62091600,
      "step": 2890,
      "time_per_iteration": 2.792116165161133
    },
    {
      "auxiliary_loss_clip": 0.01152676,
      "auxiliary_loss_mlp": 0.01035971,
      "balance_loss_clip": 1.05083287,
      "balance_loss_mlp": 1.02605319,
      "epoch": 0.347622196837612,
      "flos": 29095004885760.0,
      "grad_norm": 2.243178749572877,
      "language_loss": 0.7594465,
      "learning_rate": 3.031982192546238e-06,
      "loss": 0.78133303,
      "num_input_tokens_seen": 62114695,
      "step": 2891,
      "time_per_iteration": 2.7642650604248047
    },
    {
      "auxiliary_loss_clip": 0.01183718,
      "auxiliary_loss_mlp": 0.0102844,
      "balance_loss_clip": 1.05355501,
      "balance_loss_mlp": 1.02051854,
      "epoch": 0.3477424397282511,
      "flos": 22455732758400.0,
      "grad_norm": 8.420097990764068,
      "language_loss": 0.95007885,
      "learning_rate": 3.0313148518903696e-06,
      "loss": 0.97220039,
      "num_input_tokens_seen": 62134520,
      "step": 2892,
      "time_per_iteration": 2.6494662761688232
    },
    {
      "auxiliary_loss_clip": 0.01174852,
      "auxiliary_loss_mlp": 0.01031806,
      "balance_loss_clip": 1.05749059,
      "balance_loss_mlp": 1.02345514,
      "epoch": 0.34786268261889014,
      "flos": 15778790242560.0,
      "grad_norm": 2.479159479969363,
      "language_loss": 0.81102443,
      "learning_rate": 3.030647354784859e-06,
      "loss": 0.83309102,
      "num_input_tokens_seen": 62151560,
      "step": 2893,
      "time_per_iteration": 2.7526423931121826
    },
    {
      "auxiliary_loss_clip": 0.01148995,
      "auxiliary_loss_mlp": 0.0102987,
      "balance_loss_clip": 1.04957628,
      "balance_loss_mlp": 1.02206171,
      "epoch": 0.34798292550952925,
      "flos": 20777627214720.0,
      "grad_norm": 1.8509895514881833,
      "language_loss": 0.77278435,
      "learning_rate": 3.029979701330964e-06,
      "loss": 0.79457301,
      "num_input_tokens_seen": 62170985,
      "step": 2894,
      "time_per_iteration": 2.723860263824463
    },
    {
      "auxiliary_loss_clip": 0.01172855,
      "auxiliary_loss_mlp": 0.01032695,
      "balance_loss_clip": 1.05386794,
      "balance_loss_mlp": 1.02377212,
      "epoch": 0.34810316840016836,
      "flos": 19937820257280.0,
      "grad_norm": 2.5211748898191777,
      "language_loss": 0.8044644,
      "learning_rate": 3.029311891629966e-06,
      "loss": 0.82651985,
      "num_input_tokens_seen": 62189440,
      "step": 2895,
      "time_per_iteration": 2.6895737648010254
    },
    {
      "auxiliary_loss_clip": 0.01164186,
      "auxiliary_loss_mlp": 0.01029764,
      "balance_loss_clip": 1.05314589,
      "balance_loss_mlp": 1.02088308,
      "epoch": 0.3482234112908074,
      "flos": 23623296341760.0,
      "grad_norm": 1.8313039277942187,
      "language_loss": 0.74818659,
      "learning_rate": 3.0286439257831744e-06,
      "loss": 0.7701261,
      "num_input_tokens_seen": 62208910,
      "step": 2896,
      "time_per_iteration": 3.612828016281128
    },
    {
      "auxiliary_loss_clip": 0.01199023,
      "auxiliary_loss_mlp": 0.01032185,
      "balance_loss_clip": 1.05663133,
      "balance_loss_mlp": 1.02208161,
      "epoch": 0.3483436541814465,
      "flos": 23986712194560.0,
      "grad_norm": 2.15714261934903,
      "language_loss": 0.71846879,
      "learning_rate": 3.0279758038919156e-06,
      "loss": 0.74078083,
      "num_input_tokens_seen": 62227135,
      "step": 2897,
      "time_per_iteration": 2.6443543434143066
    },
    {
      "auxiliary_loss_clip": 0.01185435,
      "auxiliary_loss_mlp": 0.01029107,
      "balance_loss_clip": 1.05736327,
      "balance_loss_mlp": 1.01977921,
      "epoch": 0.34846389707208564,
      "flos": 22638338524800.0,
      "grad_norm": 2.0154863267991954,
      "language_loss": 0.7854352,
      "learning_rate": 3.0273075260575455e-06,
      "loss": 0.80758065,
      "num_input_tokens_seen": 62246035,
      "step": 2898,
      "time_per_iteration": 2.6476950645446777
    },
    {
      "auxiliary_loss_clip": 0.01169868,
      "auxiliary_loss_mlp": 0.010316,
      "balance_loss_clip": 1.05182791,
      "balance_loss_mlp": 1.02248633,
      "epoch": 0.3485841399627247,
      "flos": 21792857218560.0,
      "grad_norm": 1.8491760669656492,
      "language_loss": 0.80910677,
      "learning_rate": 3.0266390923814396e-06,
      "loss": 0.8311215,
      "num_input_tokens_seen": 62264095,
      "step": 2899,
      "time_per_iteration": 2.67299485206604
    },
    {
      "auxiliary_loss_clip": 0.0117492,
      "auxiliary_loss_mlp": 0.01030865,
      "balance_loss_clip": 1.05857134,
      "balance_loss_mlp": 1.02204347,
      "epoch": 0.3487043828533638,
      "flos": 17019036996480.0,
      "grad_norm": 1.820856832493506,
      "language_loss": 0.8239553,
      "learning_rate": 3.0259705029650008e-06,
      "loss": 0.84601313,
      "num_input_tokens_seen": 62282025,
      "step": 2900,
      "time_per_iteration": 2.7655794620513916
    },
    {
      "auxiliary_loss_clip": 0.0118347,
      "auxiliary_loss_mlp": 0.01026336,
      "balance_loss_clip": 1.05371737,
      "balance_loss_mlp": 1.01784849,
      "epoch": 0.34882462574400286,
      "flos": 22601135013120.0,
      "grad_norm": 1.7937924106543541,
      "language_loss": 0.729379,
      "learning_rate": 3.025301757909652e-06,
      "loss": 0.75147712,
      "num_input_tokens_seen": 62302220,
      "step": 2901,
      "time_per_iteration": 3.5888545513153076
    },
    {
      "auxiliary_loss_clip": 0.01151367,
      "auxiliary_loss_mlp": 0.00764075,
      "balance_loss_clip": 1.05110073,
      "balance_loss_mlp": 1.00029755,
      "epoch": 0.34894486863464197,
      "flos": 29861518141440.0,
      "grad_norm": 1.4592093581033876,
      "language_loss": 0.80749345,
      "learning_rate": 3.024632857316842e-06,
      "loss": 0.82664788,
      "num_input_tokens_seen": 62323535,
      "step": 2902,
      "time_per_iteration": 2.8104183673858643
    },
    {
      "auxiliary_loss_clip": 0.01184938,
      "auxiliary_loss_mlp": 0.01038986,
      "balance_loss_clip": 1.05744886,
      "balance_loss_mlp": 1.03061104,
      "epoch": 0.3490651115252811,
      "flos": 22122265870080.0,
      "grad_norm": 2.5239466025280417,
      "language_loss": 0.77306384,
      "learning_rate": 3.0239638012880412e-06,
      "loss": 0.79530305,
      "num_input_tokens_seen": 62343430,
      "step": 2903,
      "time_per_iteration": 2.6399543285369873
    },
    {
      "auxiliary_loss_clip": 0.01131492,
      "auxiliary_loss_mlp": 0.01030589,
      "balance_loss_clip": 1.0480957,
      "balance_loss_mlp": 1.02149963,
      "epoch": 0.34918535441592014,
      "flos": 12676682943360.0,
      "grad_norm": 2.4889272468711616,
      "language_loss": 0.81751055,
      "learning_rate": 3.0232945899247466e-06,
      "loss": 0.83913136,
      "num_input_tokens_seen": 62360365,
      "step": 2904,
      "time_per_iteration": 2.7688820362091064
    },
    {
      "auxiliary_loss_clip": 0.01183776,
      "auxiliary_loss_mlp": 0.01032673,
      "balance_loss_clip": 1.05317509,
      "balance_loss_mlp": 1.02325523,
      "epoch": 0.34930559730655925,
      "flos": 23185617120000.0,
      "grad_norm": 2.0882445401698613,
      "language_loss": 0.77496684,
      "learning_rate": 3.022625223328476e-06,
      "loss": 0.7971313,
      "num_input_tokens_seen": 62382105,
      "step": 2905,
      "time_per_iteration": 3.6787290573120117
    },
    {
      "auxiliary_loss_clip": 0.01190399,
      "auxiliary_loss_mlp": 0.01032856,
      "balance_loss_clip": 1.05583966,
      "balance_loss_mlp": 1.02366543,
      "epoch": 0.34942584019719836,
      "flos": 22855023319680.0,
      "grad_norm": 2.657250919549311,
      "language_loss": 0.69455576,
      "learning_rate": 3.0219557016007723e-06,
      "loss": 0.71678835,
      "num_input_tokens_seen": 62402235,
      "step": 2906,
      "time_per_iteration": 2.6393113136291504
    },
    {
      "auxiliary_loss_clip": 0.01179451,
      "auxiliary_loss_mlp": 0.01035452,
      "balance_loss_clip": 1.05653214,
      "balance_loss_mlp": 1.02650571,
      "epoch": 0.3495460830878374,
      "flos": 24426043441920.0,
      "grad_norm": 1.8787776544060832,
      "language_loss": 0.69687843,
      "learning_rate": 3.021286024843202e-06,
      "loss": 0.71902746,
      "num_input_tokens_seen": 62420430,
      "step": 2907,
      "time_per_iteration": 2.7353363037109375
    },
    {
      "auxiliary_loss_clip": 0.01099926,
      "auxiliary_loss_mlp": 0.01001339,
      "balance_loss_clip": 1.02659392,
      "balance_loss_mlp": 0.99967051,
      "epoch": 0.3496663259784765,
      "flos": 70008749389440.0,
      "grad_norm": 1.072254078965234,
      "language_loss": 0.64745384,
      "learning_rate": 3.0206161931573526e-06,
      "loss": 0.66846651,
      "num_input_tokens_seen": 62472980,
      "step": 2908,
      "time_per_iteration": 3.146068811416626
    },
    {
      "auxiliary_loss_clip": 0.01163665,
      "auxiliary_loss_mlp": 0.01031917,
      "balance_loss_clip": 1.04975796,
      "balance_loss_mlp": 1.02343512,
      "epoch": 0.34978656886911563,
      "flos": 28692805322880.0,
      "grad_norm": 1.6110440033108082,
      "language_loss": 0.93010294,
      "learning_rate": 3.0199462066448388e-06,
      "loss": 0.95205879,
      "num_input_tokens_seen": 62495175,
      "step": 2909,
      "time_per_iteration": 2.770305633544922
    },
    {
      "auxiliary_loss_clip": 0.01186676,
      "auxiliary_loss_mlp": 0.01030429,
      "balance_loss_clip": 1.05657959,
      "balance_loss_mlp": 1.02156019,
      "epoch": 0.3499068117597547,
      "flos": 21142156389120.0,
      "grad_norm": 1.8147112413376805,
      "language_loss": 0.69623852,
      "learning_rate": 3.019276065407296e-06,
      "loss": 0.7184096,
      "num_input_tokens_seen": 62514295,
      "step": 2910,
      "time_per_iteration": 2.6421573162078857
    },
    {
      "auxiliary_loss_clip": 0.01141609,
      "auxiliary_loss_mlp": 0.01028789,
      "balance_loss_clip": 1.05036974,
      "balance_loss_mlp": 1.01972306,
      "epoch": 0.3500270546503938,
      "flos": 22782699285120.0,
      "grad_norm": 1.7737035787021966,
      "language_loss": 0.80629694,
      "learning_rate": 3.018605769546385e-06,
      "loss": 0.8280009,
      "num_input_tokens_seen": 62534850,
      "step": 2911,
      "time_per_iteration": 2.8473026752471924
    },
    {
      "auxiliary_loss_clip": 0.01179569,
      "auxiliary_loss_mlp": 0.01035185,
      "balance_loss_clip": 1.05129838,
      "balance_loss_mlp": 1.02572536,
      "epoch": 0.3501472975410329,
      "flos": 22894058424960.0,
      "grad_norm": 2.1656836444925,
      "language_loss": 0.79656351,
      "learning_rate": 3.017935319163788e-06,
      "loss": 0.81871098,
      "num_input_tokens_seen": 62553810,
      "step": 2912,
      "time_per_iteration": 2.706317901611328
    },
    {
      "auxiliary_loss_clip": 0.01183612,
      "auxiliary_loss_mlp": 0.01030637,
      "balance_loss_clip": 1.05419207,
      "balance_loss_mlp": 1.01988411,
      "epoch": 0.35026754043167196,
      "flos": 25446588658560.0,
      "grad_norm": 1.8699807260351646,
      "language_loss": 0.70505321,
      "learning_rate": 3.017264714361213e-06,
      "loss": 0.72719568,
      "num_input_tokens_seen": 62573460,
      "step": 2913,
      "time_per_iteration": 2.6591193675994873
    },
    {
      "auxiliary_loss_clip": 0.01165466,
      "auxiliary_loss_mlp": 0.0076402,
      "balance_loss_clip": 1.05277324,
      "balance_loss_mlp": 1.00029826,
      "epoch": 0.3503877833223111,
      "flos": 19573757959680.0,
      "grad_norm": 2.080590051632714,
      "language_loss": 0.82362032,
      "learning_rate": 3.016593955240389e-06,
      "loss": 0.84291518,
      "num_input_tokens_seen": 62592150,
      "step": 2914,
      "time_per_iteration": 2.699784994125366
    },
    {
      "auxiliary_loss_clip": 0.01081561,
      "auxiliary_loss_mlp": 0.01001596,
      "balance_loss_clip": 1.02177024,
      "balance_loss_mlp": 0.99977189,
      "epoch": 0.3505080262129502,
      "flos": 65072075880960.0,
      "grad_norm": 0.8233668207484838,
      "language_loss": 0.63670266,
      "learning_rate": 3.015923041903071e-06,
      "loss": 0.65753424,
      "num_input_tokens_seen": 62658275,
      "step": 2915,
      "time_per_iteration": 3.2850747108459473
    },
    {
      "auxiliary_loss_clip": 0.0118212,
      "auxiliary_loss_mlp": 0.01035759,
      "balance_loss_clip": 1.05563402,
      "balance_loss_mlp": 1.02715778,
      "epoch": 0.35062826910358924,
      "flos": 29314562768640.0,
      "grad_norm": 2.2978005890296314,
      "language_loss": 0.83528304,
      "learning_rate": 3.0152519744510347e-06,
      "loss": 0.85746181,
      "num_input_tokens_seen": 62678075,
      "step": 2916,
      "time_per_iteration": 2.82252836227417
    },
    {
      "auxiliary_loss_clip": 0.01154689,
      "auxiliary_loss_mlp": 0.01028916,
      "balance_loss_clip": 1.05044389,
      "balance_loss_mlp": 1.02029085,
      "epoch": 0.35074851199422835,
      "flos": 23987717775360.0,
      "grad_norm": 2.448849498275923,
      "language_loss": 0.83083051,
      "learning_rate": 3.014580752986081e-06,
      "loss": 0.85266656,
      "num_input_tokens_seen": 62696950,
      "step": 2917,
      "time_per_iteration": 2.690643787384033
    },
    {
      "auxiliary_loss_clip": 0.01141484,
      "auxiliary_loss_mlp": 0.01035136,
      "balance_loss_clip": 1.04971433,
      "balance_loss_mlp": 1.02630806,
      "epoch": 0.3508687548848674,
      "flos": 15224436668160.0,
      "grad_norm": 1.9955559047212241,
      "language_loss": 0.78702056,
      "learning_rate": 3.0139093776100345e-06,
      "loss": 0.80878675,
      "num_input_tokens_seen": 62713540,
      "step": 2918,
      "time_per_iteration": 2.6704654693603516
    },
    {
      "auxiliary_loss_clip": 0.01194166,
      "auxiliary_loss_mlp": 0.01031172,
      "balance_loss_clip": 1.05414379,
      "balance_loss_mlp": 1.02225506,
      "epoch": 0.3509889977755065,
      "flos": 21361750185600.0,
      "grad_norm": 1.9795531191064755,
      "language_loss": 0.75727868,
      "learning_rate": 3.013237848424741e-06,
      "loss": 0.77953213,
      "num_input_tokens_seen": 62732925,
      "step": 2919,
      "time_per_iteration": 2.612204074859619
    },
    {
      "auxiliary_loss_clip": 0.01173285,
      "auxiliary_loss_mlp": 0.01028415,
      "balance_loss_clip": 1.05589366,
      "balance_loss_mlp": 1.01943886,
      "epoch": 0.35110924066614563,
      "flos": 19135360465920.0,
      "grad_norm": 3.354863537911646,
      "language_loss": 0.74948627,
      "learning_rate": 3.012566165532072e-06,
      "loss": 0.77150327,
      "num_input_tokens_seen": 62751715,
      "step": 2920,
      "time_per_iteration": 2.6619694232940674
    },
    {
      "auxiliary_loss_clip": 0.01132317,
      "auxiliary_loss_mlp": 0.01037597,
      "balance_loss_clip": 1.04908228,
      "balance_loss_mlp": 1.02813208,
      "epoch": 0.3512294835567847,
      "flos": 21980885938560.0,
      "grad_norm": 2.1243630269617655,
      "language_loss": 0.76251388,
      "learning_rate": 3.0118943290339207e-06,
      "loss": 0.78421307,
      "num_input_tokens_seen": 62771925,
      "step": 2921,
      "time_per_iteration": 2.761431932449341
    },
    {
      "auxiliary_loss_clip": 0.01139551,
      "auxiliary_loss_mlp": 0.01036488,
      "balance_loss_clip": 1.04478097,
      "balance_loss_mlp": 1.02693975,
      "epoch": 0.3513497264474238,
      "flos": 17817294896640.0,
      "grad_norm": 1.8750901537890243,
      "language_loss": 0.68380392,
      "learning_rate": 3.011222339032204e-06,
      "loss": 0.70556432,
      "num_input_tokens_seen": 62790075,
      "step": 2922,
      "time_per_iteration": 3.6116373538970947
    },
    {
      "auxiliary_loss_clip": 0.01197587,
      "auxiliary_loss_mlp": 0.01037314,
      "balance_loss_clip": 1.05790448,
      "balance_loss_mlp": 1.0283556,
      "epoch": 0.3514699693380629,
      "flos": 26943417239040.0,
      "grad_norm": 1.9635529349910479,
      "language_loss": 0.69381917,
      "learning_rate": 3.0105501956288626e-06,
      "loss": 0.71616822,
      "num_input_tokens_seen": 62810545,
      "step": 2923,
      "time_per_iteration": 2.6699538230895996
    },
    {
      "auxiliary_loss_clip": 0.01185471,
      "auxiliary_loss_mlp": 0.01030773,
      "balance_loss_clip": 1.05320644,
      "balance_loss_mlp": 1.02208877,
      "epoch": 0.35159021222870196,
      "flos": 15267565923840.0,
      "grad_norm": 2.0745590666074114,
      "language_loss": 0.72743207,
      "learning_rate": 3.0098778989258602e-06,
      "loss": 0.74959445,
      "num_input_tokens_seen": 62829155,
      "step": 2924,
      "time_per_iteration": 2.6024160385131836
    },
    {
      "auxiliary_loss_clip": 0.01145722,
      "auxiliary_loss_mlp": 0.01028771,
      "balance_loss_clip": 1.04951608,
      "balance_loss_mlp": 1.01985979,
      "epoch": 0.35171045511934107,
      "flos": 13984154000640.0,
      "grad_norm": 3.1909506649341934,
      "language_loss": 0.88458627,
      "learning_rate": 3.009205449025183e-06,
      "loss": 0.90633118,
      "num_input_tokens_seen": 62845350,
      "step": 2925,
      "time_per_iteration": 2.6759209632873535
    },
    {
      "auxiliary_loss_clip": 0.01146602,
      "auxiliary_loss_mlp": 0.01030287,
      "balance_loss_clip": 1.04666054,
      "balance_loss_mlp": 1.02137041,
      "epoch": 0.3518306980099802,
      "flos": 14283434119680.0,
      "grad_norm": 1.9159730601500395,
      "language_loss": 0.63191628,
      "learning_rate": 3.008532846028842e-06,
      "loss": 0.65368521,
      "num_input_tokens_seen": 62862110,
      "step": 2926,
      "time_per_iteration": 2.6243698596954346
    },
    {
      "auxiliary_loss_clip": 0.01198938,
      "auxiliary_loss_mlp": 0.01038394,
      "balance_loss_clip": 1.05803025,
      "balance_loss_mlp": 1.02858901,
      "epoch": 0.35195094090061924,
      "flos": 27052872958080.0,
      "grad_norm": 3.1806536793345406,
      "language_loss": 0.71862161,
      "learning_rate": 3.0078600900388694e-06,
      "loss": 0.74099493,
      "num_input_tokens_seen": 62882415,
      "step": 2927,
      "time_per_iteration": 3.6043879985809326
    },
    {
      "auxiliary_loss_clip": 0.01140422,
      "auxiliary_loss_mlp": 0.01034156,
      "balance_loss_clip": 1.04548383,
      "balance_loss_mlp": 1.02478635,
      "epoch": 0.35207118379125835,
      "flos": 25629266252160.0,
      "grad_norm": 3.3311178340406404,
      "language_loss": 0.74353015,
      "learning_rate": 3.007187181157323e-06,
      "loss": 0.76527596,
      "num_input_tokens_seen": 62902425,
      "step": 2928,
      "time_per_iteration": 2.8107335567474365
    },
    {
      "auxiliary_loss_clip": 0.01110793,
      "auxiliary_loss_mlp": 0.01042651,
      "balance_loss_clip": 1.04371452,
      "balance_loss_mlp": 1.03344214,
      "epoch": 0.35219142668189746,
      "flos": 18004713085440.0,
      "grad_norm": 2.867851687076375,
      "language_loss": 0.6805799,
      "learning_rate": 3.006514119486282e-06,
      "loss": 0.70211434,
      "num_input_tokens_seen": 62919255,
      "step": 2929,
      "time_per_iteration": 2.7337088584899902
    },
    {
      "auxiliary_loss_clip": 0.01141407,
      "auxiliary_loss_mlp": 0.01028597,
      "balance_loss_clip": 1.04712343,
      "balance_loss_mlp": 1.01941192,
      "epoch": 0.3523116695725365,
      "flos": 14028109269120.0,
      "grad_norm": 1.914410051371642,
      "language_loss": 0.70152932,
      "learning_rate": 3.005840905127849e-06,
      "loss": 0.72322941,
      "num_input_tokens_seen": 62936160,
      "step": 2930,
      "time_per_iteration": 3.568060874938965
    },
    {
      "auxiliary_loss_clip": 0.01196596,
      "auxiliary_loss_mlp": 0.01035767,
      "balance_loss_clip": 1.05724788,
      "balance_loss_mlp": 1.02614713,
      "epoch": 0.3524319124631756,
      "flos": 21433966479360.0,
      "grad_norm": 3.00388824873739,
      "language_loss": 0.867975,
      "learning_rate": 3.0051675381841516e-06,
      "loss": 0.8902986,
      "num_input_tokens_seen": 62953470,
      "step": 2931,
      "time_per_iteration": 3.5451879501342773
    },
    {
      "auxiliary_loss_clip": 0.01108099,
      "auxiliary_loss_mlp": 0.00764991,
      "balance_loss_clip": 1.04323912,
      "balance_loss_mlp": 1.0002172,
      "epoch": 0.3525521553538147,
      "flos": 26322773114880.0,
      "grad_norm": 1.5167175861539668,
      "language_loss": 0.76852381,
      "learning_rate": 3.0044940187573363e-06,
      "loss": 0.78725475,
      "num_input_tokens_seen": 62974480,
      "step": 2932,
      "time_per_iteration": 2.834671974182129
    },
    {
      "auxiliary_loss_clip": 0.01183304,
      "auxiliary_loss_mlp": 0.01028607,
      "balance_loss_clip": 1.05249751,
      "balance_loss_mlp": 1.01926136,
      "epoch": 0.3526723982444538,
      "flos": 21543314457600.0,
      "grad_norm": 2.0190965462306574,
      "language_loss": 0.65252453,
      "learning_rate": 3.003820346949578e-06,
      "loss": 0.67464364,
      "num_input_tokens_seen": 62992560,
      "step": 2933,
      "time_per_iteration": 2.703700065612793
    },
    {
      "auxiliary_loss_clip": 0.01196393,
      "auxiliary_loss_mlp": 0.01032651,
      "balance_loss_clip": 1.05455184,
      "balance_loss_mlp": 1.02429485,
      "epoch": 0.3527926411350929,
      "flos": 23733649900800.0,
      "grad_norm": 2.3141155908921736,
      "language_loss": 0.79646766,
      "learning_rate": 3.003146522863071e-06,
      "loss": 0.81875813,
      "num_input_tokens_seen": 63013445,
      "step": 2934,
      "time_per_iteration": 2.6311252117156982
    },
    {
      "auxiliary_loss_clip": 0.01165958,
      "auxiliary_loss_mlp": 0.01031845,
      "balance_loss_clip": 1.05341005,
      "balance_loss_mlp": 1.02276671,
      "epoch": 0.35291288402573195,
      "flos": 30445461544320.0,
      "grad_norm": 2.5313947431030357,
      "language_loss": 0.85920423,
      "learning_rate": 3.0024725466000345e-06,
      "loss": 0.88118225,
      "num_input_tokens_seen": 63033400,
      "step": 2935,
      "time_per_iteration": 2.7431368827819824
    },
    {
      "auxiliary_loss_clip": 0.01180677,
      "auxiliary_loss_mlp": 0.01039198,
      "balance_loss_clip": 1.05456233,
      "balance_loss_mlp": 1.03026891,
      "epoch": 0.35303312691637107,
      "flos": 23112179763840.0,
      "grad_norm": 1.757103577465668,
      "language_loss": 0.78776145,
      "learning_rate": 3.0017984182627087e-06,
      "loss": 0.80996013,
      "num_input_tokens_seen": 63052725,
      "step": 2936,
      "time_per_iteration": 2.6730575561523438
    },
    {
      "auxiliary_loss_clip": 0.01149331,
      "auxiliary_loss_mlp": 0.00764209,
      "balance_loss_clip": 1.04834092,
      "balance_loss_mlp": 1.00029624,
      "epoch": 0.3531533698070102,
      "flos": 21835699165440.0,
      "grad_norm": 2.0160478149532395,
      "language_loss": 0.82409286,
      "learning_rate": 3.00112413795336e-06,
      "loss": 0.84322822,
      "num_input_tokens_seen": 63072560,
      "step": 2937,
      "time_per_iteration": 2.794373035430908
    },
    {
      "auxiliary_loss_clip": 0.01163917,
      "auxiliary_loss_mlp": 0.01029182,
      "balance_loss_clip": 1.04769254,
      "balance_loss_mlp": 1.01972914,
      "epoch": 0.35327361269764923,
      "flos": 15778969810560.0,
      "grad_norm": 2.2810593737799554,
      "language_loss": 0.80245125,
      "learning_rate": 3.000449705774275e-06,
      "loss": 0.82438231,
      "num_input_tokens_seen": 63090800,
      "step": 2938,
      "time_per_iteration": 2.676865339279175
    },
    {
      "auxiliary_loss_clip": 0.01185927,
      "auxiliary_loss_mlp": 0.01032865,
      "balance_loss_clip": 1.05614805,
      "balance_loss_mlp": 1.02408528,
      "epoch": 0.35339385558828834,
      "flos": 22090413484800.0,
      "grad_norm": 1.8500346653595896,
      "language_loss": 0.71661198,
      "learning_rate": 2.9997751218277654e-06,
      "loss": 0.73879993,
      "num_input_tokens_seen": 63108955,
      "step": 2939,
      "time_per_iteration": 2.6270527839660645
    },
    {
      "auxiliary_loss_clip": 0.01197623,
      "auxiliary_loss_mlp": 0.01029208,
      "balance_loss_clip": 1.05613792,
      "balance_loss_mlp": 1.02003503,
      "epoch": 0.35351409847892745,
      "flos": 24165008328960.0,
      "grad_norm": 3.5706132456323463,
      "language_loss": 0.78341234,
      "learning_rate": 2.999100386216166e-06,
      "loss": 0.80568075,
      "num_input_tokens_seen": 63127895,
      "step": 2940,
      "time_per_iteration": 2.629265546798706
    },
    {
      "auxiliary_loss_clip": 0.01167449,
      "auxiliary_loss_mlp": 0.01030479,
      "balance_loss_clip": 1.05194783,
      "balance_loss_mlp": 1.02193809,
      "epoch": 0.3536343413695665,
      "flos": 27052298340480.0,
      "grad_norm": 2.098110009771241,
      "language_loss": 0.74429965,
      "learning_rate": 2.998425499041831e-06,
      "loss": 0.76627886,
      "num_input_tokens_seen": 63148410,
      "step": 2941,
      "time_per_iteration": 2.72666335105896
    },
    {
      "auxiliary_loss_clip": 0.01081388,
      "auxiliary_loss_mlp": 0.01003662,
      "balance_loss_clip": 1.02041698,
      "balance_loss_mlp": 1.00193357,
      "epoch": 0.3537545842602056,
      "flos": 65991066370560.0,
      "grad_norm": 1.2687750889226708,
      "language_loss": 0.64489585,
      "learning_rate": 2.997750460407142e-06,
      "loss": 0.66574639,
      "num_input_tokens_seen": 63209765,
      "step": 2942,
      "time_per_iteration": 3.2771735191345215
    },
    {
      "auxiliary_loss_clip": 0.01156593,
      "auxiliary_loss_mlp": 0.01027231,
      "balance_loss_clip": 1.04961658,
      "balance_loss_mlp": 1.01791453,
      "epoch": 0.35387482715084473,
      "flos": 18436897526400.0,
      "grad_norm": 2.1614805445511625,
      "language_loss": 0.69833148,
      "learning_rate": 2.997075270414501e-06,
      "loss": 0.72016966,
      "num_input_tokens_seen": 63226980,
      "step": 2943,
      "time_per_iteration": 2.7233757972717285
    },
    {
      "auxiliary_loss_clip": 0.01070333,
      "auxiliary_loss_mlp": 0.01004111,
      "balance_loss_clip": 1.02097201,
      "balance_loss_mlp": 1.0024178,
      "epoch": 0.3539950700414838,
      "flos": 65588579498880.0,
      "grad_norm": 0.7090042950016123,
      "language_loss": 0.5773083,
      "learning_rate": 2.9963999291663347e-06,
      "loss": 0.59805274,
      "num_input_tokens_seen": 63292760,
      "step": 2944,
      "time_per_iteration": 3.246868848800659
    },
    {
      "auxiliary_loss_clip": 0.01138095,
      "auxiliary_loss_mlp": 0.0103187,
      "balance_loss_clip": 1.05094242,
      "balance_loss_mlp": 1.0227983,
      "epoch": 0.3541153129321229,
      "flos": 20521655919360.0,
      "grad_norm": 3.679678595111307,
      "language_loss": 0.74211258,
      "learning_rate": 2.9957244367650915e-06,
      "loss": 0.76381218,
      "num_input_tokens_seen": 63309005,
      "step": 2945,
      "time_per_iteration": 2.727323055267334
    },
    {
      "auxiliary_loss_clip": 0.01135532,
      "auxiliary_loss_mlp": 0.01034962,
      "balance_loss_clip": 1.0508635,
      "balance_loss_mlp": 1.02619362,
      "epoch": 0.354235555822762,
      "flos": 19573578391680.0,
      "grad_norm": 1.9885413575104811,
      "language_loss": 0.83715379,
      "learning_rate": 2.9950487933132425e-06,
      "loss": 0.8588587,
      "num_input_tokens_seen": 63326420,
      "step": 2946,
      "time_per_iteration": 2.7159605026245117
    },
    {
      "auxiliary_loss_clip": 0.01185076,
      "auxiliary_loss_mlp": 0.01035365,
      "balance_loss_clip": 1.05432892,
      "balance_loss_mlp": 1.02658498,
      "epoch": 0.35435579871340106,
      "flos": 20777268078720.0,
      "grad_norm": 2.677571011943611,
      "language_loss": 0.71448016,
      "learning_rate": 2.994372998913283e-06,
      "loss": 0.73668456,
      "num_input_tokens_seen": 63344925,
      "step": 2947,
      "time_per_iteration": 2.6551353931427
    },
    {
      "auxiliary_loss_clip": 0.0116804,
      "auxiliary_loss_mlp": 0.01033449,
      "balance_loss_clip": 1.05508173,
      "balance_loss_mlp": 1.02469862,
      "epoch": 0.35447604160404017,
      "flos": 23951807153280.0,
      "grad_norm": 2.807209481944425,
      "language_loss": 0.62374508,
      "learning_rate": 2.99369705366773e-06,
      "loss": 0.64575994,
      "num_input_tokens_seen": 63365170,
      "step": 2948,
      "time_per_iteration": 3.621102809906006
    },
    {
      "auxiliary_loss_clip": 0.01163733,
      "auxiliary_loss_mlp": 0.0103461,
      "balance_loss_clip": 1.05232048,
      "balance_loss_mlp": 1.02550268,
      "epoch": 0.3545962844946792,
      "flos": 23435662671360.0,
      "grad_norm": 1.8980530493076788,
      "language_loss": 0.82112199,
      "learning_rate": 2.9930209576791244e-06,
      "loss": 0.84310544,
      "num_input_tokens_seen": 63383645,
      "step": 2949,
      "time_per_iteration": 2.7469656467437744
    },
    {
      "auxiliary_loss_clip": 0.01177694,
      "auxiliary_loss_mlp": 0.01030058,
      "balance_loss_clip": 1.05332756,
      "balance_loss_mlp": 1.0206759,
      "epoch": 0.35471652738531834,
      "flos": 22085134185600.0,
      "grad_norm": 2.4523922925816857,
      "language_loss": 0.63599706,
      "learning_rate": 2.9923447110500285e-06,
      "loss": 0.65807462,
      "num_input_tokens_seen": 63402390,
      "step": 2950,
      "time_per_iteration": 2.61568021774292
    },
    {
      "auxiliary_loss_clip": 0.01170972,
      "auxiliary_loss_mlp": 0.01032445,
      "balance_loss_clip": 1.05293775,
      "balance_loss_mlp": 1.02368879,
      "epoch": 0.35483677027595745,
      "flos": 27341881787520.0,
      "grad_norm": 1.6251560381479642,
      "language_loss": 0.75455844,
      "learning_rate": 2.9916683138830295e-06,
      "loss": 0.77659267,
      "num_input_tokens_seen": 63423055,
      "step": 2951,
      "time_per_iteration": 2.7109241485595703
    },
    {
      "auxiliary_loss_clip": 0.01164807,
      "auxiliary_loss_mlp": 0.01030045,
      "balance_loss_clip": 1.05272043,
      "balance_loss_mlp": 1.02131343,
      "epoch": 0.3549570131665965,
      "flos": 13516166678400.0,
      "grad_norm": 1.9315953052432235,
      "language_loss": 0.81155837,
      "learning_rate": 2.9909917662807353e-06,
      "loss": 0.83350688,
      "num_input_tokens_seen": 63440855,
      "step": 2952,
      "time_per_iteration": 3.6108484268188477
    },
    {
      "auxiliary_loss_clip": 0.0117927,
      "auxiliary_loss_mlp": 0.01034896,
      "balance_loss_clip": 1.05400109,
      "balance_loss_mlp": 1.02633107,
      "epoch": 0.3550772560572356,
      "flos": 20887549810560.0,
      "grad_norm": 2.111690326077062,
      "language_loss": 0.69758821,
      "learning_rate": 2.9903150683457783e-06,
      "loss": 0.7197299,
      "num_input_tokens_seen": 63459400,
      "step": 2953,
      "time_per_iteration": 2.5993800163269043
    },
    {
      "auxiliary_loss_clip": 0.01165926,
      "auxiliary_loss_mlp": 0.01033871,
      "balance_loss_clip": 1.04934025,
      "balance_loss_mlp": 1.02436352,
      "epoch": 0.3551974989478747,
      "flos": 20194042947840.0,
      "grad_norm": 2.3535776781244695,
      "language_loss": 0.6548934,
      "learning_rate": 2.9896382201808126e-06,
      "loss": 0.67689139,
      "num_input_tokens_seen": 63476800,
      "step": 2954,
      "time_per_iteration": 2.654672145843506
    },
    {
      "auxiliary_loss_clip": 0.01196657,
      "auxiliary_loss_mlp": 0.01027703,
      "balance_loss_clip": 1.05602098,
      "balance_loss_mlp": 1.01869082,
      "epoch": 0.3553177418385138,
      "flos": 19828831415040.0,
      "grad_norm": 2.512205705408409,
      "language_loss": 0.81470621,
      "learning_rate": 2.988961221888516e-06,
      "loss": 0.83694983,
      "num_input_tokens_seen": 63493475,
      "step": 2955,
      "time_per_iteration": 2.565715789794922
    },
    {
      "auxiliary_loss_clip": 0.01143441,
      "auxiliary_loss_mlp": 0.01031439,
      "balance_loss_clip": 1.04883492,
      "balance_loss_mlp": 1.02312374,
      "epoch": 0.3554379847291529,
      "flos": 14829132516480.0,
      "grad_norm": 3.256802443448596,
      "language_loss": 0.7898249,
      "learning_rate": 2.988284073571589e-06,
      "loss": 0.81157374,
      "num_input_tokens_seen": 63509560,
      "step": 2956,
      "time_per_iteration": 3.601400852203369
    },
    {
      "auxiliary_loss_clip": 0.01182662,
      "auxiliary_loss_mlp": 0.00764378,
      "balance_loss_clip": 1.05432057,
      "balance_loss_mlp": 1.00025332,
      "epoch": 0.355558227619792,
      "flos": 20485350247680.0,
      "grad_norm": 2.333207943624781,
      "language_loss": 0.73181719,
      "learning_rate": 2.9876067753327528e-06,
      "loss": 0.75128758,
      "num_input_tokens_seen": 63527290,
      "step": 2957,
      "time_per_iteration": 3.508537530899048
    },
    {
      "auxiliary_loss_clip": 0.01185728,
      "auxiliary_loss_mlp": 0.01031117,
      "balance_loss_clip": 1.05342591,
      "balance_loss_mlp": 1.02219427,
      "epoch": 0.35567847051043106,
      "flos": 37663613256960.0,
      "grad_norm": 2.108432696218772,
      "language_loss": 0.80613005,
      "learning_rate": 2.986929327274754e-06,
      "loss": 0.82829845,
      "num_input_tokens_seen": 63547870,
      "step": 2958,
      "time_per_iteration": 2.7309229373931885
    },
    {
      "auxiliary_loss_clip": 0.0117729,
      "auxiliary_loss_mlp": 0.01031246,
      "balance_loss_clip": 1.05306399,
      "balance_loss_mlp": 1.02237689,
      "epoch": 0.35579871340107017,
      "flos": 26943058103040.0,
      "grad_norm": 1.6750687991197504,
      "language_loss": 0.78900027,
      "learning_rate": 2.9862517295003617e-06,
      "loss": 0.81108564,
      "num_input_tokens_seen": 63568285,
      "step": 2959,
      "time_per_iteration": 2.709367513656616
    },
    {
      "auxiliary_loss_clip": 0.01149279,
      "auxiliary_loss_mlp": 0.01033383,
      "balance_loss_clip": 1.04696083,
      "balance_loss_mlp": 1.02463949,
      "epoch": 0.3559189562917093,
      "flos": 28293335193600.0,
      "grad_norm": 3.433574473161194,
      "language_loss": 0.72887468,
      "learning_rate": 2.9855739821123654e-06,
      "loss": 0.75070137,
      "num_input_tokens_seen": 63589865,
      "step": 2960,
      "time_per_iteration": 2.8028271198272705
    },
    {
      "auxiliary_loss_clip": 0.01179244,
      "auxiliary_loss_mlp": 0.01035552,
      "balance_loss_clip": 1.05487168,
      "balance_loss_mlp": 1.02702272,
      "epoch": 0.35603919918234833,
      "flos": 25664063552640.0,
      "grad_norm": 2.2144469309302535,
      "language_loss": 0.82033348,
      "learning_rate": 2.98489608521358e-06,
      "loss": 0.84248149,
      "num_input_tokens_seen": 63609805,
      "step": 2961,
      "time_per_iteration": 2.713576078414917
    },
    {
      "auxiliary_loss_clip": 0.01187003,
      "auxiliary_loss_mlp": 0.00763761,
      "balance_loss_clip": 1.05611074,
      "balance_loss_mlp": 1.00027084,
      "epoch": 0.35615944207298744,
      "flos": 23000856537600.0,
      "grad_norm": 2.4947540724011756,
      "language_loss": 0.79494023,
      "learning_rate": 2.9842180389068425e-06,
      "loss": 0.81444788,
      "num_input_tokens_seen": 63627115,
      "step": 2962,
      "time_per_iteration": 2.6376824378967285
    },
    {
      "auxiliary_loss_clip": 0.01065162,
      "auxiliary_loss_mlp": 0.01012221,
      "balance_loss_clip": 1.02843785,
      "balance_loss_mlp": 1.01039755,
      "epoch": 0.35627968496362655,
      "flos": 68251283723520.0,
      "grad_norm": 0.766997578516349,
      "language_loss": 0.59225297,
      "learning_rate": 2.98353984329501e-06,
      "loss": 0.61302686,
      "num_input_tokens_seen": 63691460,
      "step": 2963,
      "time_per_iteration": 3.315363645553589
    },
    {
      "auxiliary_loss_clip": 0.01166103,
      "auxiliary_loss_mlp": 0.01032488,
      "balance_loss_clip": 1.0536716,
      "balance_loss_mlp": 1.02261722,
      "epoch": 0.3563999278542656,
      "flos": 22641714403200.0,
      "grad_norm": 1.663735700407347,
      "language_loss": 0.7056365,
      "learning_rate": 2.982861498480965e-06,
      "loss": 0.72762239,
      "num_input_tokens_seen": 63713840,
      "step": 2964,
      "time_per_iteration": 2.7815921306610107
    },
    {
      "auxiliary_loss_clip": 0.01146039,
      "auxiliary_loss_mlp": 0.01025731,
      "balance_loss_clip": 1.0475843,
      "balance_loss_mlp": 1.01711786,
      "epoch": 0.3565201707449047,
      "flos": 25952533678080.0,
      "grad_norm": 1.7013612272510852,
      "language_loss": 0.82699519,
      "learning_rate": 2.9821830045676122e-06,
      "loss": 0.84871286,
      "num_input_tokens_seen": 63733540,
      "step": 2965,
      "time_per_iteration": 2.7753610610961914
    },
    {
      "auxiliary_loss_clip": 0.01199598,
      "auxiliary_loss_mlp": 0.01033445,
      "balance_loss_clip": 1.05775881,
      "balance_loss_mlp": 1.02360415,
      "epoch": 0.3566404136355438,
      "flos": 28475725478400.0,
      "grad_norm": 1.805597824194273,
      "language_loss": 0.72968823,
      "learning_rate": 2.9815043616578793e-06,
      "loss": 0.75201863,
      "num_input_tokens_seen": 63754335,
      "step": 2966,
      "time_per_iteration": 2.6762287616729736
    },
    {
      "auxiliary_loss_clip": 0.01148386,
      "auxiliary_loss_mlp": 0.01031952,
      "balance_loss_clip": 1.04961085,
      "balance_loss_mlp": 1.02366066,
      "epoch": 0.3567606565261829,
      "flos": 38363117690880.0,
      "grad_norm": 2.0328047610945768,
      "language_loss": 0.77292073,
      "learning_rate": 2.9808255698547145e-06,
      "loss": 0.79472411,
      "num_input_tokens_seen": 63777135,
      "step": 2967,
      "time_per_iteration": 2.8567299842834473
    },
    {
      "auxiliary_loss_clip": 0.01182731,
      "auxiliary_loss_mlp": 0.01030138,
      "balance_loss_clip": 1.05764782,
      "balance_loss_mlp": 1.02159083,
      "epoch": 0.356880899416822,
      "flos": 21981029592960.0,
      "grad_norm": 3.096418581797977,
      "language_loss": 0.7966423,
      "learning_rate": 2.9801466292610913e-06,
      "loss": 0.818771,
      "num_input_tokens_seen": 63797020,
      "step": 2968,
      "time_per_iteration": 2.7444419860839844
    },
    {
      "auxiliary_loss_clip": 0.01180556,
      "auxiliary_loss_mlp": 0.01026527,
      "balance_loss_clip": 1.05362666,
      "balance_loss_mlp": 1.0179919,
      "epoch": 0.35700114230746105,
      "flos": 18989132198400.0,
      "grad_norm": 2.1545782062715655,
      "language_loss": 0.81112742,
      "learning_rate": 2.979467539980003e-06,
      "loss": 0.83319825,
      "num_input_tokens_seen": 63813810,
      "step": 2969,
      "time_per_iteration": 2.695913791656494
    },
    {
      "auxiliary_loss_clip": 0.01183702,
      "auxiliary_loss_mlp": 0.01033977,
      "balance_loss_clip": 1.05573773,
      "balance_loss_mlp": 1.02495241,
      "epoch": 0.35712138519810016,
      "flos": 19756112330880.0,
      "grad_norm": 1.9876642146353385,
      "language_loss": 0.76902425,
      "learning_rate": 2.978788302114468e-06,
      "loss": 0.79120111,
      "num_input_tokens_seen": 63830925,
      "step": 2970,
      "time_per_iteration": 2.82220458984375
    },
    {
      "auxiliary_loss_clip": 0.01177019,
      "auxiliary_loss_mlp": 0.01041491,
      "balance_loss_clip": 1.05295455,
      "balance_loss_mlp": 1.0319066,
      "epoch": 0.35724162808873927,
      "flos": 35183012008320.0,
      "grad_norm": 2.194411800081643,
      "language_loss": 0.83444369,
      "learning_rate": 2.9781089157675255e-06,
      "loss": 0.85662878,
      "num_input_tokens_seen": 63849385,
      "step": 2971,
      "time_per_iteration": 2.781601667404175
    },
    {
      "auxiliary_loss_clip": 0.01178638,
      "auxiliary_loss_mlp": 0.0103105,
      "balance_loss_clip": 1.05693793,
      "balance_loss_mlp": 1.0227468,
      "epoch": 0.3573618709793783,
      "flos": 25556726736000.0,
      "grad_norm": 1.5909647843914085,
      "language_loss": 0.88462132,
      "learning_rate": 2.977429381042238e-06,
      "loss": 0.90671813,
      "num_input_tokens_seen": 63870060,
      "step": 2972,
      "time_per_iteration": 2.805368423461914
    },
    {
      "auxiliary_loss_clip": 0.01168981,
      "auxiliary_loss_mlp": 0.01029465,
      "balance_loss_clip": 1.05505311,
      "balance_loss_mlp": 1.02157331,
      "epoch": 0.35748211387001744,
      "flos": 29132352051840.0,
      "grad_norm": 2.1520768308544045,
      "language_loss": 0.89103043,
      "learning_rate": 2.9767496980416913e-06,
      "loss": 0.91301489,
      "num_input_tokens_seen": 63889355,
      "step": 2973,
      "time_per_iteration": 2.72102427482605
    },
    {
      "auxiliary_loss_clip": 0.01158509,
      "auxiliary_loss_mlp": 0.01026882,
      "balance_loss_clip": 1.04836512,
      "balance_loss_mlp": 1.01720786,
      "epoch": 0.35760235676065655,
      "flos": 13954169122560.0,
      "grad_norm": 2.1066987669919244,
      "language_loss": 0.8104344,
      "learning_rate": 2.9760698668689914e-06,
      "loss": 0.83228832,
      "num_input_tokens_seen": 63905580,
      "step": 2974,
      "time_per_iteration": 2.713947296142578
    },
    {
      "auxiliary_loss_clip": 0.0117757,
      "auxiliary_loss_mlp": 0.01029941,
      "balance_loss_clip": 1.05274701,
      "balance_loss_mlp": 1.02176905,
      "epoch": 0.3577225996512956,
      "flos": 44018688977280.0,
      "grad_norm": 1.9084311372201959,
      "language_loss": 0.71604347,
      "learning_rate": 2.975389887627269e-06,
      "loss": 0.73811865,
      "num_input_tokens_seen": 63928180,
      "step": 2975,
      "time_per_iteration": 3.7302658557891846
    },
    {
      "auxiliary_loss_clip": 0.01154512,
      "auxiliary_loss_mlp": 0.01031504,
      "balance_loss_clip": 1.04933119,
      "balance_loss_mlp": 1.02321935,
      "epoch": 0.3578428425419347,
      "flos": 17055199013760.0,
      "grad_norm": 2.2951536408469964,
      "language_loss": 0.89737028,
      "learning_rate": 2.9747097604196764e-06,
      "loss": 0.91923046,
      "num_input_tokens_seen": 63944825,
      "step": 2976,
      "time_per_iteration": 2.669689893722534
    },
    {
      "auxiliary_loss_clip": 0.01050488,
      "auxiliary_loss_mlp": 0.01015512,
      "balance_loss_clip": 1.02361274,
      "balance_loss_mlp": 1.0138911,
      "epoch": 0.3579630854325738,
      "flos": 71676550707840.0,
      "grad_norm": 0.7033145679024358,
      "language_loss": 0.56614566,
      "learning_rate": 2.9740294853493875e-06,
      "loss": 0.5868057,
      "num_input_tokens_seen": 64016385,
      "step": 2977,
      "time_per_iteration": 3.570793867111206
    },
    {
      "auxiliary_loss_clip": 0.01142414,
      "auxiliary_loss_mlp": 0.01031721,
      "balance_loss_clip": 1.048558,
      "balance_loss_mlp": 1.02279806,
      "epoch": 0.3580833283232129,
      "flos": 25046651652480.0,
      "grad_norm": 2.7261054759323877,
      "language_loss": 0.6734587,
      "learning_rate": 2.9733490625196008e-06,
      "loss": 0.69520009,
      "num_input_tokens_seen": 64036245,
      "step": 2978,
      "time_per_iteration": 3.6809422969818115
    },
    {
      "auxiliary_loss_clip": 0.0114368,
      "auxiliary_loss_mlp": 0.01035235,
      "balance_loss_clip": 1.04972458,
      "balance_loss_mlp": 1.02704477,
      "epoch": 0.358203571213852,
      "flos": 13953127628160.0,
      "grad_norm": 3.003121916110987,
      "language_loss": 0.76546204,
      "learning_rate": 2.9726684920335353e-06,
      "loss": 0.78725111,
      "num_input_tokens_seen": 64054110,
      "step": 2979,
      "time_per_iteration": 2.688033103942871
    },
    {
      "auxiliary_loss_clip": 0.01194467,
      "auxiliary_loss_mlp": 0.00764242,
      "balance_loss_clip": 1.05461645,
      "balance_loss_mlp": 1.00023985,
      "epoch": 0.35832381410449105,
      "flos": 20302457172480.0,
      "grad_norm": 2.168891584958668,
      "language_loss": 0.82312334,
      "learning_rate": 2.971987773994432e-06,
      "loss": 0.84271044,
      "num_input_tokens_seen": 64070295,
      "step": 2980,
      "time_per_iteration": 2.6672730445861816
    },
    {
      "auxiliary_loss_clip": 0.01171287,
      "auxiliary_loss_mlp": 0.01030714,
      "balance_loss_clip": 1.05096018,
      "balance_loss_mlp": 1.02221394,
      "epoch": 0.35844405699513016,
      "flos": 16983234115200.0,
      "grad_norm": 1.9685711916746054,
      "language_loss": 0.83232069,
      "learning_rate": 2.9713069085055566e-06,
      "loss": 0.85434067,
      "num_input_tokens_seen": 64088605,
      "step": 2981,
      "time_per_iteration": 2.731266975402832
    },
    {
      "auxiliary_loss_clip": 0.01151413,
      "auxiliary_loss_mlp": 0.01024755,
      "balance_loss_clip": 1.05167234,
      "balance_loss_mlp": 1.01678002,
      "epoch": 0.35856429988576927,
      "flos": 23216858974080.0,
      "grad_norm": 1.9175138221577082,
      "language_loss": 0.79310322,
      "learning_rate": 2.9706258956701958e-06,
      "loss": 0.81486487,
      "num_input_tokens_seen": 64108595,
      "step": 2982,
      "time_per_iteration": 3.6448233127593994
    },
    {
      "auxiliary_loss_clip": 0.01179126,
      "auxiliary_loss_mlp": 0.01029428,
      "balance_loss_clip": 1.05255318,
      "balance_loss_mlp": 1.02003431,
      "epoch": 0.3586845427764083,
      "flos": 23034576430080.0,
      "grad_norm": 55.016705483691055,
      "language_loss": 0.77734053,
      "learning_rate": 2.9699447355916575e-06,
      "loss": 0.79942608,
      "num_input_tokens_seen": 64127405,
      "step": 2983,
      "time_per_iteration": 3.591041326522827
    },
    {
      "auxiliary_loss_clip": 0.01194474,
      "auxiliary_loss_mlp": 0.00763996,
      "balance_loss_clip": 1.056885,
      "balance_loss_mlp": 1.00018907,
      "epoch": 0.35880478566704743,
      "flos": 20010682995840.0,
      "grad_norm": 1.9935260306153408,
      "language_loss": 0.7410599,
      "learning_rate": 2.969263428373275e-06,
      "loss": 0.76064467,
      "num_input_tokens_seen": 64145755,
      "step": 2984,
      "time_per_iteration": 2.593135356903076
    },
    {
      "auxiliary_loss_clip": 0.01167368,
      "auxiliary_loss_mlp": 0.01032851,
      "balance_loss_clip": 1.05174839,
      "balance_loss_mlp": 1.02420235,
      "epoch": 0.35892502855768654,
      "flos": 13699095667200.0,
      "grad_norm": 2.173234855886229,
      "language_loss": 0.79472613,
      "learning_rate": 2.9685819741184007e-06,
      "loss": 0.81672835,
      "num_input_tokens_seen": 64164195,
      "step": 2985,
      "time_per_iteration": 2.657317638397217
    },
    {
      "auxiliary_loss_clip": 0.01145295,
      "auxiliary_loss_mlp": 0.01030979,
      "balance_loss_clip": 1.05099905,
      "balance_loss_mlp": 1.02275395,
      "epoch": 0.3590452714483256,
      "flos": 18114096977280.0,
      "grad_norm": 2.689639831757914,
      "language_loss": 0.68873572,
      "learning_rate": 2.967900372930411e-06,
      "loss": 0.71049845,
      "num_input_tokens_seen": 64182705,
      "step": 2986,
      "time_per_iteration": 2.7035906314849854
    },
    {
      "auxiliary_loss_clip": 0.01160683,
      "auxiliary_loss_mlp": 0.01033049,
      "balance_loss_clip": 1.05188048,
      "balance_loss_mlp": 1.0245676,
      "epoch": 0.3591655143389647,
      "flos": 17749352321280.0,
      "grad_norm": 2.321788593443789,
      "language_loss": 0.79679108,
      "learning_rate": 2.9672186249127046e-06,
      "loss": 0.81872839,
      "num_input_tokens_seen": 64202170,
      "step": 2987,
      "time_per_iteration": 2.6422650814056396
    },
    {
      "auxiliary_loss_clip": 0.01167314,
      "auxiliary_loss_mlp": 0.01032796,
      "balance_loss_clip": 1.05391419,
      "balance_loss_mlp": 1.02463579,
      "epoch": 0.3592857572296038,
      "flos": 25224409082880.0,
      "grad_norm": 1.9545723729178024,
      "language_loss": 0.79115868,
      "learning_rate": 2.9665367301687014e-06,
      "loss": 0.81315982,
      "num_input_tokens_seen": 64220415,
      "step": 2988,
      "time_per_iteration": 2.6516716480255127
    },
    {
      "auxiliary_loss_clip": 0.01158918,
      "auxiliary_loss_mlp": 0.0102366,
      "balance_loss_clip": 1.05003059,
      "balance_loss_mlp": 1.01517797,
      "epoch": 0.3594060001202429,
      "flos": 29384408764800.0,
      "grad_norm": 2.1339009421681054,
      "language_loss": 0.76974463,
      "learning_rate": 2.965854688801845e-06,
      "loss": 0.79157043,
      "num_input_tokens_seen": 64242475,
      "step": 2989,
      "time_per_iteration": 2.7399585247039795
    },
    {
      "auxiliary_loss_clip": 0.01176372,
      "auxiliary_loss_mlp": 0.01030739,
      "balance_loss_clip": 1.04959679,
      "balance_loss_mlp": 1.02210867,
      "epoch": 0.359526243010882,
      "flos": 17052900543360.0,
      "grad_norm": 1.7239232497674615,
      "language_loss": 0.76499951,
      "learning_rate": 2.9651725009156005e-06,
      "loss": 0.78707063,
      "num_input_tokens_seen": 64260220,
      "step": 2990,
      "time_per_iteration": 2.6978909969329834
    },
    {
      "auxiliary_loss_clip": 0.01155974,
      "auxiliary_loss_mlp": 0.01034352,
      "balance_loss_clip": 1.04845119,
      "balance_loss_mlp": 1.02578664,
      "epoch": 0.3596464859015211,
      "flos": 22965089569920.0,
      "grad_norm": 1.7576081413897597,
      "language_loss": 0.74503267,
      "learning_rate": 2.964490166613454e-06,
      "loss": 0.76693588,
      "num_input_tokens_seen": 64280145,
      "step": 2991,
      "time_per_iteration": 2.695620536804199
    },
    {
      "auxiliary_loss_clip": 0.01101402,
      "auxiliary_loss_mlp": 0.01002363,
      "balance_loss_clip": 1.02935791,
      "balance_loss_mlp": 1.00076604,
      "epoch": 0.35976672879216015,
      "flos": 54739462590720.0,
      "grad_norm": 0.7555452525745984,
      "language_loss": 0.57741505,
      "learning_rate": 2.963807685998917e-06,
      "loss": 0.59845275,
      "num_input_tokens_seen": 64336010,
      "step": 2992,
      "time_per_iteration": 3.0939486026763916
    },
    {
      "auxiliary_loss_clip": 0.01137982,
      "auxiliary_loss_mlp": 0.01030561,
      "balance_loss_clip": 1.04829085,
      "balance_loss_mlp": 1.02193642,
      "epoch": 0.35988697168279926,
      "flos": 43139020901760.0,
      "grad_norm": 1.6591960963864554,
      "language_loss": 0.78054154,
      "learning_rate": 2.9631250591755196e-06,
      "loss": 0.8022269,
      "num_input_tokens_seen": 64358725,
      "step": 2993,
      "time_per_iteration": 2.9499051570892334
    },
    {
      "auxiliary_loss_clip": 0.01162957,
      "auxiliary_loss_mlp": 0.01032594,
      "balance_loss_clip": 1.05399168,
      "balance_loss_mlp": 1.02428508,
      "epoch": 0.36000721457343837,
      "flos": 35845600239360.0,
      "grad_norm": 1.9813116716206676,
      "language_loss": 0.57557082,
      "learning_rate": 2.962442286246817e-06,
      "loss": 0.59752631,
      "num_input_tokens_seen": 64381555,
      "step": 2994,
      "time_per_iteration": 2.817528247833252
    },
    {
      "auxiliary_loss_clip": 0.01169461,
      "auxiliary_loss_mlp": 0.01032506,
      "balance_loss_clip": 1.05247378,
      "balance_loss_mlp": 1.02353525,
      "epoch": 0.3601274574640774,
      "flos": 18291100222080.0,
      "grad_norm": 1.6488867344931484,
      "language_loss": 0.69759023,
      "learning_rate": 2.9617593673163853e-06,
      "loss": 0.71960992,
      "num_input_tokens_seen": 64400375,
      "step": 2995,
      "time_per_iteration": 2.832857131958008
    },
    {
      "auxiliary_loss_clip": 0.01168339,
      "auxiliary_loss_mlp": 0.01024825,
      "balance_loss_clip": 1.05024576,
      "balance_loss_mlp": 1.01638508,
      "epoch": 0.36024770035471654,
      "flos": 13333955961600.0,
      "grad_norm": 2.20907951348623,
      "language_loss": 0.77256602,
      "learning_rate": 2.9610763024878216e-06,
      "loss": 0.79449767,
      "num_input_tokens_seen": 64415880,
      "step": 2996,
      "time_per_iteration": 2.816661834716797
    },
    {
      "auxiliary_loss_clip": 0.01159807,
      "auxiliary_loss_mlp": 0.01028538,
      "balance_loss_clip": 1.05028975,
      "balance_loss_mlp": 1.02045608,
      "epoch": 0.3603679432453556,
      "flos": 20267013427200.0,
      "grad_norm": 1.9168041613923466,
      "language_loss": 0.91684777,
      "learning_rate": 2.960393091864747e-06,
      "loss": 0.93873125,
      "num_input_tokens_seen": 64434260,
      "step": 2997,
      "time_per_iteration": 2.783811330795288
    },
    {
      "auxiliary_loss_clip": 0.01166248,
      "auxiliary_loss_mlp": 0.01030466,
      "balance_loss_clip": 1.05232882,
      "balance_loss_mlp": 1.02227664,
      "epoch": 0.3604881861359947,
      "flos": 22451135817600.0,
      "grad_norm": 1.7615305240322987,
      "language_loss": 0.74957097,
      "learning_rate": 2.959709735550804e-06,
      "loss": 0.77153814,
      "num_input_tokens_seen": 64453855,
      "step": 2998,
      "time_per_iteration": 2.726381778717041
    },
    {
      "auxiliary_loss_clip": 0.01139389,
      "auxiliary_loss_mlp": 0.01029253,
      "balance_loss_clip": 1.04740024,
      "balance_loss_mlp": 1.02052653,
      "epoch": 0.3606084290266338,
      "flos": 22054251467520.0,
      "grad_norm": 1.9018128265148717,
      "language_loss": 0.75972396,
      "learning_rate": 2.9590262336496575e-06,
      "loss": 0.7814104,
      "num_input_tokens_seen": 64473585,
      "step": 2999,
      "time_per_iteration": 2.885101795196533
    },
    {
      "auxiliary_loss_clip": 0.01148507,
      "auxiliary_loss_mlp": 0.01025989,
      "balance_loss_clip": 1.05280638,
      "balance_loss_mlp": 1.01771045,
      "epoch": 0.36072867191727287,
      "flos": 15632921111040.0,
      "grad_norm": 2.597915275242869,
      "language_loss": 0.85597467,
      "learning_rate": 2.9583425862649936e-06,
      "loss": 0.87771964,
      "num_input_tokens_seen": 64491720,
      "step": 3000,
      "time_per_iteration": 4.237358808517456
    },
    {
      "auxiliary_loss_clip": 0.01198027,
      "auxiliary_loss_mlp": 0.01037805,
      "balance_loss_clip": 1.05814552,
      "balance_loss_mlp": 1.02883434,
      "epoch": 0.360848914807912,
      "flos": 19677000625920.0,
      "grad_norm": 2.529947337355544,
      "language_loss": 0.74085289,
      "learning_rate": 2.9576587935005215e-06,
      "loss": 0.76321113,
      "num_input_tokens_seen": 64509800,
      "step": 3001,
      "time_per_iteration": 2.6573708057403564
    },
    {
      "auxiliary_loss_clip": 0.01181382,
      "auxiliary_loss_mlp": 0.01036996,
      "balance_loss_clip": 1.05431259,
      "balance_loss_mlp": 1.02744722,
      "epoch": 0.3609691576985511,
      "flos": 18877808972160.0,
      "grad_norm": 2.3283847168288596,
      "language_loss": 0.72103858,
      "learning_rate": 2.9569748554599713e-06,
      "loss": 0.74322236,
      "num_input_tokens_seen": 64525410,
      "step": 3002,
      "time_per_iteration": 2.660202980041504
    },
    {
      "auxiliary_loss_clip": 0.01164384,
      "auxiliary_loss_mlp": 0.01030584,
      "balance_loss_clip": 1.05124807,
      "balance_loss_mlp": 1.02226877,
      "epoch": 0.36108940058919015,
      "flos": 42224088648960.0,
      "grad_norm": 2.207089238317271,
      "language_loss": 0.73260808,
      "learning_rate": 2.956290772247097e-06,
      "loss": 0.75455773,
      "num_input_tokens_seen": 64544085,
      "step": 3003,
      "time_per_iteration": 2.917388439178467
    },
    {
      "auxiliary_loss_clip": 0.01130587,
      "auxiliary_loss_mlp": 0.01028938,
      "balance_loss_clip": 1.04983497,
      "balance_loss_mlp": 1.02092743,
      "epoch": 0.36120964347982926,
      "flos": 23185150243200.0,
      "grad_norm": 2.5290326974899964,
      "language_loss": 0.73434436,
      "learning_rate": 2.9556065439656724e-06,
      "loss": 0.7559396,
      "num_input_tokens_seen": 64563135,
      "step": 3004,
      "time_per_iteration": 3.7215256690979004
    },
    {
      "auxiliary_loss_clip": 0.01110826,
      "auxiliary_loss_mlp": 0.01034675,
      "balance_loss_clip": 1.04129195,
      "balance_loss_mlp": 1.02657509,
      "epoch": 0.36132988637046837,
      "flos": 18113055482880.0,
      "grad_norm": 1.7127789026907507,
      "language_loss": 0.81939715,
      "learning_rate": 2.9549221707194952e-06,
      "loss": 0.84085214,
      "num_input_tokens_seen": 64581985,
      "step": 3005,
      "time_per_iteration": 2.858633279800415
    },
    {
      "auxiliary_loss_clip": 0.01182771,
      "auxiliary_loss_mlp": 0.01031475,
      "balance_loss_clip": 1.0549686,
      "balance_loss_mlp": 1.02244496,
      "epoch": 0.3614501292611074,
      "flos": 27813101333760.0,
      "grad_norm": 1.9966521993581747,
      "language_loss": 0.72795415,
      "learning_rate": 2.954237652612384e-06,
      "loss": 0.75009656,
      "num_input_tokens_seen": 64601035,
      "step": 3006,
      "time_per_iteration": 2.714794874191284
    },
    {
      "auxiliary_loss_clip": 0.01163728,
      "auxiliary_loss_mlp": 0.01029571,
      "balance_loss_clip": 1.05238163,
      "balance_loss_mlp": 1.02184319,
      "epoch": 0.36157037215174653,
      "flos": 22634926732800.0,
      "grad_norm": 1.9477136769913024,
      "language_loss": 0.846645,
      "learning_rate": 2.9535529897481796e-06,
      "loss": 0.86857802,
      "num_input_tokens_seen": 64618580,
      "step": 3007,
      "time_per_iteration": 2.754840612411499
    },
    {
      "auxiliary_loss_clip": 0.0119362,
      "auxiliary_loss_mlp": 0.01031055,
      "balance_loss_clip": 1.05366576,
      "balance_loss_mlp": 1.02211988,
      "epoch": 0.36169061504238564,
      "flos": 12600839376000.0,
      "grad_norm": 2.367705580839698,
      "language_loss": 0.77432978,
      "learning_rate": 2.9528681822307446e-06,
      "loss": 0.7965765,
      "num_input_tokens_seen": 64635430,
      "step": 3008,
      "time_per_iteration": 3.592416524887085
    },
    {
      "auxiliary_loss_clip": 0.01177008,
      "auxiliary_loss_mlp": 0.00763314,
      "balance_loss_clip": 1.05602264,
      "balance_loss_mlp": 1.00020218,
      "epoch": 0.3618108579330247,
      "flos": 26684644682880.0,
      "grad_norm": 2.1046109992905153,
      "language_loss": 0.8238008,
      "learning_rate": 2.952183230163964e-06,
      "loss": 0.84320402,
      "num_input_tokens_seen": 64655005,
      "step": 3009,
      "time_per_iteration": 3.5639495849609375
    },
    {
      "auxiliary_loss_clip": 0.01140361,
      "auxiliary_loss_mlp": 0.01028193,
      "balance_loss_clip": 1.04554725,
      "balance_loss_mlp": 1.01915669,
      "epoch": 0.3619311008236638,
      "flos": 22817029708800.0,
      "grad_norm": 1.936470142533772,
      "language_loss": 0.7368924,
      "learning_rate": 2.9514981336517448e-06,
      "loss": 0.75857794,
      "num_input_tokens_seen": 64674775,
      "step": 3010,
      "time_per_iteration": 2.799267530441284
    },
    {
      "auxiliary_loss_clip": 0.01180217,
      "auxiliary_loss_mlp": 0.01030578,
      "balance_loss_clip": 1.05647635,
      "balance_loss_mlp": 1.02193546,
      "epoch": 0.36205134371430286,
      "flos": 25919603884800.0,
      "grad_norm": 2.663304808391954,
      "language_loss": 0.81681979,
      "learning_rate": 2.950812892798015e-06,
      "loss": 0.83892775,
      "num_input_tokens_seen": 64695670,
      "step": 3011,
      "time_per_iteration": 2.695112705230713
    },
    {
      "auxiliary_loss_clip": 0.01126266,
      "auxiliary_loss_mlp": 0.00763669,
      "balance_loss_clip": 1.04763138,
      "balance_loss_mlp": 1.00022089,
      "epoch": 0.362171586604942,
      "flos": 26139592730880.0,
      "grad_norm": 3.75012947896066,
      "language_loss": 0.87312937,
      "learning_rate": 2.9501275077067256e-06,
      "loss": 0.89202881,
      "num_input_tokens_seen": 64716290,
      "step": 3012,
      "time_per_iteration": 2.8496928215026855
    },
    {
      "auxiliary_loss_clip": 0.01102656,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 1.04200137,
      "balance_loss_mlp": 1.01886547,
      "epoch": 0.3622918294955811,
      "flos": 28074208273920.0,
      "grad_norm": 1.7594165928459538,
      "language_loss": 0.88533318,
      "learning_rate": 2.949441978481848e-06,
      "loss": 0.9066292,
      "num_input_tokens_seen": 64737190,
      "step": 3013,
      "time_per_iteration": 2.8808069229125977
    },
    {
      "auxiliary_loss_clip": 0.01154577,
      "auxiliary_loss_mlp": 0.01032856,
      "balance_loss_clip": 1.04930639,
      "balance_loss_mlp": 1.02359962,
      "epoch": 0.36241207238622014,
      "flos": 19828005402240.0,
      "grad_norm": 2.0412071320149003,
      "language_loss": 0.80003285,
      "learning_rate": 2.9487563052273778e-06,
      "loss": 0.82190716,
      "num_input_tokens_seen": 64753950,
      "step": 3014,
      "time_per_iteration": 2.8149890899658203
    },
    {
      "auxiliary_loss_clip": 0.01178817,
      "auxiliary_loss_mlp": 0.01025504,
      "balance_loss_clip": 1.05724621,
      "balance_loss_mlp": 1.01701069,
      "epoch": 0.36253231527685925,
      "flos": 21397158017280.0,
      "grad_norm": 1.6861201308886742,
      "language_loss": 0.85621542,
      "learning_rate": 2.94807048804733e-06,
      "loss": 0.87825859,
      "num_input_tokens_seen": 64773570,
      "step": 3015,
      "time_per_iteration": 2.7524633407592773
    },
    {
      "auxiliary_loss_clip": 0.01155257,
      "auxiliary_loss_mlp": 0.01034526,
      "balance_loss_clip": 1.04898763,
      "balance_loss_mlp": 1.02510881,
      "epoch": 0.36265255816749836,
      "flos": 18362885552640.0,
      "grad_norm": 2.429001613076155,
      "language_loss": 0.90285921,
      "learning_rate": 2.9473845270457434e-06,
      "loss": 0.92475712,
      "num_input_tokens_seen": 64790385,
      "step": 3016,
      "time_per_iteration": 2.914992332458496
    },
    {
      "auxiliary_loss_clip": 0.01158287,
      "auxiliary_loss_mlp": 0.01030987,
      "balance_loss_clip": 1.05055952,
      "balance_loss_mlp": 1.02260637,
      "epoch": 0.3627728010581374,
      "flos": 18660046769280.0,
      "grad_norm": 3.041654709976678,
      "language_loss": 0.70047027,
      "learning_rate": 2.946698422326677e-06,
      "loss": 0.722363,
      "num_input_tokens_seen": 64807845,
      "step": 3017,
      "time_per_iteration": 3.052154064178467
    },
    {
      "auxiliary_loss_clip": 0.01133808,
      "auxiliary_loss_mlp": 0.01029214,
      "balance_loss_clip": 1.046731,
      "balance_loss_mlp": 1.01995778,
      "epoch": 0.36289304394877653,
      "flos": 27524272072320.0,
      "grad_norm": 2.0677791717600322,
      "language_loss": 0.80265737,
      "learning_rate": 2.946012173994213e-06,
      "loss": 0.82428765,
      "num_input_tokens_seen": 64827630,
      "step": 3018,
      "time_per_iteration": 2.96711802482605
    },
    {
      "auxiliary_loss_clip": 0.01175718,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.0564785,
      "balance_loss_mlp": 1.01832271,
      "epoch": 0.36301328683941564,
      "flos": 34533244932480.0,
      "grad_norm": 2.032708331925737,
      "language_loss": 0.67871171,
      "learning_rate": 2.945325782152454e-06,
      "loss": 0.70073795,
      "num_input_tokens_seen": 64850665,
      "step": 3019,
      "time_per_iteration": 2.9132590293884277
    },
    {
      "auxiliary_loss_clip": 0.01167853,
      "auxiliary_loss_mlp": 0.01025417,
      "balance_loss_clip": 1.05207038,
      "balance_loss_mlp": 1.01748395,
      "epoch": 0.3631335297300547,
      "flos": 19025976574080.0,
      "grad_norm": 2.4441300716797074,
      "language_loss": 0.79039603,
      "learning_rate": 2.9446392469055257e-06,
      "loss": 0.8123287,
      "num_input_tokens_seen": 64868700,
      "step": 3020,
      "time_per_iteration": 2.835613965988159
    },
    {
      "auxiliary_loss_clip": 0.01151761,
      "auxiliary_loss_mlp": 0.01028764,
      "balance_loss_clip": 1.05658543,
      "balance_loss_mlp": 1.02031815,
      "epoch": 0.3632537726206938,
      "flos": 19536769929600.0,
      "grad_norm": 1.803843155378383,
      "language_loss": 0.79762447,
      "learning_rate": 2.9439525683575745e-06,
      "loss": 0.81942976,
      "num_input_tokens_seen": 64887620,
      "step": 3021,
      "time_per_iteration": 2.8791427612304688
    },
    {
      "auxiliary_loss_clip": 0.01199858,
      "auxiliary_loss_mlp": 0.01025202,
      "balance_loss_clip": 1.0591253,
      "balance_loss_mlp": 1.01642799,
      "epoch": 0.3633740155113329,
      "flos": 21068611292160.0,
      "grad_norm": 2.6326434021022433,
      "language_loss": 0.7487964,
      "learning_rate": 2.9432657466127694e-06,
      "loss": 0.771047,
      "num_input_tokens_seen": 64907190,
      "step": 3022,
      "time_per_iteration": 2.721956491470337
    },
    {
      "auxiliary_loss_clip": 0.01140573,
      "auxiliary_loss_mlp": 0.01031037,
      "balance_loss_clip": 1.05553961,
      "balance_loss_mlp": 1.02302873,
      "epoch": 0.36349425840197197,
      "flos": 20298722158080.0,
      "grad_norm": 1.9308089420836378,
      "language_loss": 0.76693016,
      "learning_rate": 2.9425787817753007e-06,
      "loss": 0.78864622,
      "num_input_tokens_seen": 64925850,
      "step": 3023,
      "time_per_iteration": 2.8514485359191895
    },
    {
      "auxiliary_loss_clip": 0.01152094,
      "auxiliary_loss_mlp": 0.01033744,
      "balance_loss_clip": 1.05125368,
      "balance_loss_mlp": 1.02499402,
      "epoch": 0.3636145012926111,
      "flos": 29716762331520.0,
      "grad_norm": 1.4737271804959662,
      "language_loss": 0.71712446,
      "learning_rate": 2.94189167394938e-06,
      "loss": 0.7389828,
      "num_input_tokens_seen": 64948285,
      "step": 3024,
      "time_per_iteration": 2.9803271293640137
    },
    {
      "auxiliary_loss_clip": 0.01196271,
      "auxiliary_loss_mlp": 0.01026125,
      "balance_loss_clip": 1.05892563,
      "balance_loss_mlp": 1.01726151,
      "epoch": 0.3637347441832502,
      "flos": 21431847576960.0,
      "grad_norm": 1.9553254160462754,
      "language_loss": 0.81424248,
      "learning_rate": 2.941204423239241e-06,
      "loss": 0.83646637,
      "num_input_tokens_seen": 64967160,
      "step": 3025,
      "time_per_iteration": 2.76127552986145
    },
    {
      "auxiliary_loss_clip": 0.01178395,
      "auxiliary_loss_mlp": 0.01029962,
      "balance_loss_clip": 1.05551052,
      "balance_loss_mlp": 1.02109265,
      "epoch": 0.36385498707388925,
      "flos": 29533941083520.0,
      "grad_norm": 2.2313277704435457,
      "language_loss": 0.76082414,
      "learning_rate": 2.9405170297491395e-06,
      "loss": 0.78290772,
      "num_input_tokens_seen": 64987155,
      "step": 3026,
      "time_per_iteration": 3.9228198528289795
    },
    {
      "auxiliary_loss_clip": 0.0111501,
      "auxiliary_loss_mlp": 0.00764052,
      "balance_loss_clip": 1.05293345,
      "balance_loss_mlp": 1.00023627,
      "epoch": 0.36397522996452836,
      "flos": 22236569925120.0,
      "grad_norm": 2.4687679240571896,
      "language_loss": 0.80481339,
      "learning_rate": 2.939829493583353e-06,
      "loss": 0.82360399,
      "num_input_tokens_seen": 65003800,
      "step": 3027,
      "time_per_iteration": 3.0461294651031494
    },
    {
      "auxiliary_loss_clip": 0.01143561,
      "auxiliary_loss_mlp": 0.0102487,
      "balance_loss_clip": 1.04680467,
      "balance_loss_mlp": 1.01678181,
      "epoch": 0.3640954728551674,
      "flos": 21506505995520.0,
      "grad_norm": 2.452790317211603,
      "language_loss": 0.83476084,
      "learning_rate": 2.939141814846179e-06,
      "loss": 0.85644519,
      "num_input_tokens_seen": 65021215,
      "step": 3028,
      "time_per_iteration": 2.7621371746063232
    },
    {
      "auxiliary_loss_clip": 0.01166684,
      "auxiliary_loss_mlp": 0.01029214,
      "balance_loss_clip": 1.05299115,
      "balance_loss_mlp": 1.02060127,
      "epoch": 0.3642157157458065,
      "flos": 17712867081600.0,
      "grad_norm": 1.5733046126238512,
      "language_loss": 0.82668245,
      "learning_rate": 2.938453993641938e-06,
      "loss": 0.84864146,
      "num_input_tokens_seen": 65039590,
      "step": 3029,
      "time_per_iteration": 2.7919559478759766
    },
    {
      "auxiliary_loss_clip": 0.01164358,
      "auxiliary_loss_mlp": 0.01030736,
      "balance_loss_clip": 1.05610812,
      "balance_loss_mlp": 1.02174211,
      "epoch": 0.36433595863644563,
      "flos": 17639537466240.0,
      "grad_norm": 2.1227228105526064,
      "language_loss": 0.70844746,
      "learning_rate": 2.937766030074973e-06,
      "loss": 0.73039836,
      "num_input_tokens_seen": 65056845,
      "step": 3030,
      "time_per_iteration": 3.725620746612549
    },
    {
      "auxiliary_loss_clip": 0.01157961,
      "auxiliary_loss_mlp": 0.01030488,
      "balance_loss_clip": 1.05329752,
      "balance_loss_mlp": 1.02142215,
      "epoch": 0.3644562015270847,
      "flos": 26833279161600.0,
      "grad_norm": 2.0512648300478764,
      "language_loss": 0.82850456,
      "learning_rate": 2.937077924249646e-06,
      "loss": 0.85038906,
      "num_input_tokens_seen": 65079435,
      "step": 3031,
      "time_per_iteration": 2.8781497478485107
    },
    {
      "auxiliary_loss_clip": 0.01171258,
      "auxiliary_loss_mlp": 0.01032745,
      "balance_loss_clip": 1.05261672,
      "balance_loss_mlp": 1.0237329,
      "epoch": 0.3645764444177238,
      "flos": 14282715847680.0,
      "grad_norm": 2.192152377107307,
      "language_loss": 0.76043856,
      "learning_rate": 2.9363896762703443e-06,
      "loss": 0.78247857,
      "num_input_tokens_seen": 65096500,
      "step": 3032,
      "time_per_iteration": 2.866676092147827
    },
    {
      "auxiliary_loss_clip": 0.01195249,
      "auxiliary_loss_mlp": 0.0102649,
      "balance_loss_clip": 1.05673051,
      "balance_loss_mlp": 1.01774597,
      "epoch": 0.3646966873083629,
      "flos": 20667489137280.0,
      "grad_norm": 1.6930484979076361,
      "language_loss": 0.84800828,
      "learning_rate": 2.9357012862414725e-06,
      "loss": 0.87022567,
      "num_input_tokens_seen": 65115860,
      "step": 3033,
      "time_per_iteration": 2.7183830738067627
    },
    {
      "auxiliary_loss_clip": 0.01182945,
      "auxiliary_loss_mlp": 0.01031621,
      "balance_loss_clip": 1.05687213,
      "balance_loss_mlp": 1.02325857,
      "epoch": 0.36481693019900197,
      "flos": 27782613665280.0,
      "grad_norm": 2.058529981792853,
      "language_loss": 0.72044933,
      "learning_rate": 2.9350127542674593e-06,
      "loss": 0.74259496,
      "num_input_tokens_seen": 65138070,
      "step": 3034,
      "time_per_iteration": 4.58564019203186
    },
    {
      "auxiliary_loss_clip": 0.01172563,
      "auxiliary_loss_mlp": 0.01029838,
      "balance_loss_clip": 1.05740809,
      "balance_loss_mlp": 1.02100468,
      "epoch": 0.3649371730896411,
      "flos": 19712588025600.0,
      "grad_norm": 2.5815760642663537,
      "language_loss": 0.76589555,
      "learning_rate": 2.934324080452755e-06,
      "loss": 0.78791958,
      "num_input_tokens_seen": 65155860,
      "step": 3035,
      "time_per_iteration": 2.9124064445495605
    },
    {
      "auxiliary_loss_clip": 0.01141011,
      "auxiliary_loss_mlp": 0.00765008,
      "balance_loss_clip": 1.04920566,
      "balance_loss_mlp": 1.00017214,
      "epoch": 0.3650574159802802,
      "flos": 24750496016640.0,
      "grad_norm": 2.1381602425065545,
      "language_loss": 0.78404009,
      "learning_rate": 2.9336352649018307e-06,
      "loss": 0.80310035,
      "num_input_tokens_seen": 65175930,
      "step": 3036,
      "time_per_iteration": 2.8277554512023926
    },
    {
      "auxiliary_loss_clip": 0.01165841,
      "auxiliary_loss_mlp": 0.01030323,
      "balance_loss_clip": 1.05298531,
      "balance_loss_mlp": 1.02163255,
      "epoch": 0.36517765887091924,
      "flos": 32853487363200.0,
      "grad_norm": 1.7769811768693267,
      "language_loss": 0.70255965,
      "learning_rate": 2.9329463077191783e-06,
      "loss": 0.72452128,
      "num_input_tokens_seen": 65199305,
      "step": 3037,
      "time_per_iteration": 2.9407994747161865
    },
    {
      "auxiliary_loss_clip": 0.01134185,
      "auxiliary_loss_mlp": 0.01028907,
      "balance_loss_clip": 1.04877472,
      "balance_loss_mlp": 1.02030611,
      "epoch": 0.36529790176155835,
      "flos": 20120318282880.0,
      "grad_norm": 2.211960444867724,
      "language_loss": 0.64097971,
      "learning_rate": 2.9322572090093135e-06,
      "loss": 0.66261065,
      "num_input_tokens_seen": 65218010,
      "step": 3038,
      "time_per_iteration": 2.8294153213500977
    },
    {
      "auxiliary_loss_clip": 0.01133923,
      "auxiliary_loss_mlp": 0.01030244,
      "balance_loss_clip": 1.04891205,
      "balance_loss_mlp": 1.02175033,
      "epoch": 0.36541814465219746,
      "flos": 17639573379840.0,
      "grad_norm": 3.647554331010622,
      "language_loss": 0.76461101,
      "learning_rate": 2.9315679688767713e-06,
      "loss": 0.78625268,
      "num_input_tokens_seen": 65236020,
      "step": 3039,
      "time_per_iteration": 2.8355793952941895
    },
    {
      "auxiliary_loss_clip": 0.01161784,
      "auxiliary_loss_mlp": 0.01027185,
      "balance_loss_clip": 1.05158126,
      "balance_loss_mlp": 1.01924527,
      "epoch": 0.3655383875428365,
      "flos": 22674356887680.0,
      "grad_norm": 1.543477004064135,
      "language_loss": 0.66670322,
      "learning_rate": 2.9308785874261085e-06,
      "loss": 0.68859291,
      "num_input_tokens_seen": 65256210,
      "step": 3040,
      "time_per_iteration": 2.8440096378326416
    },
    {
      "auxiliary_loss_clip": 0.0119812,
      "auxiliary_loss_mlp": 0.01028893,
      "balance_loss_clip": 1.05913401,
      "balance_loss_mlp": 1.02025044,
      "epoch": 0.36565863043347563,
      "flos": 21981173247360.0,
      "grad_norm": 1.6425336510538318,
      "language_loss": 0.8169533,
      "learning_rate": 2.9301890647619045e-06,
      "loss": 0.8392235,
      "num_input_tokens_seen": 65275505,
      "step": 3041,
      "time_per_iteration": 2.7902135848999023
    },
    {
      "auxiliary_loss_clip": 0.01172245,
      "auxiliary_loss_mlp": 0.01027829,
      "balance_loss_clip": 1.05529749,
      "balance_loss_mlp": 1.01913881,
      "epoch": 0.36577887332411474,
      "flos": 24827632473600.0,
      "grad_norm": 2.1218694753116787,
      "language_loss": 0.80589843,
      "learning_rate": 2.929499400988759e-06,
      "loss": 0.82789922,
      "num_input_tokens_seen": 65296665,
      "step": 3042,
      "time_per_iteration": 2.818767786026001
    },
    {
      "auxiliary_loss_clip": 0.01179454,
      "auxiliary_loss_mlp": 0.0103167,
      "balance_loss_clip": 1.05665982,
      "balance_loss_mlp": 1.02252698,
      "epoch": 0.3658991162147538,
      "flos": 28293191539200.0,
      "grad_norm": 1.6459572618933778,
      "language_loss": 0.65432727,
      "learning_rate": 2.9288095962112927e-06,
      "loss": 0.67643851,
      "num_input_tokens_seen": 65317370,
      "step": 3043,
      "time_per_iteration": 2.8477816581726074
    },
    {
      "auxiliary_loss_clip": 0.01197074,
      "auxiliary_loss_mlp": 0.01029479,
      "balance_loss_clip": 1.05853009,
      "balance_loss_mlp": 1.02055573,
      "epoch": 0.3660193591053929,
      "flos": 17785550252160.0,
      "grad_norm": 1.94478186464327,
      "language_loss": 0.85252208,
      "learning_rate": 2.9281196505341503e-06,
      "loss": 0.87478757,
      "num_input_tokens_seen": 65334540,
      "step": 3044,
      "time_per_iteration": 2.767505407333374
    },
    {
      "auxiliary_loss_clip": 0.01136151,
      "auxiliary_loss_mlp": 0.00763724,
      "balance_loss_clip": 1.05300426,
      "balance_loss_mlp": 1.00021577,
      "epoch": 0.36613960199603196,
      "flos": 10342776839040.0,
      "grad_norm": 2.041767060352574,
      "language_loss": 0.78719819,
      "learning_rate": 2.9274295640619946e-06,
      "loss": 0.80619699,
      "num_input_tokens_seen": 65351670,
      "step": 3045,
      "time_per_iteration": 2.8605852127075195
    },
    {
      "auxiliary_loss_clip": 0.01150582,
      "auxiliary_loss_mlp": 0.0102851,
      "balance_loss_clip": 1.04909348,
      "balance_loss_mlp": 1.02076149,
      "epoch": 0.36625984488667107,
      "flos": 19755609540480.0,
      "grad_norm": 1.752359137090996,
      "language_loss": 0.7872802,
      "learning_rate": 2.9267393368995103e-06,
      "loss": 0.80907112,
      "num_input_tokens_seen": 65370900,
      "step": 3046,
      "time_per_iteration": 2.8611161708831787
    },
    {
      "auxiliary_loss_clip": 0.01196942,
      "auxiliary_loss_mlp": 0.01033517,
      "balance_loss_clip": 1.05752325,
      "balance_loss_mlp": 1.0250051,
      "epoch": 0.3663800877773102,
      "flos": 17674262939520.0,
      "grad_norm": 2.5661232337030944,
      "language_loss": 0.7488867,
      "learning_rate": 2.926048969151407e-06,
      "loss": 0.7711913,
      "num_input_tokens_seen": 65388185,
      "step": 3047,
      "time_per_iteration": 2.678215742111206
    },
    {
      "auxiliary_loss_clip": 0.01133008,
      "auxiliary_loss_mlp": 0.01026255,
      "balance_loss_clip": 1.0536648,
      "balance_loss_mlp": 1.0174334,
      "epoch": 0.36650033066794924,
      "flos": 20303606407680.0,
      "grad_norm": 1.7864753397349535,
      "language_loss": 0.68705332,
      "learning_rate": 2.92535846092241e-06,
      "loss": 0.70864594,
      "num_input_tokens_seen": 65407200,
      "step": 3048,
      "time_per_iteration": 2.851482391357422
    },
    {
      "auxiliary_loss_clip": 0.01168551,
      "auxiliary_loss_mlp": 0.01029984,
      "balance_loss_clip": 1.05483961,
      "balance_loss_mlp": 1.02090645,
      "epoch": 0.36662057355858835,
      "flos": 24716237420160.0,
      "grad_norm": 1.6883106797194478,
      "language_loss": 0.82673126,
      "learning_rate": 2.9246678123172704e-06,
      "loss": 0.84871656,
      "num_input_tokens_seen": 65427290,
      "step": 3049,
      "time_per_iteration": 2.8687281608581543
    },
    {
      "auxiliary_loss_clip": 0.01199931,
      "auxiliary_loss_mlp": 0.01029932,
      "balance_loss_clip": 1.05981731,
      "balance_loss_mlp": 1.02073526,
      "epoch": 0.36674081644922746,
      "flos": 12385267902720.0,
      "grad_norm": 2.203208828200931,
      "language_loss": 0.74581409,
      "learning_rate": 2.9239770234407596e-06,
      "loss": 0.76811272,
      "num_input_tokens_seen": 65445595,
      "step": 3050,
      "time_per_iteration": 2.662234306335449
    },
    {
      "auxiliary_loss_clip": 0.01182408,
      "auxiliary_loss_mlp": 0.01028291,
      "balance_loss_clip": 1.05566895,
      "balance_loss_mlp": 1.01949906,
      "epoch": 0.3668610593398665,
      "flos": 21105922544640.0,
      "grad_norm": 1.806790746651207,
      "language_loss": 0.68115407,
      "learning_rate": 2.9232860943976686e-06,
      "loss": 0.70326108,
      "num_input_tokens_seen": 65466330,
      "step": 3051,
      "time_per_iteration": 2.759047269821167
    },
    {
      "auxiliary_loss_clip": 0.0116463,
      "auxiliary_loss_mlp": 0.01028207,
      "balance_loss_clip": 1.05499923,
      "balance_loss_mlp": 1.01990354,
      "epoch": 0.3669813022305056,
      "flos": 26758082039040.0,
      "grad_norm": 1.6857267510734768,
      "language_loss": 0.84490538,
      "learning_rate": 2.9225950252928115e-06,
      "loss": 0.86683381,
      "num_input_tokens_seen": 65487180,
      "step": 3052,
      "time_per_iteration": 3.854109287261963
    },
    {
      "auxiliary_loss_clip": 0.01180262,
      "auxiliary_loss_mlp": 0.01036073,
      "balance_loss_clip": 1.05632687,
      "balance_loss_mlp": 1.02737653,
      "epoch": 0.36710154512114473,
      "flos": 19099521671040.0,
      "grad_norm": 2.9130872327647954,
      "language_loss": 0.81811649,
      "learning_rate": 2.9219038162310217e-06,
      "loss": 0.84027982,
      "num_input_tokens_seen": 65505380,
      "step": 3053,
      "time_per_iteration": 2.680817127227783
    },
    {
      "auxiliary_loss_clip": 0.01111145,
      "auxiliary_loss_mlp": 0.00764017,
      "balance_loss_clip": 1.04791605,
      "balance_loss_mlp": 1.00021982,
      "epoch": 0.3672217880117838,
      "flos": 20812029465600.0,
      "grad_norm": 1.9475270594694967,
      "language_loss": 0.82426548,
      "learning_rate": 2.921212467317157e-06,
      "loss": 0.8430171,
      "num_input_tokens_seen": 65524825,
      "step": 3054,
      "time_per_iteration": 2.9275104999542236
    },
    {
      "auxiliary_loss_clip": 0.01150354,
      "auxiliary_loss_mlp": 0.01036795,
      "balance_loss_clip": 1.04862833,
      "balance_loss_mlp": 1.02756238,
      "epoch": 0.3673420309024229,
      "flos": 13590394133760.0,
      "grad_norm": 1.8045060207874672,
      "language_loss": 0.8026917,
      "learning_rate": 2.920520978656093e-06,
      "loss": 0.82456326,
      "num_input_tokens_seen": 65541790,
      "step": 3055,
      "time_per_iteration": 3.610708236694336
    },
    {
      "auxiliary_loss_clip": 0.01193588,
      "auxiliary_loss_mlp": 0.00763872,
      "balance_loss_clip": 1.0563587,
      "balance_loss_mlp": 1.00020671,
      "epoch": 0.367462273793062,
      "flos": 28986877969920.0,
      "grad_norm": 1.9426802500743918,
      "language_loss": 0.76685506,
      "learning_rate": 2.919829350352729e-06,
      "loss": 0.78642964,
      "num_input_tokens_seen": 65563395,
      "step": 3056,
      "time_per_iteration": 2.7628262042999268
    },
    {
      "auxiliary_loss_clip": 0.01100171,
      "auxiliary_loss_mlp": 0.01003911,
      "balance_loss_clip": 1.02711916,
      "balance_loss_mlp": 1.00236118,
      "epoch": 0.36758251668370107,
      "flos": 62643148346880.0,
      "grad_norm": 0.7555108572020115,
      "language_loss": 0.5992595,
      "learning_rate": 2.919137582511983e-06,
      "loss": 0.62030029,
      "num_input_tokens_seen": 65619835,
      "step": 3057,
      "time_per_iteration": 3.1424074172973633
    },
    {
      "auxiliary_loss_clip": 0.01161529,
      "auxiliary_loss_mlp": 0.01028168,
      "balance_loss_clip": 1.05715001,
      "balance_loss_mlp": 1.01949573,
      "epoch": 0.3677027595743402,
      "flos": 12713886455040.0,
      "grad_norm": 1.7508229426224924,
      "language_loss": 0.64130658,
      "learning_rate": 2.918445675238797e-06,
      "loss": 0.66320354,
      "num_input_tokens_seen": 65636760,
      "step": 3058,
      "time_per_iteration": 2.6767497062683105
    },
    {
      "auxiliary_loss_clip": 0.0119555,
      "auxiliary_loss_mlp": 0.01031318,
      "balance_loss_clip": 1.05581057,
      "balance_loss_mlp": 1.02350426,
      "epoch": 0.36782300246497923,
      "flos": 25046579825280.0,
      "grad_norm": 2.424419482803805,
      "language_loss": 0.70084631,
      "learning_rate": 2.917753628638132e-06,
      "loss": 0.72311497,
      "num_input_tokens_seen": 65657065,
      "step": 3059,
      "time_per_iteration": 2.6764109134674072
    },
    {
      "auxiliary_loss_clip": 0.01168892,
      "auxiliary_loss_mlp": 0.01035045,
      "balance_loss_clip": 1.05538952,
      "balance_loss_mlp": 1.02663481,
      "epoch": 0.36794324535561834,
      "flos": 17419512706560.0,
      "grad_norm": 2.18570161282597,
      "language_loss": 0.70237249,
      "learning_rate": 2.9170614428149716e-06,
      "loss": 0.72441185,
      "num_input_tokens_seen": 65675400,
      "step": 3060,
      "time_per_iteration": 4.51049017906189
    },
    {
      "auxiliary_loss_clip": 0.01152551,
      "auxiliary_loss_mlp": 0.01030423,
      "balance_loss_clip": 1.05562949,
      "balance_loss_mlp": 1.02139866,
      "epoch": 0.36806348824625745,
      "flos": 24089128848000.0,
      "grad_norm": 2.192932537871339,
      "language_loss": 0.8638804,
      "learning_rate": 2.9163691178743195e-06,
      "loss": 0.88571012,
      "num_input_tokens_seen": 65694050,
      "step": 3061,
      "time_per_iteration": 2.750082015991211
    },
    {
      "auxiliary_loss_clip": 0.01179095,
      "auxiliary_loss_mlp": 0.01027644,
      "balance_loss_clip": 1.05640459,
      "balance_loss_mlp": 1.01963329,
      "epoch": 0.3681837311368965,
      "flos": 20521871400960.0,
      "grad_norm": 2.0633169664747215,
      "language_loss": 0.77733457,
      "learning_rate": 2.9156766539212006e-06,
      "loss": 0.799402,
      "num_input_tokens_seen": 65711695,
      "step": 3062,
      "time_per_iteration": 2.6384732723236084
    },
    {
      "auxiliary_loss_clip": 0.01182585,
      "auxiliary_loss_mlp": 0.01033204,
      "balance_loss_clip": 1.05304742,
      "balance_loss_mlp": 1.02489471,
      "epoch": 0.3683039740275356,
      "flos": 21466644877440.0,
      "grad_norm": 2.172812970818998,
      "language_loss": 0.72023094,
      "learning_rate": 2.9149840510606614e-06,
      "loss": 0.74238884,
      "num_input_tokens_seen": 65730350,
      "step": 3063,
      "time_per_iteration": 2.707871675491333
    },
    {
      "auxiliary_loss_clip": 0.01082311,
      "auxiliary_loss_mlp": 0.00755715,
      "balance_loss_clip": 1.02279305,
      "balance_loss_mlp": 1.00041103,
      "epoch": 0.36842421691817473,
      "flos": 70380999987840.0,
      "grad_norm": 2.2307007250479054,
      "language_loss": 0.64146864,
      "learning_rate": 2.914291309397769e-06,
      "loss": 0.65984887,
      "num_input_tokens_seen": 65787820,
      "step": 3064,
      "time_per_iteration": 3.4315669536590576
    },
    {
      "auxiliary_loss_clip": 0.01109501,
      "auxiliary_loss_mlp": 0.01028949,
      "balance_loss_clip": 1.04499769,
      "balance_loss_mlp": 1.02057493,
      "epoch": 0.3685444598088138,
      "flos": 23331378510720.0,
      "grad_norm": 2.148740462374405,
      "language_loss": 0.78882915,
      "learning_rate": 2.9135984290376117e-06,
      "loss": 0.81021363,
      "num_input_tokens_seen": 65806685,
      "step": 3065,
      "time_per_iteration": 2.779705762863159
    },
    {
      "auxiliary_loss_clip": 0.0111795,
      "auxiliary_loss_mlp": 0.01030711,
      "balance_loss_clip": 1.04649103,
      "balance_loss_mlp": 1.02173996,
      "epoch": 0.3686647026994529,
      "flos": 23070271570560.0,
      "grad_norm": 1.692092650318129,
      "language_loss": 0.82916558,
      "learning_rate": 2.9129054100853e-06,
      "loss": 0.85065216,
      "num_input_tokens_seen": 65825525,
      "step": 3066,
      "time_per_iteration": 2.8168540000915527
    },
    {
      "auxiliary_loss_clip": 0.01167685,
      "auxiliary_loss_mlp": 0.01032408,
      "balance_loss_clip": 1.05275428,
      "balance_loss_mlp": 1.02381933,
      "epoch": 0.368784945590092,
      "flos": 25119909440640.0,
      "grad_norm": 1.640213611373079,
      "language_loss": 0.76296139,
      "learning_rate": 2.912212252645963e-06,
      "loss": 0.78496236,
      "num_input_tokens_seen": 65848110,
      "step": 3067,
      "time_per_iteration": 2.685175895690918
    },
    {
      "auxiliary_loss_clip": 0.01184961,
      "auxiliary_loss_mlp": 0.01030642,
      "balance_loss_clip": 1.05390596,
      "balance_loss_mlp": 1.02162969,
      "epoch": 0.36890518848073106,
      "flos": 18442284566400.0,
      "grad_norm": 1.994615964227408,
      "language_loss": 0.76792657,
      "learning_rate": 2.9115189568247523e-06,
      "loss": 0.79008257,
      "num_input_tokens_seen": 65865670,
      "step": 3068,
      "time_per_iteration": 2.6370818614959717
    },
    {
      "auxiliary_loss_clip": 0.01133791,
      "auxiliary_loss_mlp": 0.01033506,
      "balance_loss_clip": 1.05685234,
      "balance_loss_mlp": 1.02522671,
      "epoch": 0.36902543137137017,
      "flos": 16362446336640.0,
      "grad_norm": 1.8809974313331506,
      "language_loss": 0.92508709,
      "learning_rate": 2.910825522726841e-06,
      "loss": 0.94676006,
      "num_input_tokens_seen": 65883195,
      "step": 3069,
      "time_per_iteration": 2.793888807296753
    },
    {
      "auxiliary_loss_clip": 0.0112873,
      "auxiliary_loss_mlp": 0.0102808,
      "balance_loss_clip": 1.04677069,
      "balance_loss_mlp": 1.01959836,
      "epoch": 0.3691456742620093,
      "flos": 12275596702080.0,
      "grad_norm": 2.195110598555077,
      "language_loss": 0.77404684,
      "learning_rate": 2.9101319504574215e-06,
      "loss": 0.79561496,
      "num_input_tokens_seen": 65899635,
      "step": 3070,
      "time_per_iteration": 2.658608913421631
    },
    {
      "auxiliary_loss_clip": 0.01169611,
      "auxiliary_loss_mlp": 0.01035667,
      "balance_loss_clip": 1.04989696,
      "balance_loss_mlp": 1.02670884,
      "epoch": 0.36926591715264834,
      "flos": 17786412178560.0,
      "grad_norm": 1.6130309965195073,
      "language_loss": 0.76268196,
      "learning_rate": 2.909438240121709e-06,
      "loss": 0.78473473,
      "num_input_tokens_seen": 65919910,
      "step": 3071,
      "time_per_iteration": 2.746612787246704
    },
    {
      "auxiliary_loss_clip": 0.01161052,
      "auxiliary_loss_mlp": 0.01028342,
      "balance_loss_clip": 1.05272245,
      "balance_loss_mlp": 1.01959229,
      "epoch": 0.36938616004328745,
      "flos": 28948309741440.0,
      "grad_norm": 1.7529545056745583,
      "language_loss": 0.70313042,
      "learning_rate": 2.908744391824939e-06,
      "loss": 0.72502434,
      "num_input_tokens_seen": 65940930,
      "step": 3072,
      "time_per_iteration": 2.713883638381958
    },
    {
      "auxiliary_loss_clip": 0.01123527,
      "auxiliary_loss_mlp": 0.01029216,
      "balance_loss_clip": 1.04567122,
      "balance_loss_mlp": 1.02093744,
      "epoch": 0.36950640293392656,
      "flos": 29205394358400.0,
      "grad_norm": 1.7241076735374798,
      "language_loss": 0.7927959,
      "learning_rate": 2.908050405672367e-06,
      "loss": 0.81432331,
      "num_input_tokens_seen": 65960475,
      "step": 3073,
      "time_per_iteration": 2.910400867462158
    },
    {
      "auxiliary_loss_clip": 0.01171893,
      "auxiliary_loss_mlp": 0.01030202,
      "balance_loss_clip": 1.0501771,
      "balance_loss_mlp": 1.02226233,
      "epoch": 0.3696266458245656,
      "flos": 24827776128000.0,
      "grad_norm": 1.8169211174192257,
      "language_loss": 0.79017371,
      "learning_rate": 2.9073562817692703e-06,
      "loss": 0.81219465,
      "num_input_tokens_seen": 65979160,
      "step": 3074,
      "time_per_iteration": 2.7333781719207764
    },
    {
      "auxiliary_loss_clip": 0.01056736,
      "auxiliary_loss_mlp": 0.01009013,
      "balance_loss_clip": 1.02024841,
      "balance_loss_mlp": 1.00737977,
      "epoch": 0.3697468887152047,
      "flos": 59887257264000.0,
      "grad_norm": 0.7213550041027801,
      "language_loss": 0.56526554,
      "learning_rate": 2.9066620202209468e-06,
      "loss": 0.58592308,
      "num_input_tokens_seen": 66041650,
      "step": 3075,
      "time_per_iteration": 3.25860595703125
    },
    {
      "auxiliary_loss_clip": 0.01141447,
      "auxiliary_loss_mlp": 0.01028125,
      "balance_loss_clip": 1.04960465,
      "balance_loss_mlp": 1.01913035,
      "epoch": 0.3698671316058438,
      "flos": 26137581569280.0,
      "grad_norm": 2.1337151708874322,
      "language_loss": 0.77557176,
      "learning_rate": 2.905967621132716e-06,
      "loss": 0.7972675,
      "num_input_tokens_seen": 66059260,
      "step": 3076,
      "time_per_iteration": 2.6882569789886475
    },
    {
      "auxiliary_loss_clip": 0.0117004,
      "auxiliary_loss_mlp": 0.01031588,
      "balance_loss_clip": 1.05203331,
      "balance_loss_mlp": 1.02190828,
      "epoch": 0.3699873744964829,
      "flos": 24607464059520.0,
      "grad_norm": 2.0340322670214355,
      "language_loss": 0.75474054,
      "learning_rate": 2.9052730846099172e-06,
      "loss": 0.77675676,
      "num_input_tokens_seen": 66080605,
      "step": 3077,
      "time_per_iteration": 2.8291192054748535
    },
    {
      "auxiliary_loss_clip": 0.01074273,
      "auxiliary_loss_mlp": 0.01000686,
      "balance_loss_clip": 1.02217853,
      "balance_loss_mlp": 0.99902898,
      "epoch": 0.370107617387122,
      "flos": 64885340050560.0,
      "grad_norm": 0.8619617994667297,
      "language_loss": 0.60957265,
      "learning_rate": 2.9045784107579123e-06,
      "loss": 0.63032222,
      "num_input_tokens_seen": 66140710,
      "step": 3078,
      "time_per_iteration": 4.373821258544922
    },
    {
      "auxiliary_loss_clip": 0.01195788,
      "auxiliary_loss_mlp": 0.01025729,
      "balance_loss_clip": 1.05781329,
      "balance_loss_mlp": 1.01665759,
      "epoch": 0.37022786027776106,
      "flos": 15961683317760.0,
      "grad_norm": 1.994230855456401,
      "language_loss": 0.67068684,
      "learning_rate": 2.9038835996820807e-06,
      "loss": 0.69290197,
      "num_input_tokens_seen": 66158320,
      "step": 3079,
      "time_per_iteration": 2.7236931324005127
    },
    {
      "auxiliary_loss_clip": 0.01155942,
      "auxiliary_loss_mlp": 0.01039903,
      "balance_loss_clip": 1.04866683,
      "balance_loss_mlp": 1.03131974,
      "epoch": 0.37034810316840017,
      "flos": 18546927863040.0,
      "grad_norm": 1.9941432821420537,
      "language_loss": 0.79834127,
      "learning_rate": 2.903188651487826e-06,
      "loss": 0.82029974,
      "num_input_tokens_seen": 66176875,
      "step": 3080,
      "time_per_iteration": 2.680091142654419
    },
    {
      "auxiliary_loss_clip": 0.01184626,
      "auxiliary_loss_mlp": 0.01030138,
      "balance_loss_clip": 1.05577981,
      "balance_loss_mlp": 1.02216315,
      "epoch": 0.3704683460590393,
      "flos": 17821927751040.0,
      "grad_norm": 2.2960773612329253,
      "language_loss": 0.86952579,
      "learning_rate": 2.902493566280571e-06,
      "loss": 0.89167345,
      "num_input_tokens_seen": 66194980,
      "step": 3081,
      "time_per_iteration": 3.516897439956665
    },
    {
      "auxiliary_loss_clip": 0.01164699,
      "auxiliary_loss_mlp": 0.0104016,
      "balance_loss_clip": 1.05263412,
      "balance_loss_mlp": 1.03124285,
      "epoch": 0.37058858894967833,
      "flos": 14134081368960.0,
      "grad_norm": 2.0452844680893874,
      "language_loss": 0.81307817,
      "learning_rate": 2.9017983441657595e-06,
      "loss": 0.8351267,
      "num_input_tokens_seen": 66212310,
      "step": 3082,
      "time_per_iteration": 2.6495707035064697
    },
    {
      "auxiliary_loss_clip": 0.01138117,
      "auxiliary_loss_mlp": 0.01032239,
      "balance_loss_clip": 1.04749334,
      "balance_loss_mlp": 1.02363181,
      "epoch": 0.37070883184031744,
      "flos": 13954492344960.0,
      "grad_norm": 3.039449288428381,
      "language_loss": 0.74758214,
      "learning_rate": 2.9011029852488564e-06,
      "loss": 0.76928568,
      "num_input_tokens_seen": 66229545,
      "step": 3083,
      "time_per_iteration": 2.7994070053100586
    },
    {
      "auxiliary_loss_clip": 0.0109503,
      "auxiliary_loss_mlp": 0.01000863,
      "balance_loss_clip": 1.02243996,
      "balance_loss_mlp": 0.99924183,
      "epoch": 0.37082907473095655,
      "flos": 52315419306240.0,
      "grad_norm": 0.9919923221261641,
      "language_loss": 0.62466234,
      "learning_rate": 2.9004074896353465e-06,
      "loss": 0.64562124,
      "num_input_tokens_seen": 66283545,
      "step": 3084,
      "time_per_iteration": 3.100482225418091
    },
    {
      "auxiliary_loss_clip": 0.01196637,
      "auxiliary_loss_mlp": 0.0102684,
      "balance_loss_clip": 1.0603919,
      "balance_loss_mlp": 1.01791084,
      "epoch": 0.3709493176215956,
      "flos": 15998096730240.0,
      "grad_norm": 1.7790861009011831,
      "language_loss": 0.81869829,
      "learning_rate": 2.8997118574307362e-06,
      "loss": 0.84093308,
      "num_input_tokens_seen": 66300500,
      "step": 3085,
      "time_per_iteration": 3.525451898574829
    },
    {
      "auxiliary_loss_clip": 0.01159554,
      "auxiliary_loss_mlp": 0.01028474,
      "balance_loss_clip": 1.05384707,
      "balance_loss_mlp": 1.0198015,
      "epoch": 0.3710695605122347,
      "flos": 20959837931520.0,
      "grad_norm": 2.4595475372905358,
      "language_loss": 0.74485022,
      "learning_rate": 2.899016088740553e-06,
      "loss": 0.76673049,
      "num_input_tokens_seen": 66318610,
      "step": 3086,
      "time_per_iteration": 3.524543523788452
    },
    {
      "auxiliary_loss_clip": 0.01136116,
      "auxiliary_loss_mlp": 0.01032024,
      "balance_loss_clip": 1.05152333,
      "balance_loss_mlp": 1.02358413,
      "epoch": 0.37118980340287383,
      "flos": 14355578586240.0,
      "grad_norm": 2.1117755790104393,
      "language_loss": 0.79866171,
      "learning_rate": 2.898320183670344e-06,
      "loss": 0.82034314,
      "num_input_tokens_seen": 66336025,
      "step": 3087,
      "time_per_iteration": 2.74395489692688
    },
    {
      "auxiliary_loss_clip": 0.01136176,
      "auxiliary_loss_mlp": 0.01029502,
      "balance_loss_clip": 1.05065429,
      "balance_loss_mlp": 1.02097845,
      "epoch": 0.3713100462935129,
      "flos": 25885381201920.0,
      "grad_norm": 1.7130075465696646,
      "language_loss": 0.89045918,
      "learning_rate": 2.8976241423256767e-06,
      "loss": 0.91211605,
      "num_input_tokens_seen": 66356120,
      "step": 3088,
      "time_per_iteration": 2.8897385597229004
    },
    {
      "auxiliary_loss_clip": 0.01161198,
      "auxiliary_loss_mlp": 0.01031227,
      "balance_loss_clip": 1.052858,
      "balance_loss_mlp": 1.02335274,
      "epoch": 0.371430289184152,
      "flos": 30518934814080.0,
      "grad_norm": 2.8403154910916992,
      "language_loss": 0.68163151,
      "learning_rate": 2.896927964812142e-06,
      "loss": 0.7035557,
      "num_input_tokens_seen": 66376685,
      "step": 3089,
      "time_per_iteration": 2.729454278945923
    },
    {
      "auxiliary_loss_clip": 0.0116921,
      "auxiliary_loss_mlp": 0.0102852,
      "balance_loss_clip": 1.05689132,
      "balance_loss_mlp": 1.01962709,
      "epoch": 0.37155053207479105,
      "flos": 15742233175680.0,
      "grad_norm": 2.6993800336594327,
      "language_loss": 0.75011301,
      "learning_rate": 2.8962316512353465e-06,
      "loss": 0.77209032,
      "num_input_tokens_seen": 66394230,
      "step": 3090,
      "time_per_iteration": 2.6533782482147217
    },
    {
      "auxiliary_loss_clip": 0.01118095,
      "auxiliary_loss_mlp": 0.01027062,
      "balance_loss_clip": 1.04640317,
      "balance_loss_mlp": 1.01889658,
      "epoch": 0.37167077496543016,
      "flos": 23404061681280.0,
      "grad_norm": 3.432138814668347,
      "language_loss": 0.74989343,
      "learning_rate": 2.8955352017009233e-06,
      "loss": 0.77134502,
      "num_input_tokens_seen": 66413475,
      "step": 3091,
      "time_per_iteration": 2.836380958557129
    },
    {
      "auxiliary_loss_clip": 0.01167266,
      "auxiliary_loss_mlp": 0.01028031,
      "balance_loss_clip": 1.056898,
      "balance_loss_mlp": 1.01953745,
      "epoch": 0.3717910178560693,
      "flos": 22088653718400.0,
      "grad_norm": 2.4708831387069514,
      "language_loss": 0.77352983,
      "learning_rate": 2.8948386163145212e-06,
      "loss": 0.79548281,
      "num_input_tokens_seen": 66432685,
      "step": 3092,
      "time_per_iteration": 2.6632354259490967
    },
    {
      "auxiliary_loss_clip": 0.011867,
      "auxiliary_loss_mlp": 0.01031044,
      "balance_loss_clip": 1.05734193,
      "balance_loss_mlp": 1.02306867,
      "epoch": 0.3719112607467083,
      "flos": 26939969533440.0,
      "grad_norm": 1.8941405710429402,
      "language_loss": 0.79500437,
      "learning_rate": 2.8941418951818135e-06,
      "loss": 0.81718183,
      "num_input_tokens_seen": 66452245,
      "step": 3093,
      "time_per_iteration": 2.7025389671325684
    },
    {
      "auxiliary_loss_clip": 0.01153771,
      "auxiliary_loss_mlp": 0.01032169,
      "balance_loss_clip": 1.05309701,
      "balance_loss_mlp": 1.02386045,
      "epoch": 0.37203150363734744,
      "flos": 12166500119040.0,
      "grad_norm": 2.59169488633503,
      "language_loss": 0.71184647,
      "learning_rate": 2.8934450384084903e-06,
      "loss": 0.73370594,
      "num_input_tokens_seen": 66469760,
      "step": 3094,
      "time_per_iteration": 2.811070680618286
    },
    {
      "auxiliary_loss_clip": 0.01157357,
      "auxiliary_loss_mlp": 0.01034165,
      "balance_loss_clip": 1.04926109,
      "balance_loss_mlp": 1.02490807,
      "epoch": 0.37215174652798655,
      "flos": 23697595624320.0,
      "grad_norm": 2.0298270751156875,
      "language_loss": 0.69744933,
      "learning_rate": 2.8927480461002653e-06,
      "loss": 0.71936452,
      "num_input_tokens_seen": 66489730,
      "step": 3095,
      "time_per_iteration": 2.7436277866363525
    },
    {
      "auxiliary_loss_clip": 0.01163562,
      "auxiliary_loss_mlp": 0.01033734,
      "balance_loss_clip": 1.05045223,
      "balance_loss_mlp": 1.02437592,
      "epoch": 0.3722719894186256,
      "flos": 17887751424000.0,
      "grad_norm": 2.345533385277675,
      "language_loss": 0.8597436,
      "learning_rate": 2.892050918362872e-06,
      "loss": 0.88171661,
      "num_input_tokens_seen": 66504785,
      "step": 3096,
      "time_per_iteration": 2.630793809890747
    },
    {
      "auxiliary_loss_clip": 0.01021836,
      "auxiliary_loss_mlp": 0.01008044,
      "balance_loss_clip": 1.01770747,
      "balance_loss_mlp": 1.00642288,
      "epoch": 0.3723922323092647,
      "flos": 62419891363200.0,
      "grad_norm": 0.8455447584831847,
      "language_loss": 0.55807292,
      "learning_rate": 2.8913536553020626e-06,
      "loss": 0.57837176,
      "num_input_tokens_seen": 66558840,
      "step": 3097,
      "time_per_iteration": 3.3812026977539062
    },
    {
      "auxiliary_loss_clip": 0.01126562,
      "auxiliary_loss_mlp": 0.01029088,
      "balance_loss_clip": 1.04595876,
      "balance_loss_mlp": 1.02080858,
      "epoch": 0.3725124751999038,
      "flos": 23039747988480.0,
      "grad_norm": 1.9735876152899559,
      "language_loss": 0.85033166,
      "learning_rate": 2.8906562570236137e-06,
      "loss": 0.87188816,
      "num_input_tokens_seen": 66576750,
      "step": 3098,
      "time_per_iteration": 2.9260549545288086
    },
    {
      "auxiliary_loss_clip": 0.01121096,
      "auxiliary_loss_mlp": 0.01030152,
      "balance_loss_clip": 1.04550052,
      "balance_loss_mlp": 1.02180123,
      "epoch": 0.3726327180905429,
      "flos": 20920551431040.0,
      "grad_norm": 1.4857896792319343,
      "language_loss": 0.76524508,
      "learning_rate": 2.889958723633318e-06,
      "loss": 0.78675759,
      "num_input_tokens_seen": 66595690,
      "step": 3099,
      "time_per_iteration": 2.817406415939331
    },
    {
      "auxiliary_loss_clip": 0.01151893,
      "auxiliary_loss_mlp": 0.01035688,
      "balance_loss_clip": 1.05190349,
      "balance_loss_mlp": 1.02804637,
      "epoch": 0.372752960981182,
      "flos": 30592156688640.0,
      "grad_norm": 1.64983343824259,
      "language_loss": 0.73755586,
      "learning_rate": 2.889261055236992e-06,
      "loss": 0.75943166,
      "num_input_tokens_seen": 66617905,
      "step": 3100,
      "time_per_iteration": 2.8219411373138428
    },
    {
      "auxiliary_loss_clip": 0.01162799,
      "auxiliary_loss_mlp": 0.01028478,
      "balance_loss_clip": 1.05445969,
      "balance_loss_mlp": 1.0194602,
      "epoch": 0.3728732038718211,
      "flos": 25116749043840.0,
      "grad_norm": 1.7902986495082749,
      "language_loss": 0.82910311,
      "learning_rate": 2.8885632519404704e-06,
      "loss": 0.85101593,
      "num_input_tokens_seen": 66638175,
      "step": 3101,
      "time_per_iteration": 2.7250545024871826
    },
    {
      "auxiliary_loss_clip": 0.01162883,
      "auxiliary_loss_mlp": 0.01029365,
      "balance_loss_clip": 1.05293632,
      "balance_loss_mlp": 1.02100217,
      "epoch": 0.37299344676246016,
      "flos": 25302048330240.0,
      "grad_norm": 1.9237264746332927,
      "language_loss": 0.75658518,
      "learning_rate": 2.8878653138496107e-06,
      "loss": 0.77850771,
      "num_input_tokens_seen": 66658670,
      "step": 3102,
      "time_per_iteration": 2.7599291801452637
    },
    {
      "auxiliary_loss_clip": 0.01115292,
      "auxiliary_loss_mlp": 0.01023896,
      "balance_loss_clip": 1.04254127,
      "balance_loss_mlp": 1.01519382,
      "epoch": 0.37311368965309927,
      "flos": 23842531002240.0,
      "grad_norm": 2.6006699952504824,
      "language_loss": 0.76203072,
      "learning_rate": 2.8871672410702878e-06,
      "loss": 0.78342259,
      "num_input_tokens_seen": 66676030,
      "step": 3103,
      "time_per_iteration": 2.786940336227417
    },
    {
      "auxiliary_loss_clip": 0.01157996,
      "auxiliary_loss_mlp": 0.0103372,
      "balance_loss_clip": 1.05008709,
      "balance_loss_mlp": 1.02460623,
      "epoch": 0.3732339325437384,
      "flos": 25811943845760.0,
      "grad_norm": 1.918152970527542,
      "language_loss": 0.82559925,
      "learning_rate": 2.8864690337084008e-06,
      "loss": 0.84751642,
      "num_input_tokens_seen": 66695305,
      "step": 3104,
      "time_per_iteration": 3.782810926437378
    },
    {
      "auxiliary_loss_clip": 0.01173969,
      "auxiliary_loss_mlp": 0.01030762,
      "balance_loss_clip": 1.05192888,
      "balance_loss_mlp": 1.02281082,
      "epoch": 0.37335417543437743,
      "flos": 26208433146240.0,
      "grad_norm": 2.4380417375740806,
      "language_loss": 0.78052223,
      "learning_rate": 2.885770691869866e-06,
      "loss": 0.80256957,
      "num_input_tokens_seen": 66716185,
      "step": 3105,
      "time_per_iteration": 2.8215389251708984
    },
    {
      "auxiliary_loss_clip": 0.01177016,
      "auxiliary_loss_mlp": 0.0102864,
      "balance_loss_clip": 1.05439639,
      "balance_loss_mlp": 1.02098131,
      "epoch": 0.37347441832501654,
      "flos": 24023879792640.0,
      "grad_norm": 2.0087004852746544,
      "language_loss": 0.7424919,
      "learning_rate": 2.8850722156606207e-06,
      "loss": 0.76454842,
      "num_input_tokens_seen": 66734575,
      "step": 3106,
      "time_per_iteration": 2.674243688583374
    },
    {
      "auxiliary_loss_clip": 0.01171183,
      "auxiliary_loss_mlp": 0.01030642,
      "balance_loss_clip": 1.05207372,
      "balance_loss_mlp": 1.02232075,
      "epoch": 0.3735946612156556,
      "flos": 19714922409600.0,
      "grad_norm": 1.5450383144588031,
      "language_loss": 0.67305291,
      "learning_rate": 2.8843736051866252e-06,
      "loss": 0.6950711,
      "num_input_tokens_seen": 66753500,
      "step": 3107,
      "time_per_iteration": 3.5739047527313232
    },
    {
      "auxiliary_loss_clip": 0.01131353,
      "auxiliary_loss_mlp": 0.00763624,
      "balance_loss_clip": 1.04784453,
      "balance_loss_mlp": 1.00021279,
      "epoch": 0.3737149041062947,
      "flos": 23039604334080.0,
      "grad_norm": 1.5283885625990974,
      "language_loss": 0.6932227,
      "learning_rate": 2.8836748605538557e-06,
      "loss": 0.71217245,
      "num_input_tokens_seen": 66775140,
      "step": 3108,
      "time_per_iteration": 2.7971391677856445
    },
    {
      "auxiliary_loss_clip": 0.01167554,
      "auxiliary_loss_mlp": 0.01030753,
      "balance_loss_clip": 1.05192971,
      "balance_loss_mlp": 1.02243829,
      "epoch": 0.3738351469969338,
      "flos": 34678108483200.0,
      "grad_norm": 4.863172078664337,
      "language_loss": 0.63635671,
      "learning_rate": 2.882975981868313e-06,
      "loss": 0.6583398,
      "num_input_tokens_seen": 66795525,
      "step": 3109,
      "time_per_iteration": 2.8542168140411377
    },
    {
      "auxiliary_loss_clip": 0.01179868,
      "auxiliary_loss_mlp": 0.0102696,
      "balance_loss_clip": 1.05568862,
      "balance_loss_mlp": 1.01879418,
      "epoch": 0.3739553898875729,
      "flos": 43507967448960.0,
      "grad_norm": 2.3676359308169825,
      "language_loss": 0.68668866,
      "learning_rate": 2.882276969236016e-06,
      "loss": 0.70875692,
      "num_input_tokens_seen": 66816885,
      "step": 3110,
      "time_per_iteration": 2.868337392807007
    },
    {
      "auxiliary_loss_clip": 0.01158373,
      "auxiliary_loss_mlp": 0.01035216,
      "balance_loss_clip": 1.05120826,
      "balance_loss_mlp": 1.02680612,
      "epoch": 0.374075632778212,
      "flos": 12856487448960.0,
      "grad_norm": 1.881200579700236,
      "language_loss": 0.762061,
      "learning_rate": 2.881577822763005e-06,
      "loss": 0.78399694,
      "num_input_tokens_seen": 66834835,
      "step": 3111,
      "time_per_iteration": 3.5591001510620117
    },
    {
      "auxiliary_loss_clip": 0.01176791,
      "auxiliary_loss_mlp": 0.01023758,
      "balance_loss_clip": 1.05259013,
      "balance_loss_mlp": 1.0156157,
      "epoch": 0.3741958756688511,
      "flos": 26024031699840.0,
      "grad_norm": 1.8963390414213954,
      "language_loss": 0.87422931,
      "learning_rate": 2.880878542555338e-06,
      "loss": 0.89623475,
      "num_input_tokens_seen": 66852600,
      "step": 3112,
      "time_per_iteration": 3.618725061416626
    },
    {
      "auxiliary_loss_clip": 0.01193453,
      "auxiliary_loss_mlp": 0.01028611,
      "balance_loss_clip": 1.05467439,
      "balance_loss_mlp": 1.01958048,
      "epoch": 0.37431611855949015,
      "flos": 21433894652160.0,
      "grad_norm": 1.9774108060355124,
      "language_loss": 0.80533004,
      "learning_rate": 2.8801791287190976e-06,
      "loss": 0.82755065,
      "num_input_tokens_seen": 66870595,
      "step": 3113,
      "time_per_iteration": 2.633082866668701
    },
    {
      "auxiliary_loss_clip": 0.01181986,
      "auxiliary_loss_mlp": 0.01030075,
      "balance_loss_clip": 1.05386019,
      "balance_loss_mlp": 1.0211339,
      "epoch": 0.37443636145012926,
      "flos": 24207096090240.0,
      "grad_norm": 4.262007071234999,
      "language_loss": 0.86295319,
      "learning_rate": 2.8794795813603817e-06,
      "loss": 0.88507378,
      "num_input_tokens_seen": 66886060,
      "step": 3114,
      "time_per_iteration": 2.662886381149292
    },
    {
      "auxiliary_loss_clip": 0.01181132,
      "auxiliary_loss_mlp": 0.01032354,
      "balance_loss_clip": 1.05161321,
      "balance_loss_mlp": 1.02403355,
      "epoch": 0.3745566043407684,
      "flos": 15378601841280.0,
      "grad_norm": 1.8196089630324046,
      "language_loss": 0.8162294,
      "learning_rate": 2.878779900585314e-06,
      "loss": 0.8383643,
      "num_input_tokens_seen": 66903900,
      "step": 3115,
      "time_per_iteration": 2.6496787071228027
    },
    {
      "auxiliary_loss_clip": 0.01172023,
      "auxiliary_loss_mlp": 0.01031443,
      "balance_loss_clip": 1.05440211,
      "balance_loss_mlp": 1.02310419,
      "epoch": 0.37467684723140743,
      "flos": 24608218245120.0,
      "grad_norm": 1.4904887391934618,
      "language_loss": 0.75238293,
      "learning_rate": 2.8780800865000336e-06,
      "loss": 0.77441758,
      "num_input_tokens_seen": 66925210,
      "step": 3116,
      "time_per_iteration": 2.7056055068969727
    },
    {
      "auxiliary_loss_clip": 0.0108247,
      "auxiliary_loss_mlp": 0.01000789,
      "balance_loss_clip": 1.02050471,
      "balance_loss_mlp": 0.9992038,
      "epoch": 0.37479709012204654,
      "flos": 64377491610240.0,
      "grad_norm": 0.9760962482515769,
      "language_loss": 0.5918715,
      "learning_rate": 2.877380139210702e-06,
      "loss": 0.61270404,
      "num_input_tokens_seen": 66983880,
      "step": 3117,
      "time_per_iteration": 3.275131940841675
    },
    {
      "auxiliary_loss_clip": 0.01157959,
      "auxiliary_loss_mlp": 0.01035616,
      "balance_loss_clip": 1.0536319,
      "balance_loss_mlp": 1.02681828,
      "epoch": 0.37491733301268565,
      "flos": 23803962773760.0,
      "grad_norm": 1.807271101212916,
      "language_loss": 0.76701802,
      "learning_rate": 2.876680058823501e-06,
      "loss": 0.78895378,
      "num_input_tokens_seen": 67004280,
      "step": 3118,
      "time_per_iteration": 2.727661609649658
    },
    {
      "auxiliary_loss_clip": 0.01156902,
      "auxiliary_loss_mlp": 0.01031574,
      "balance_loss_clip": 1.05291247,
      "balance_loss_mlp": 1.02253234,
      "epoch": 0.3750375759033247,
      "flos": 32160950167680.0,
      "grad_norm": 2.1904064825227945,
      "language_loss": 0.66449189,
      "learning_rate": 2.8759798454446314e-06,
      "loss": 0.68637669,
      "num_input_tokens_seen": 67027445,
      "step": 3119,
      "time_per_iteration": 2.7748589515686035
    },
    {
      "auxiliary_loss_clip": 0.01182183,
      "auxiliary_loss_mlp": 0.01034159,
      "balance_loss_clip": 1.05466759,
      "balance_loss_mlp": 1.02561164,
      "epoch": 0.3751578187939638,
      "flos": 23367791923200.0,
      "grad_norm": 1.7697248166125905,
      "language_loss": 0.81505036,
      "learning_rate": 2.8752794991803173e-06,
      "loss": 0.83721375,
      "num_input_tokens_seen": 67045130,
      "step": 3120,
      "time_per_iteration": 2.7114031314849854
    },
    {
      "auxiliary_loss_clip": 0.01163649,
      "auxiliary_loss_mlp": 0.01027532,
      "balance_loss_clip": 1.05263603,
      "balance_loss_mlp": 1.01887143,
      "epoch": 0.37527806168460287,
      "flos": 14605731878400.0,
      "grad_norm": 2.026411523433748,
      "language_loss": 0.75176764,
      "learning_rate": 2.8745790201367976e-06,
      "loss": 0.77367949,
      "num_input_tokens_seen": 67060885,
      "step": 3121,
      "time_per_iteration": 2.587972640991211
    },
    {
      "auxiliary_loss_clip": 0.01196371,
      "auxiliary_loss_mlp": 0.01032436,
      "balance_loss_clip": 1.05602455,
      "balance_loss_mlp": 1.0232389,
      "epoch": 0.375398304575242,
      "flos": 26390823431040.0,
      "grad_norm": 8.577034695704883,
      "language_loss": 0.84443271,
      "learning_rate": 2.8738784084203373e-06,
      "loss": 0.86672074,
      "num_input_tokens_seen": 67080960,
      "step": 3122,
      "time_per_iteration": 2.672083854675293
    },
    {
      "auxiliary_loss_clip": 0.01155512,
      "auxiliary_loss_mlp": 0.01030859,
      "balance_loss_clip": 1.04764867,
      "balance_loss_mlp": 1.02259135,
      "epoch": 0.3755185474658811,
      "flos": 22236605838720.0,
      "grad_norm": 1.7553019653810469,
      "language_loss": 0.79029465,
      "learning_rate": 2.873177664137216e-06,
      "loss": 0.81215835,
      "num_input_tokens_seen": 67101890,
      "step": 3123,
      "time_per_iteration": 2.6508126258850098
    },
    {
      "auxiliary_loss_clip": 0.0114958,
      "auxiliary_loss_mlp": 0.01036668,
      "balance_loss_clip": 1.05331016,
      "balance_loss_mlp": 1.02800751,
      "epoch": 0.37563879035652015,
      "flos": 30812935633920.0,
      "grad_norm": 1.7501432794802991,
      "language_loss": 0.69329947,
      "learning_rate": 2.8724767873937384e-06,
      "loss": 0.71516192,
      "num_input_tokens_seen": 67126010,
      "step": 3124,
      "time_per_iteration": 2.8310606479644775
    },
    {
      "auxiliary_loss_clip": 0.01164438,
      "auxiliary_loss_mlp": 0.01030889,
      "balance_loss_clip": 1.05159676,
      "balance_loss_mlp": 1.02275908,
      "epoch": 0.37575903324715926,
      "flos": 20773533064320.0,
      "grad_norm": 2.581565948390894,
      "language_loss": 0.87481266,
      "learning_rate": 2.871775778296225e-06,
      "loss": 0.89676595,
      "num_input_tokens_seen": 67143100,
      "step": 3125,
      "time_per_iteration": 2.6660566329956055
    },
    {
      "auxiliary_loss_clip": 0.01180991,
      "auxiliary_loss_mlp": 0.01027032,
      "balance_loss_clip": 1.05578864,
      "balance_loss_mlp": 1.01729882,
      "epoch": 0.37587927613779837,
      "flos": 18697681244160.0,
      "grad_norm": 2.0610033912095265,
      "language_loss": 0.78552485,
      "learning_rate": 2.8710746369510196e-06,
      "loss": 0.80760509,
      "num_input_tokens_seen": 67161085,
      "step": 3126,
      "time_per_iteration": 2.667849540710449
    },
    {
      "auxiliary_loss_clip": 0.01162476,
      "auxiliary_loss_mlp": 0.01029073,
      "balance_loss_clip": 1.05355203,
      "balance_loss_mlp": 1.01976824,
      "epoch": 0.3759995190284374,
      "flos": 13624796384640.0,
      "grad_norm": 2.6049600964994912,
      "language_loss": 0.83501852,
      "learning_rate": 2.8703733634644846e-06,
      "loss": 0.85693395,
      "num_input_tokens_seen": 67175840,
      "step": 3127,
      "time_per_iteration": 2.6155519485473633
    },
    {
      "auxiliary_loss_clip": 0.01198274,
      "auxiliary_loss_mlp": 0.01027662,
      "balance_loss_clip": 1.05948746,
      "balance_loss_mlp": 1.0197407,
      "epoch": 0.37611976191907653,
      "flos": 20484847457280.0,
      "grad_norm": 1.830247754910172,
      "language_loss": 0.79140902,
      "learning_rate": 2.869671957943002e-06,
      "loss": 0.81366837,
      "num_input_tokens_seen": 67194995,
      "step": 3128,
      "time_per_iteration": 2.6157338619232178
    },
    {
      "auxiliary_loss_clip": 0.01160336,
      "auxiliary_loss_mlp": 0.01028825,
      "balance_loss_clip": 1.05487418,
      "balance_loss_mlp": 1.02045083,
      "epoch": 0.37624000480971564,
      "flos": 21141797253120.0,
      "grad_norm": 1.7846701141967873,
      "language_loss": 0.74643123,
      "learning_rate": 2.8689704204929747e-06,
      "loss": 0.76832289,
      "num_input_tokens_seen": 67214175,
      "step": 3129,
      "time_per_iteration": 2.710440158843994
    },
    {
      "auxiliary_loss_clip": 0.01195293,
      "auxiliary_loss_mlp": 0.01034164,
      "balance_loss_clip": 1.05632782,
      "balance_loss_mlp": 1.02569413,
      "epoch": 0.3763602477003547,
      "flos": 22564470205440.0,
      "grad_norm": 1.958982572280581,
      "language_loss": 0.81077927,
      "learning_rate": 2.8682687512208253e-06,
      "loss": 0.83307385,
      "num_input_tokens_seen": 67233185,
      "step": 3130,
      "time_per_iteration": 3.753727674484253
    },
    {
      "auxiliary_loss_clip": 0.01185883,
      "auxiliary_loss_mlp": 0.0103215,
      "balance_loss_clip": 1.05420303,
      "balance_loss_mlp": 1.02386546,
      "epoch": 0.3764804905909938,
      "flos": 27526857851520.0,
      "grad_norm": 1.912458500436718,
      "language_loss": 0.80656719,
      "learning_rate": 2.8675669502329972e-06,
      "loss": 0.82874745,
      "num_input_tokens_seen": 67254715,
      "step": 3131,
      "time_per_iteration": 2.761674165725708
    },
    {
      "auxiliary_loss_clip": 0.01178234,
      "auxiliary_loss_mlp": 0.00763508,
      "balance_loss_clip": 1.05333257,
      "balance_loss_mlp": 1.00017619,
      "epoch": 0.3766007334816329,
      "flos": 22528092706560.0,
      "grad_norm": 2.3234397376545615,
      "language_loss": 0.86246705,
      "learning_rate": 2.866865017635952e-06,
      "loss": 0.88188446,
      "num_input_tokens_seen": 67272535,
      "step": 3132,
      "time_per_iteration": 2.8670501708984375
    },
    {
      "auxiliary_loss_clip": 0.01153393,
      "auxiliary_loss_mlp": 0.01031577,
      "balance_loss_clip": 1.05634403,
      "balance_loss_mlp": 1.02311897,
      "epoch": 0.376720976372272,
      "flos": 25957166532480.0,
      "grad_norm": 1.5806862004513775,
      "language_loss": 0.79378366,
      "learning_rate": 2.866162953536174e-06,
      "loss": 0.8156333,
      "num_input_tokens_seen": 67293505,
      "step": 3133,
      "time_per_iteration": 3.633573532104492
    },
    {
      "auxiliary_loss_clip": 0.01164897,
      "auxiliary_loss_mlp": 0.00763422,
      "balance_loss_clip": 1.05125606,
      "balance_loss_mlp": 1.00014925,
      "epoch": 0.3768412192629111,
      "flos": 18041162411520.0,
      "grad_norm": 1.8028845128396391,
      "language_loss": 0.74989998,
      "learning_rate": 2.8654607580401634e-06,
      "loss": 0.76918316,
      "num_input_tokens_seen": 67313240,
      "step": 3134,
      "time_per_iteration": 2.6736795902252197
    },
    {
      "auxiliary_loss_clip": 0.01080224,
      "auxiliary_loss_mlp": 0.01000205,
      "balance_loss_clip": 1.02035439,
      "balance_loss_mlp": 0.99858361,
      "epoch": 0.3769614621535502,
      "flos": 62989472304000.0,
      "grad_norm": 0.9603080408296224,
      "language_loss": 0.65157372,
      "learning_rate": 2.8647584312544446e-06,
      "loss": 0.67237806,
      "num_input_tokens_seen": 67378445,
      "step": 3135,
      "time_per_iteration": 3.2186474800109863
    },
    {
      "auxiliary_loss_clip": 0.01143374,
      "auxiliary_loss_mlp": 0.00762758,
      "balance_loss_clip": 1.04701495,
      "balance_loss_mlp": 1.00015044,
      "epoch": 0.37708170504418925,
      "flos": 23661685002240.0,
      "grad_norm": 1.5116638911097344,
      "language_loss": 0.85566509,
      "learning_rate": 2.864055973285559e-06,
      "loss": 0.87472641,
      "num_input_tokens_seen": 67400445,
      "step": 3136,
      "time_per_iteration": 2.781022787094116
    },
    {
      "auxiliary_loss_clip": 0.01151374,
      "auxiliary_loss_mlp": 0.01033672,
      "balance_loss_clip": 1.04845476,
      "balance_loss_mlp": 1.02551782,
      "epoch": 0.37720194793482836,
      "flos": 24423170353920.0,
      "grad_norm": 1.8348489084765285,
      "language_loss": 0.86465299,
      "learning_rate": 2.8633533842400698e-06,
      "loss": 0.88650346,
      "num_input_tokens_seen": 67420645,
      "step": 3137,
      "time_per_iteration": 3.629549980163574
    },
    {
      "auxiliary_loss_clip": 0.01181924,
      "auxiliary_loss_mlp": 0.00764261,
      "balance_loss_clip": 1.05563366,
      "balance_loss_mlp": 1.00024498,
      "epoch": 0.3773221908254674,
      "flos": 20996502739200.0,
      "grad_norm": 1.9327771069270452,
      "language_loss": 0.77805209,
      "learning_rate": 2.862650664224558e-06,
      "loss": 0.79751396,
      "num_input_tokens_seen": 67439495,
      "step": 3138,
      "time_per_iteration": 3.5395736694335938
    },
    {
      "auxiliary_loss_clip": 0.01178925,
      "auxiliary_loss_mlp": 0.01030092,
      "balance_loss_clip": 1.0574317,
      "balance_loss_mlp": 1.02205157,
      "epoch": 0.37744243371610653,
      "flos": 37631724958080.0,
      "grad_norm": 1.6208341410494207,
      "language_loss": 0.6969803,
      "learning_rate": 2.861947813345627e-06,
      "loss": 0.71907043,
      "num_input_tokens_seen": 67462195,
      "step": 3139,
      "time_per_iteration": 2.8611836433410645
    },
    {
      "auxiliary_loss_clip": 0.01193407,
      "auxiliary_loss_mlp": 0.00764332,
      "balance_loss_clip": 1.05504501,
      "balance_loss_mlp": 1.00018239,
      "epoch": 0.37756267660674564,
      "flos": 26140526484480.0,
      "grad_norm": 1.579126424608873,
      "language_loss": 0.72380638,
      "learning_rate": 2.8612448317098974e-06,
      "loss": 0.74338377,
      "num_input_tokens_seen": 67482530,
      "step": 3140,
      "time_per_iteration": 2.6237947940826416
    },
    {
      "auxiliary_loss_clip": 0.01149032,
      "auxiliary_loss_mlp": 0.00764152,
      "balance_loss_clip": 1.0475564,
      "balance_loss_mlp": 1.00011706,
      "epoch": 0.3776829194973847,
      "flos": 19427888828160.0,
      "grad_norm": 1.939347168214533,
      "language_loss": 0.83495647,
      "learning_rate": 2.8605417194240114e-06,
      "loss": 0.85408831,
      "num_input_tokens_seen": 67500890,
      "step": 3141,
      "time_per_iteration": 2.757746934890747
    },
    {
      "auxiliary_loss_clip": 0.01171696,
      "auxiliary_loss_mlp": 0.01031846,
      "balance_loss_clip": 1.05080664,
      "balance_loss_mlp": 1.0235368,
      "epoch": 0.3778031623880238,
      "flos": 17382309194880.0,
      "grad_norm": 1.8482666696067072,
      "language_loss": 0.79258883,
      "learning_rate": 2.8598384765946315e-06,
      "loss": 0.81462419,
      "num_input_tokens_seen": 67519545,
      "step": 3142,
      "time_per_iteration": 2.6092066764831543
    },
    {
      "auxiliary_loss_clip": 0.01188558,
      "auxiliary_loss_mlp": 0.01031425,
      "balance_loss_clip": 1.05177474,
      "balance_loss_mlp": 1.02362323,
      "epoch": 0.3779234052786629,
      "flos": 27125843437440.0,
      "grad_norm": 2.3725289941613914,
      "language_loss": 0.71881294,
      "learning_rate": 2.8591351033284377e-06,
      "loss": 0.74101281,
      "num_input_tokens_seen": 67539275,
      "step": 3143,
      "time_per_iteration": 2.7344861030578613
    },
    {
      "auxiliary_loss_clip": 0.01180097,
      "auxiliary_loss_mlp": 0.01030203,
      "balance_loss_clip": 1.05203199,
      "balance_loss_mlp": 1.02157807,
      "epoch": 0.37804364816930197,
      "flos": 19682639061120.0,
      "grad_norm": 2.341116283893558,
      "language_loss": 0.83755016,
      "learning_rate": 2.8584315997321325e-06,
      "loss": 0.85965323,
      "num_input_tokens_seen": 67558280,
      "step": 3144,
      "time_per_iteration": 2.6385817527770996
    },
    {
      "auxiliary_loss_clip": 0.01191815,
      "auxiliary_loss_mlp": 0.01026672,
      "balance_loss_clip": 1.05338979,
      "balance_loss_mlp": 1.01800001,
      "epoch": 0.3781638910599411,
      "flos": 22702905221760.0,
      "grad_norm": 2.332977081904027,
      "language_loss": 0.777179,
      "learning_rate": 2.8577279659124356e-06,
      "loss": 0.79936385,
      "num_input_tokens_seen": 67575955,
      "step": 3145,
      "time_per_iteration": 2.627830743789673
    },
    {
      "auxiliary_loss_clip": 0.011725,
      "auxiliary_loss_mlp": 0.01032591,
      "balance_loss_clip": 1.0506736,
      "balance_loss_mlp": 1.02455044,
      "epoch": 0.3782841339505802,
      "flos": 14647604158080.0,
      "grad_norm": 1.8916992724775787,
      "language_loss": 0.83538592,
      "learning_rate": 2.857024201976089e-06,
      "loss": 0.8574369,
      "num_input_tokens_seen": 67593515,
      "step": 3146,
      "time_per_iteration": 2.5955352783203125
    },
    {
      "auxiliary_loss_clip": 0.01160654,
      "auxiliary_loss_mlp": 0.0103154,
      "balance_loss_clip": 1.05041838,
      "balance_loss_mlp": 1.0220983,
      "epoch": 0.37840437684121925,
      "flos": 32818223185920.0,
      "grad_norm": 1.8290199561080984,
      "language_loss": 0.73117507,
      "learning_rate": 2.8563203080298516e-06,
      "loss": 0.75309694,
      "num_input_tokens_seen": 67614290,
      "step": 3147,
      "time_per_iteration": 2.755469799041748
    },
    {
      "auxiliary_loss_clip": 0.01161636,
      "auxiliary_loss_mlp": 0.00762857,
      "balance_loss_clip": 1.05088592,
      "balance_loss_mlp": 1.00016737,
      "epoch": 0.37852461973185836,
      "flos": 18369206346240.0,
      "grad_norm": 4.5001346973546195,
      "language_loss": 0.89568889,
      "learning_rate": 2.855616284180505e-06,
      "loss": 0.9149338,
      "num_input_tokens_seen": 67631340,
      "step": 3148,
      "time_per_iteration": 2.6243646144866943
    },
    {
      "auxiliary_loss_clip": 0.01085318,
      "auxiliary_loss_mlp": 0.01001235,
      "balance_loss_clip": 1.02225399,
      "balance_loss_mlp": 0.99950641,
      "epoch": 0.37864486262249747,
      "flos": 59500680117120.0,
      "grad_norm": 0.8739714477496746,
      "language_loss": 0.6610744,
      "learning_rate": 2.8549121305348477e-06,
      "loss": 0.68193984,
      "num_input_tokens_seen": 67691125,
      "step": 3149,
      "time_per_iteration": 3.237043857574463
    },
    {
      "auxiliary_loss_clip": 0.01175118,
      "auxiliary_loss_mlp": 0.01028172,
      "balance_loss_clip": 1.05047703,
      "balance_loss_mlp": 1.02029216,
      "epoch": 0.3787651055131365,
      "flos": 23363015414400.0,
      "grad_norm": 2.505306319387676,
      "language_loss": 0.8317492,
      "learning_rate": 2.8542078471997006e-06,
      "loss": 0.85378206,
      "num_input_tokens_seen": 67708740,
      "step": 3150,
      "time_per_iteration": 2.6768908500671387
    },
    {
      "auxiliary_loss_clip": 0.01174134,
      "auxiliary_loss_mlp": 0.01033842,
      "balance_loss_clip": 1.0501411,
      "balance_loss_mlp": 1.02586699,
      "epoch": 0.37888534840377563,
      "flos": 24601394661120.0,
      "grad_norm": 1.8076104941695537,
      "language_loss": 0.76039851,
      "learning_rate": 2.8535034342819013e-06,
      "loss": 0.78247827,
      "num_input_tokens_seen": 67726150,
      "step": 3151,
      "time_per_iteration": 2.6127681732177734
    },
    {
      "auxiliary_loss_clip": 0.01190971,
      "auxiliary_loss_mlp": 0.0103206,
      "balance_loss_clip": 1.05401134,
      "balance_loss_mlp": 1.02323842,
      "epoch": 0.37900559129441475,
      "flos": 23986891762560.0,
      "grad_norm": 1.5942082090751741,
      "language_loss": 0.72626925,
      "learning_rate": 2.85279889188831e-06,
      "loss": 0.74849951,
      "num_input_tokens_seen": 67746525,
      "step": 3152,
      "time_per_iteration": 2.697158098220825
    },
    {
      "auxiliary_loss_clip": 0.01145896,
      "auxiliary_loss_mlp": 0.0103357,
      "balance_loss_clip": 1.04606307,
      "balance_loss_mlp": 1.02499318,
      "epoch": 0.3791258341850538,
      "flos": 24644667571200.0,
      "grad_norm": 1.693193432155523,
      "language_loss": 0.81015068,
      "learning_rate": 2.852094220125805e-06,
      "loss": 0.83194542,
      "num_input_tokens_seen": 67766035,
      "step": 3153,
      "time_per_iteration": 2.8143136501312256
    },
    {
      "auxiliary_loss_clip": 0.01178232,
      "auxiliary_loss_mlp": 0.01036513,
      "balance_loss_clip": 1.05481744,
      "balance_loss_mlp": 1.02698863,
      "epoch": 0.3792460770756929,
      "flos": 17420841509760.0,
      "grad_norm": 2.025667632126982,
      "language_loss": 0.70957363,
      "learning_rate": 2.8513894191012846e-06,
      "loss": 0.73172098,
      "num_input_tokens_seen": 67785015,
      "step": 3154,
      "time_per_iteration": 2.675022840499878
    },
    {
      "auxiliary_loss_clip": 0.01191131,
      "auxiliary_loss_mlp": 0.01037467,
      "balance_loss_clip": 1.05325866,
      "balance_loss_mlp": 1.02912807,
      "epoch": 0.37936631996633197,
      "flos": 24206557386240.0,
      "grad_norm": 1.4821687563550758,
      "language_loss": 0.78887057,
      "learning_rate": 2.8506844889216664e-06,
      "loss": 0.81115657,
      "num_input_tokens_seen": 67804400,
      "step": 3155,
      "time_per_iteration": 2.6712646484375
    },
    {
      "auxiliary_loss_clip": 0.01075467,
      "auxiliary_loss_mlp": 0.01008341,
      "balance_loss_clip": 1.01934993,
      "balance_loss_mlp": 1.00669563,
      "epoch": 0.3794865628569711,
      "flos": 70297114752000.0,
      "grad_norm": 0.8776545489840557,
      "language_loss": 0.62876135,
      "learning_rate": 2.849979429693887e-06,
      "loss": 0.64959943,
      "num_input_tokens_seen": 67865385,
      "step": 3156,
      "time_per_iteration": 3.279393434524536
    },
    {
      "auxiliary_loss_clip": 0.01190505,
      "auxiliary_loss_mlp": 0.01027062,
      "balance_loss_clip": 1.05276251,
      "balance_loss_mlp": 1.01773357,
      "epoch": 0.3796068057476102,
      "flos": 15779364860160.0,
      "grad_norm": 2.0104060466078453,
      "language_loss": 0.7457391,
      "learning_rate": 2.8492742415249042e-06,
      "loss": 0.76791471,
      "num_input_tokens_seen": 67883030,
      "step": 3157,
      "time_per_iteration": 3.729616165161133
    },
    {
      "auxiliary_loss_clip": 0.01189925,
      "auxiliary_loss_mlp": 0.01029088,
      "balance_loss_clip": 1.0529213,
      "balance_loss_mlp": 1.0209136,
      "epoch": 0.37972704863824924,
      "flos": 25191694771200.0,
      "grad_norm": 1.8394137088950613,
      "language_loss": 0.76417804,
      "learning_rate": 2.848568924521694e-06,
      "loss": 0.78636813,
      "num_input_tokens_seen": 67903810,
      "step": 3158,
      "time_per_iteration": 2.5990488529205322
    },
    {
      "auxiliary_loss_clip": 0.01168898,
      "auxiliary_loss_mlp": 0.01025913,
      "balance_loss_clip": 1.04888713,
      "balance_loss_mlp": 1.01736629,
      "epoch": 0.37984729152888835,
      "flos": 26210372480640.0,
      "grad_norm": 1.7320291602760682,
      "language_loss": 0.73592514,
      "learning_rate": 2.8478634787912526e-06,
      "loss": 0.7578733,
      "num_input_tokens_seen": 67921865,
      "step": 3159,
      "time_per_iteration": 3.6381661891937256
    },
    {
      "auxiliary_loss_clip": 0.01177326,
      "auxiliary_loss_mlp": 0.0103539,
      "balance_loss_clip": 1.05198169,
      "balance_loss_mlp": 1.02655649,
      "epoch": 0.37996753441952746,
      "flos": 25629302165760.0,
      "grad_norm": 2.5254537544887823,
      "language_loss": 0.7624349,
      "learning_rate": 2.847157904440596e-06,
      "loss": 0.78456199,
      "num_input_tokens_seen": 67941595,
      "step": 3160,
      "time_per_iteration": 2.6349852085113525
    },
    {
      "auxiliary_loss_clip": 0.01176252,
      "auxiliary_loss_mlp": 0.01031977,
      "balance_loss_clip": 1.05323398,
      "balance_loss_mlp": 1.02354884,
      "epoch": 0.3800877773101665,
      "flos": 20118414862080.0,
      "grad_norm": 1.4963191135842497,
      "language_loss": 0.74132907,
      "learning_rate": 2.846452201576759e-06,
      "loss": 0.76341128,
      "num_input_tokens_seen": 67960970,
      "step": 3161,
      "time_per_iteration": 2.590013027191162
    },
    {
      "auxiliary_loss_clip": 0.01074431,
      "auxiliary_loss_mlp": 0.01007792,
      "balance_loss_clip": 1.02203155,
      "balance_loss_mlp": 1.00623083,
      "epoch": 0.38020802020080563,
      "flos": 63053608037760.0,
      "grad_norm": 0.859587141454022,
      "language_loss": 0.62839091,
      "learning_rate": 2.845746370306795e-06,
      "loss": 0.64921319,
      "num_input_tokens_seen": 68026160,
      "step": 3162,
      "time_per_iteration": 3.3215761184692383
    },
    {
      "auxiliary_loss_clip": 0.01179961,
      "auxiliary_loss_mlp": 0.01028032,
      "balance_loss_clip": 1.05316126,
      "balance_loss_mlp": 1.01940739,
      "epoch": 0.38032826309144474,
      "flos": 21288420570240.0,
      "grad_norm": 1.8612755093614402,
      "language_loss": 0.78916025,
      "learning_rate": 2.84504041073778e-06,
      "loss": 0.8112402,
      "num_input_tokens_seen": 68044575,
      "step": 3163,
      "time_per_iteration": 3.5391719341278076
    },
    {
      "auxiliary_loss_clip": 0.01152676,
      "auxiliary_loss_mlp": 0.0103185,
      "balance_loss_clip": 1.04890847,
      "balance_loss_mlp": 1.02296352,
      "epoch": 0.3804485059820838,
      "flos": 18954119416320.0,
      "grad_norm": 1.7315060574761743,
      "language_loss": 0.79387081,
      "learning_rate": 2.844334322976806e-06,
      "loss": 0.81571603,
      "num_input_tokens_seen": 68064790,
      "step": 3164,
      "time_per_iteration": 3.597341537475586
    },
    {
      "auxiliary_loss_clip": 0.01140137,
      "auxiliary_loss_mlp": 0.01036157,
      "balance_loss_clip": 1.04583859,
      "balance_loss_mlp": 1.02670979,
      "epoch": 0.3805687488727229,
      "flos": 21833759831040.0,
      "grad_norm": 1.8899014346498153,
      "language_loss": 0.83460498,
      "learning_rate": 2.8436281071309866e-06,
      "loss": 0.85636783,
      "num_input_tokens_seen": 68083330,
      "step": 3165,
      "time_per_iteration": 2.736013412475586
    },
    {
      "auxiliary_loss_clip": 0.01057311,
      "auxiliary_loss_mlp": 0.01006465,
      "balance_loss_clip": 1.02118635,
      "balance_loss_mlp": 1.00487995,
      "epoch": 0.380688991763362,
      "flos": 58546209968640.0,
      "grad_norm": 0.7261134890584806,
      "language_loss": 0.52948314,
      "learning_rate": 2.842921763307455e-06,
      "loss": 0.55012089,
      "num_input_tokens_seen": 68146140,
      "step": 3166,
      "time_per_iteration": 3.25123929977417
    },
    {
      "auxiliary_loss_clip": 0.01156618,
      "auxiliary_loss_mlp": 0.0103385,
      "balance_loss_clip": 1.04935181,
      "balance_loss_mlp": 1.02524304,
      "epoch": 0.38080923465400107,
      "flos": 23799509487360.0,
      "grad_norm": 1.8580132402063838,
      "language_loss": 0.82377303,
      "learning_rate": 2.842215291613361e-06,
      "loss": 0.84567773,
      "num_input_tokens_seen": 68164520,
      "step": 3167,
      "time_per_iteration": 2.731501579284668
    },
    {
      "auxiliary_loss_clip": 0.01021834,
      "auxiliary_loss_mlp": 0.01004727,
      "balance_loss_clip": 1.02134824,
      "balance_loss_mlp": 1.00309372,
      "epoch": 0.3809294775446402,
      "flos": 54969866380800.0,
      "grad_norm": 0.8507008396179668,
      "language_loss": 0.5918625,
      "learning_rate": 2.8415086921558774e-06,
      "loss": 0.61212808,
      "num_input_tokens_seen": 68227945,
      "step": 3168,
      "time_per_iteration": 3.506775379180908
    },
    {
      "auxiliary_loss_clip": 0.01145487,
      "auxiliary_loss_mlp": 0.01028627,
      "balance_loss_clip": 1.04309106,
      "balance_loss_mlp": 1.01947737,
      "epoch": 0.38104972043527924,
      "flos": 24643697904000.0,
      "grad_norm": 1.531919155144609,
      "language_loss": 0.78677249,
      "learning_rate": 2.840801965042194e-06,
      "loss": 0.80851358,
      "num_input_tokens_seen": 68247405,
      "step": 3169,
      "time_per_iteration": 2.9969325065612793
    },
    {
      "auxiliary_loss_clip": 0.01151554,
      "auxiliary_loss_mlp": 0.01034526,
      "balance_loss_clip": 1.04648185,
      "balance_loss_mlp": 1.0257051,
      "epoch": 0.38116996332591835,
      "flos": 22856783086080.0,
      "grad_norm": 1.7433658373835499,
      "language_loss": 0.84023464,
      "learning_rate": 2.840095110379521e-06,
      "loss": 0.86209548,
      "num_input_tokens_seen": 68266925,
      "step": 3170,
      "time_per_iteration": 2.6760458946228027
    },
    {
      "auxiliary_loss_clip": 0.0104268,
      "auxiliary_loss_mlp": 0.01003421,
      "balance_loss_clip": 1.01907849,
      "balance_loss_mlp": 1.00187135,
      "epoch": 0.38129020621655746,
      "flos": 60836160804480.0,
      "grad_norm": 0.7519423721049622,
      "language_loss": 0.53887534,
      "learning_rate": 2.8393881282750884e-06,
      "loss": 0.5593363,
      "num_input_tokens_seen": 68329755,
      "step": 3171,
      "time_per_iteration": 3.236588716506958
    },
    {
      "auxiliary_loss_clip": 0.01163858,
      "auxiliary_loss_mlp": 0.01033599,
      "balance_loss_clip": 1.05182099,
      "balance_loss_mlp": 1.02472401,
      "epoch": 0.3814104491071965,
      "flos": 21648101408640.0,
      "grad_norm": 2.093053277990453,
      "language_loss": 0.7879312,
      "learning_rate": 2.838681018836144e-06,
      "loss": 0.80990571,
      "num_input_tokens_seen": 68347075,
      "step": 3172,
      "time_per_iteration": 2.6886661052703857
    },
    {
      "auxiliary_loss_clip": 0.01148791,
      "auxiliary_loss_mlp": 0.00763664,
      "balance_loss_clip": 1.04710937,
      "balance_loss_mlp": 1.00012398,
      "epoch": 0.3815306919978356,
      "flos": 19099090707840.0,
      "grad_norm": 2.01521044834459,
      "language_loss": 0.78790075,
      "learning_rate": 2.837973782169955e-06,
      "loss": 0.80702525,
      "num_input_tokens_seen": 68365450,
      "step": 3173,
      "time_per_iteration": 2.672671318054199
    },
    {
      "auxiliary_loss_clip": 0.01091197,
      "auxiliary_loss_mlp": 0.0100074,
      "balance_loss_clip": 1.02059817,
      "balance_loss_mlp": 0.99911898,
      "epoch": 0.38165093488847474,
      "flos": 67067918156160.0,
      "grad_norm": 0.8084814963839174,
      "language_loss": 0.59145617,
      "learning_rate": 2.8372664183838096e-06,
      "loss": 0.6123755,
      "num_input_tokens_seen": 68428470,
      "step": 3174,
      "time_per_iteration": 3.250335931777954
    },
    {
      "auxiliary_loss_clip": 0.01189428,
      "auxiliary_loss_mlp": 0.01034873,
      "balance_loss_clip": 1.05325031,
      "balance_loss_mlp": 1.02581882,
      "epoch": 0.3817711777791138,
      "flos": 22341105480960.0,
      "grad_norm": 2.186396916592698,
      "language_loss": 0.68656558,
      "learning_rate": 2.836558927585015e-06,
      "loss": 0.7088086,
      "num_input_tokens_seen": 68445440,
      "step": 3175,
      "time_per_iteration": 2.6622257232666016
    },
    {
      "auxiliary_loss_clip": 0.01178556,
      "auxiliary_loss_mlp": 0.0103522,
      "balance_loss_clip": 1.05267477,
      "balance_loss_mlp": 1.02621984,
      "epoch": 0.3818914206697529,
      "flos": 22820621068800.0,
      "grad_norm": 2.1268316864135444,
      "language_loss": 0.82605207,
      "learning_rate": 2.8358513098808957e-06,
      "loss": 0.84818983,
      "num_input_tokens_seen": 68465755,
      "step": 3176,
      "time_per_iteration": 2.616715908050537
    },
    {
      "auxiliary_loss_clip": 0.0112301,
      "auxiliary_loss_mlp": 0.01032404,
      "balance_loss_clip": 1.04748642,
      "balance_loss_mlp": 1.02329612,
      "epoch": 0.382011663560392,
      "flos": 24386074583040.0,
      "grad_norm": 1.7610499942848132,
      "language_loss": 0.77138519,
      "learning_rate": 2.835143565378798e-06,
      "loss": 0.79293931,
      "num_input_tokens_seen": 68486220,
      "step": 3177,
      "time_per_iteration": 2.777756452560425
    },
    {
      "auxiliary_loss_clip": 0.01112294,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 1.0433892,
      "balance_loss_mlp": 1.01750863,
      "epoch": 0.38213190645103107,
      "flos": 21981568296960.0,
      "grad_norm": 2.5177384145573107,
      "language_loss": 0.78591949,
      "learning_rate": 2.8344356941860847e-06,
      "loss": 0.80731654,
      "num_input_tokens_seen": 68505850,
      "step": 3178,
      "time_per_iteration": 2.7951464653015137
    },
    {
      "auxiliary_loss_clip": 0.01141974,
      "auxiliary_loss_mlp": 0.01030953,
      "balance_loss_clip": 1.04812407,
      "balance_loss_mlp": 1.02163696,
      "epoch": 0.3822521493416702,
      "flos": 35516945773440.0,
      "grad_norm": 2.6082147227891723,
      "language_loss": 0.66063166,
      "learning_rate": 2.8337276964101403e-06,
      "loss": 0.68236089,
      "num_input_tokens_seen": 68526290,
      "step": 3179,
      "time_per_iteration": 2.833160400390625
    },
    {
      "auxiliary_loss_clip": 0.01177466,
      "auxiliary_loss_mlp": 0.01034308,
      "balance_loss_clip": 1.0521301,
      "balance_loss_mlp": 1.02604663,
      "epoch": 0.3823723922323093,
      "flos": 21069904181760.0,
      "grad_norm": 1.9074507318807794,
      "language_loss": 0.76716292,
      "learning_rate": 2.833019572158367e-06,
      "loss": 0.78928065,
      "num_input_tokens_seen": 68544725,
      "step": 3180,
      "time_per_iteration": 2.648097038269043
    },
    {
      "auxiliary_loss_clip": 0.01161036,
      "auxiliary_loss_mlp": 0.01029222,
      "balance_loss_clip": 1.04947424,
      "balance_loss_mlp": 1.02096105,
      "epoch": 0.38249263512294834,
      "flos": 19789149864960.0,
      "grad_norm": 1.7703468887687286,
      "language_loss": 0.80145693,
      "learning_rate": 2.8323113215381872e-06,
      "loss": 0.82335949,
      "num_input_tokens_seen": 68563070,
      "step": 3181,
      "time_per_iteration": 2.6935696601867676
    },
    {
      "auxiliary_loss_clip": 0.01146372,
      "auxiliary_loss_mlp": 0.01038847,
      "balance_loss_clip": 1.05015421,
      "balance_loss_mlp": 1.02958477,
      "epoch": 0.38261287801358745,
      "flos": 21433930565760.0,
      "grad_norm": 1.827708278738902,
      "language_loss": 0.76338065,
      "learning_rate": 2.831602944657042e-06,
      "loss": 0.78523284,
      "num_input_tokens_seen": 68581150,
      "step": 3182,
      "time_per_iteration": 2.758544683456421
    },
    {
      "auxiliary_loss_clip": 0.01169469,
      "auxiliary_loss_mlp": 0.01032898,
      "balance_loss_clip": 1.05042887,
      "balance_loss_mlp": 1.02417779,
      "epoch": 0.38273312090422656,
      "flos": 21981568296960.0,
      "grad_norm": 2.2093715133986764,
      "language_loss": 0.74496114,
      "learning_rate": 2.830894441622391e-06,
      "loss": 0.76698482,
      "num_input_tokens_seen": 68597800,
      "step": 3183,
      "time_per_iteration": 3.6180737018585205
    },
    {
      "auxiliary_loss_clip": 0.01142653,
      "auxiliary_loss_mlp": 0.00764054,
      "balance_loss_clip": 1.04547572,
      "balance_loss_mlp": 1.00016129,
      "epoch": 0.3828533637948656,
      "flos": 24790895838720.0,
      "grad_norm": 1.8479732487857226,
      "language_loss": 0.79576862,
      "learning_rate": 2.8301858125417134e-06,
      "loss": 0.81483567,
      "num_input_tokens_seen": 68617640,
      "step": 3184,
      "time_per_iteration": 2.7778563499450684
    },
    {
      "auxiliary_loss_clip": 0.01165789,
      "auxiliary_loss_mlp": 0.01030199,
      "balance_loss_clip": 1.05384326,
      "balance_loss_mlp": 1.02206266,
      "epoch": 0.38297360668550473,
      "flos": 22455445449600.0,
      "grad_norm": 1.9828484774405977,
      "language_loss": 0.73886502,
      "learning_rate": 2.8294770575225082e-06,
      "loss": 0.76082492,
      "num_input_tokens_seen": 68637770,
      "step": 3185,
      "time_per_iteration": 3.6044039726257324
    },
    {
      "auxiliary_loss_clip": 0.01178637,
      "auxiliary_loss_mlp": 0.01038046,
      "balance_loss_clip": 1.05493069,
      "balance_loss_mlp": 1.0291115,
      "epoch": 0.3830938495761438,
      "flos": 24896903852160.0,
      "grad_norm": 1.696876237179694,
      "language_loss": 0.83492744,
      "learning_rate": 2.828768176672293e-06,
      "loss": 0.85709429,
      "num_input_tokens_seen": 68656885,
      "step": 3186,
      "time_per_iteration": 2.6498985290527344
    },
    {
      "auxiliary_loss_clip": 0.01146476,
      "auxiliary_loss_mlp": 0.01030106,
      "balance_loss_clip": 1.04659224,
      "balance_loss_mlp": 1.02045619,
      "epoch": 0.3832140924667829,
      "flos": 33036236784000.0,
      "grad_norm": 1.910138676478638,
      "language_loss": 0.71559364,
      "learning_rate": 2.8280591700986044e-06,
      "loss": 0.73735946,
      "num_input_tokens_seen": 68678750,
      "step": 3187,
      "time_per_iteration": 2.7926204204559326
    },
    {
      "auxiliary_loss_clip": 0.01163029,
      "auxiliary_loss_mlp": 0.0102931,
      "balance_loss_clip": 1.04670823,
      "balance_loss_mlp": 1.0209415,
      "epoch": 0.383334335357422,
      "flos": 31903721896320.0,
      "grad_norm": 1.7038827706281723,
      "language_loss": 0.74809444,
      "learning_rate": 2.827350037908999e-06,
      "loss": 0.77001786,
      "num_input_tokens_seen": 68698190,
      "step": 3188,
      "time_per_iteration": 2.7096970081329346
    },
    {
      "auxiliary_loss_clip": 0.01152773,
      "auxiliary_loss_mlp": 0.01035571,
      "balance_loss_clip": 1.0484848,
      "balance_loss_mlp": 1.02601624,
      "epoch": 0.38345457824806106,
      "flos": 19791915212160.0,
      "grad_norm": 3.8511360269869352,
      "language_loss": 0.79260814,
      "learning_rate": 2.8266407802110496e-06,
      "loss": 0.81449157,
      "num_input_tokens_seen": 68716445,
      "step": 3189,
      "time_per_iteration": 3.6310768127441406
    },
    {
      "auxiliary_loss_clip": 0.01109897,
      "auxiliary_loss_mlp": 0.01033811,
      "balance_loss_clip": 1.042243,
      "balance_loss_mlp": 1.02415538,
      "epoch": 0.3835748211387002,
      "flos": 22419391173120.0,
      "grad_norm": 1.982733387132145,
      "language_loss": 0.76318955,
      "learning_rate": 2.8259313971123515e-06,
      "loss": 0.7846266,
      "num_input_tokens_seen": 68737565,
      "step": 3190,
      "time_per_iteration": 3.7357072830200195
    },
    {
      "auxiliary_loss_clip": 0.01174049,
      "auxiliary_loss_mlp": 0.01036263,
      "balance_loss_clip": 1.05364466,
      "balance_loss_mlp": 1.02781081,
      "epoch": 0.3836950640293393,
      "flos": 25118436983040.0,
      "grad_norm": 1.5134947921406596,
      "language_loss": 0.78702652,
      "learning_rate": 2.8252218887205166e-06,
      "loss": 0.8091296,
      "num_input_tokens_seen": 68758255,
      "step": 3191,
      "time_per_iteration": 2.6306204795837402
    },
    {
      "auxiliary_loss_clip": 0.01121831,
      "auxiliary_loss_mlp": 0.01028886,
      "balance_loss_clip": 1.04651868,
      "balance_loss_mlp": 1.02034497,
      "epoch": 0.38381530691997834,
      "flos": 21799213925760.0,
      "grad_norm": 3.1332192380506236,
      "language_loss": 0.80661613,
      "learning_rate": 2.824512255143178e-06,
      "loss": 0.82812333,
      "num_input_tokens_seen": 68777490,
      "step": 3192,
      "time_per_iteration": 2.7528295516967773
    },
    {
      "auxiliary_loss_clip": 0.01153801,
      "auxiliary_loss_mlp": 0.01029139,
      "balance_loss_clip": 1.05056739,
      "balance_loss_mlp": 1.02078819,
      "epoch": 0.38393554981061745,
      "flos": 21252689516160.0,
      "grad_norm": 1.7110880237153903,
      "language_loss": 0.79546571,
      "learning_rate": 2.8238024964879855e-06,
      "loss": 0.81729513,
      "num_input_tokens_seen": 68798385,
      "step": 3193,
      "time_per_iteration": 2.7326602935791016
    },
    {
      "auxiliary_loss_clip": 0.01194927,
      "auxiliary_loss_mlp": 0.01029379,
      "balance_loss_clip": 1.05544078,
      "balance_loss_mlp": 1.02015245,
      "epoch": 0.38405579270125656,
      "flos": 17019360218880.0,
      "grad_norm": 2.3874984353977733,
      "language_loss": 0.77294612,
      "learning_rate": 2.8230926128626095e-06,
      "loss": 0.7951892,
      "num_input_tokens_seen": 68816880,
      "step": 3194,
      "time_per_iteration": 2.5978074073791504
    },
    {
      "auxiliary_loss_clip": 0.01152494,
      "auxiliary_loss_mlp": 0.01033259,
      "balance_loss_clip": 1.04543698,
      "balance_loss_mlp": 1.02449083,
      "epoch": 0.3841760355918956,
      "flos": 21835375943040.0,
      "grad_norm": 2.1738544362932095,
      "language_loss": 0.79701626,
      "learning_rate": 2.822382604374738e-06,
      "loss": 0.81887376,
      "num_input_tokens_seen": 68835805,
      "step": 3195,
      "time_per_iteration": 2.667903184890747
    },
    {
      "auxiliary_loss_clip": 0.01162084,
      "auxiliary_loss_mlp": 0.01032105,
      "balance_loss_clip": 1.05464983,
      "balance_loss_mlp": 1.02300942,
      "epoch": 0.3842962784825347,
      "flos": 25915114684800.0,
      "grad_norm": 1.9536734641406872,
      "language_loss": 0.65894604,
      "learning_rate": 2.8216724711320793e-06,
      "loss": 0.68088794,
      "num_input_tokens_seen": 68854930,
      "step": 3196,
      "time_per_iteration": 2.837489128112793
    },
    {
      "auxiliary_loss_clip": 0.01186299,
      "auxiliary_loss_mlp": 0.00763554,
      "balance_loss_clip": 1.05026329,
      "balance_loss_mlp": 1.00014305,
      "epoch": 0.38441652137317384,
      "flos": 25337492075520.0,
      "grad_norm": 1.5415302938835473,
      "language_loss": 0.79730481,
      "learning_rate": 2.820962213242361e-06,
      "loss": 0.81680334,
      "num_input_tokens_seen": 68874260,
      "step": 3197,
      "time_per_iteration": 2.627007007598877
    },
    {
      "auxiliary_loss_clip": 0.01174032,
      "auxiliary_loss_mlp": 0.010354,
      "balance_loss_clip": 1.05384576,
      "balance_loss_mlp": 1.02722788,
      "epoch": 0.3845367642638129,
      "flos": 18113486446080.0,
      "grad_norm": 2.036814230633612,
      "language_loss": 0.84131694,
      "learning_rate": 2.8202518308133264e-06,
      "loss": 0.86341119,
      "num_input_tokens_seen": 68891535,
      "step": 3198,
      "time_per_iteration": 2.625339984893799
    },
    {
      "auxiliary_loss_clip": 0.01190621,
      "auxiliary_loss_mlp": 0.01025973,
      "balance_loss_clip": 1.05199397,
      "balance_loss_mlp": 1.01719868,
      "epoch": 0.384657007154452,
      "flos": 25228395492480.0,
      "grad_norm": 1.8181414666782902,
      "language_loss": 0.7377221,
      "learning_rate": 2.8195413239527426e-06,
      "loss": 0.75988805,
      "num_input_tokens_seen": 68911275,
      "step": 3199,
      "time_per_iteration": 2.6491947174072266
    },
    {
      "auxiliary_loss_clip": 0.01171735,
      "auxiliary_loss_mlp": 0.010409,
      "balance_loss_clip": 1.05112731,
      "balance_loss_mlp": 1.03231096,
      "epoch": 0.38477725004509106,
      "flos": 19865855358720.0,
      "grad_norm": 3.3439566460348606,
      "language_loss": 0.8111093,
      "learning_rate": 2.8188306927683906e-06,
      "loss": 0.83323562,
      "num_input_tokens_seen": 68930745,
      "step": 3200,
      "time_per_iteration": 2.608884572982788
    },
    {
      "auxiliary_loss_clip": 0.01160007,
      "auxiliary_loss_mlp": 0.01025849,
      "balance_loss_clip": 1.04905725,
      "balance_loss_mlp": 1.01721263,
      "epoch": 0.38489749293573017,
      "flos": 18259391491200.0,
      "grad_norm": 2.4558906170138535,
      "language_loss": 0.75131214,
      "learning_rate": 2.818119937368074e-06,
      "loss": 0.77317071,
      "num_input_tokens_seen": 68949380,
      "step": 3201,
      "time_per_iteration": 2.56602144241333
    },
    {
      "auxiliary_loss_clip": 0.01180389,
      "auxiliary_loss_mlp": 0.01031255,
      "balance_loss_clip": 1.05008698,
      "balance_loss_mlp": 1.0221771,
      "epoch": 0.3850177358263693,
      "flos": 24389163152640.0,
      "grad_norm": 1.9190787408377363,
      "language_loss": 0.6554625,
      "learning_rate": 2.817409057859613e-06,
      "loss": 0.67757899,
      "num_input_tokens_seen": 68968370,
      "step": 3202,
      "time_per_iteration": 2.601261854171753
    },
    {
      "auxiliary_loss_clip": 0.01124841,
      "auxiliary_loss_mlp": 0.01033084,
      "balance_loss_clip": 1.04562283,
      "balance_loss_mlp": 1.02473378,
      "epoch": 0.38513797871700833,
      "flos": 17671533505920.0,
      "grad_norm": 1.8361040505730597,
      "language_loss": 0.79127181,
      "learning_rate": 2.8166980543508482e-06,
      "loss": 0.81285107,
      "num_input_tokens_seen": 68984260,
      "step": 3203,
      "time_per_iteration": 2.673112154006958
    },
    {
      "auxiliary_loss_clip": 0.01191134,
      "auxiliary_loss_mlp": 0.01029706,
      "balance_loss_clip": 1.05366135,
      "balance_loss_mlp": 1.02081895,
      "epoch": 0.38525822160764744,
      "flos": 25739583897600.0,
      "grad_norm": 1.8184233365351232,
      "language_loss": 0.79956901,
      "learning_rate": 2.815986926949638e-06,
      "loss": 0.8217774,
      "num_input_tokens_seen": 69002760,
      "step": 3204,
      "time_per_iteration": 2.653207778930664
    },
    {
      "auxiliary_loss_clip": 0.01174287,
      "auxiliary_loss_mlp": 0.01037926,
      "balance_loss_clip": 1.05187476,
      "balance_loss_mlp": 1.02950978,
      "epoch": 0.38537846449828655,
      "flos": 20193647898240.0,
      "grad_norm": 2.837519028569492,
      "language_loss": 0.8036052,
      "learning_rate": 2.8152756757638597e-06,
      "loss": 0.82572734,
      "num_input_tokens_seen": 69021260,
      "step": 3205,
      "time_per_iteration": 2.607947587966919
    },
    {
      "auxiliary_loss_clip": 0.01175018,
      "auxiliary_loss_mlp": 0.01031697,
      "balance_loss_clip": 1.05268836,
      "balance_loss_mlp": 1.02291107,
      "epoch": 0.3854987073889256,
      "flos": 23039352938880.0,
      "grad_norm": 1.9011011931927924,
      "language_loss": 0.84710741,
      "learning_rate": 2.8145643009014093e-06,
      "loss": 0.8691746,
      "num_input_tokens_seen": 69039755,
      "step": 3206,
      "time_per_iteration": 2.627964735031128
    },
    {
      "auxiliary_loss_clip": 0.01176804,
      "auxiliary_loss_mlp": 0.0102975,
      "balance_loss_clip": 1.05230308,
      "balance_loss_mlp": 1.02123833,
      "epoch": 0.3856189502795647,
      "flos": 20190631155840.0,
      "grad_norm": 1.8552769548485435,
      "language_loss": 0.7928136,
      "learning_rate": 2.813852802470202e-06,
      "loss": 0.81487912,
      "num_input_tokens_seen": 69057650,
      "step": 3207,
      "time_per_iteration": 2.666914224624634
    },
    {
      "auxiliary_loss_clip": 0.01154252,
      "auxiliary_loss_mlp": 0.01030572,
      "balance_loss_clip": 1.04814088,
      "balance_loss_mlp": 1.02173281,
      "epoch": 0.38573919317020383,
      "flos": 25702631781120.0,
      "grad_norm": 1.7905408898997752,
      "language_loss": 0.72663307,
      "learning_rate": 2.8131411805781717e-06,
      "loss": 0.74848127,
      "num_input_tokens_seen": 69077775,
      "step": 3208,
      "time_per_iteration": 2.6816163063049316
    },
    {
      "auxiliary_loss_clip": 0.01163344,
      "auxiliary_loss_mlp": 0.01032765,
      "balance_loss_clip": 1.05296612,
      "balance_loss_mlp": 1.02317464,
      "epoch": 0.3858594360608429,
      "flos": 29821405628160.0,
      "grad_norm": 2.6698321347285914,
      "language_loss": 0.64229977,
      "learning_rate": 2.8124294353332707e-06,
      "loss": 0.66426086,
      "num_input_tokens_seen": 69096450,
      "step": 3209,
      "time_per_iteration": 3.703415632247925
    },
    {
      "auxiliary_loss_clip": 0.01151211,
      "auxiliary_loss_mlp": 0.01037306,
      "balance_loss_clip": 1.04912984,
      "balance_loss_mlp": 1.02851987,
      "epoch": 0.385979678951482,
      "flos": 24790428961920.0,
      "grad_norm": 1.5932105167524586,
      "language_loss": 0.77227539,
      "learning_rate": 2.8117175668434713e-06,
      "loss": 0.79416054,
      "num_input_tokens_seen": 69116110,
      "step": 3210,
      "time_per_iteration": 2.8631091117858887
    },
    {
      "auxiliary_loss_clip": 0.01190213,
      "auxiliary_loss_mlp": 0.01030048,
      "balance_loss_clip": 1.05278826,
      "balance_loss_mlp": 1.02098167,
      "epoch": 0.3860999218421211,
      "flos": 21287881866240.0,
      "grad_norm": 2.3430755547299005,
      "language_loss": 0.7086159,
      "learning_rate": 2.811005575216762e-06,
      "loss": 0.73081851,
      "num_input_tokens_seen": 69134825,
      "step": 3211,
      "time_per_iteration": 3.532532215118408
    },
    {
      "auxiliary_loss_clip": 0.01140513,
      "auxiliary_loss_mlp": 0.01034532,
      "balance_loss_clip": 1.04660046,
      "balance_loss_mlp": 1.02572227,
      "epoch": 0.38622016473276016,
      "flos": 24536720223360.0,
      "grad_norm": 1.6611492537084014,
      "language_loss": 0.79139346,
      "learning_rate": 2.8102934605611513e-06,
      "loss": 0.81314385,
      "num_input_tokens_seen": 69156460,
      "step": 3212,
      "time_per_iteration": 2.711057186126709
    },
    {
      "auxiliary_loss_clip": 0.01165775,
      "auxiliary_loss_mlp": 0.01034671,
      "balance_loss_clip": 1.05136526,
      "balance_loss_mlp": 1.02616,
      "epoch": 0.3863404076233993,
      "flos": 20558212986240.0,
      "grad_norm": 2.1881246343245766,
      "language_loss": 0.67711812,
      "learning_rate": 2.8095812229846665e-06,
      "loss": 0.69912255,
      "num_input_tokens_seen": 69176420,
      "step": 3213,
      "time_per_iteration": 2.6505558490753174
    },
    {
      "auxiliary_loss_clip": 0.01160164,
      "auxiliary_loss_mlp": 0.01036176,
      "balance_loss_clip": 1.04847789,
      "balance_loss_mlp": 1.02708614,
      "epoch": 0.3864606505140384,
      "flos": 22346277039360.0,
      "grad_norm": 2.204362415290507,
      "language_loss": 0.69429123,
      "learning_rate": 2.808868862595355e-06,
      "loss": 0.71625471,
      "num_input_tokens_seen": 69196665,
      "step": 3214,
      "time_per_iteration": 2.7331881523132324
    },
    {
      "auxiliary_loss_clip": 0.0117706,
      "auxiliary_loss_mlp": 0.01025158,
      "balance_loss_clip": 1.05110514,
      "balance_loss_mlp": 1.01643205,
      "epoch": 0.38658089340467744,
      "flos": 25703601448320.0,
      "grad_norm": 1.851272183506496,
      "language_loss": 0.79854685,
      "learning_rate": 2.8081563795012795e-06,
      "loss": 0.82056904,
      "num_input_tokens_seen": 69216290,
      "step": 3215,
      "time_per_iteration": 3.599665641784668
    },
    {
      "auxiliary_loss_clip": 0.01168432,
      "auxiliary_loss_mlp": 0.01033372,
      "balance_loss_clip": 1.04951787,
      "balance_loss_mlp": 1.02468705,
      "epoch": 0.38670113629531655,
      "flos": 33802534558080.0,
      "grad_norm": 1.6875968926911713,
      "language_loss": 0.73623049,
      "learning_rate": 2.807443773810524e-06,
      "loss": 0.75824857,
      "num_input_tokens_seen": 69237550,
      "step": 3216,
      "time_per_iteration": 3.6830861568450928
    },
    {
      "auxiliary_loss_clip": 0.0114833,
      "auxiliary_loss_mlp": 0.0103048,
      "balance_loss_clip": 1.05132365,
      "balance_loss_mlp": 1.02206373,
      "epoch": 0.3868213791859556,
      "flos": 23331522165120.0,
      "grad_norm": 1.8153258837562531,
      "language_loss": 0.89890087,
      "learning_rate": 2.80673104563119e-06,
      "loss": 0.92068905,
      "num_input_tokens_seen": 69258175,
      "step": 3217,
      "time_per_iteration": 2.787142276763916
    },
    {
      "auxiliary_loss_clip": 0.01171124,
      "auxiliary_loss_mlp": 0.010274,
      "balance_loss_clip": 1.05223298,
      "balance_loss_mlp": 1.01882315,
      "epoch": 0.3869416220765947,
      "flos": 18441530380800.0,
      "grad_norm": 2.3887277283156365,
      "language_loss": 0.78964192,
      "learning_rate": 2.8060181950713976e-06,
      "loss": 0.81162721,
      "num_input_tokens_seen": 69274965,
      "step": 3218,
      "time_per_iteration": 2.641911268234253
    },
    {
      "auxiliary_loss_clip": 0.01143022,
      "auxiliary_loss_mlp": 0.01038599,
      "balance_loss_clip": 1.04640806,
      "balance_loss_mlp": 1.02921712,
      "epoch": 0.3870618649672338,
      "flos": 15632992938240.0,
      "grad_norm": 2.6183990983269094,
      "language_loss": 0.80720776,
      "learning_rate": 2.805305222239286e-06,
      "loss": 0.82902396,
      "num_input_tokens_seen": 69292220,
      "step": 3219,
      "time_per_iteration": 2.6697747707366943
    },
    {
      "auxiliary_loss_clip": 0.01158299,
      "auxiliary_loss_mlp": 0.01030012,
      "balance_loss_clip": 1.04865909,
      "balance_loss_mlp": 1.02156568,
      "epoch": 0.3871821078578729,
      "flos": 23513804709120.0,
      "grad_norm": 1.8591818636861068,
      "language_loss": 0.73706442,
      "learning_rate": 2.8045921272430118e-06,
      "loss": 0.75894749,
      "num_input_tokens_seen": 69311900,
      "step": 3220,
      "time_per_iteration": 2.6817467212677
    },
    {
      "auxiliary_loss_clip": 0.01183914,
      "auxiliary_loss_mlp": 0.01031679,
      "balance_loss_clip": 1.0530026,
      "balance_loss_mlp": 1.02289963,
      "epoch": 0.387302350748512,
      "flos": 17778259791360.0,
      "grad_norm": 2.847483525038182,
      "language_loss": 0.76721942,
      "learning_rate": 2.803878910190753e-06,
      "loss": 0.78937536,
      "num_input_tokens_seen": 69328820,
      "step": 3221,
      "time_per_iteration": 2.6325597763061523
    },
    {
      "auxiliary_loss_clip": 0.0118026,
      "auxiliary_loss_mlp": 0.01028916,
      "balance_loss_clip": 1.05212653,
      "balance_loss_mlp": 1.02029085,
      "epoch": 0.3874225936391511,
      "flos": 11503409097600.0,
      "grad_norm": 5.1580432727542656,
      "language_loss": 0.82065547,
      "learning_rate": 2.8031655711907017e-06,
      "loss": 0.84274727,
      "num_input_tokens_seen": 69342525,
      "step": 3222,
      "time_per_iteration": 2.6161258220672607
    },
    {
      "auxiliary_loss_clip": 0.01178966,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 1.05418193,
      "balance_loss_mlp": 1.01620507,
      "epoch": 0.38754283652979016,
      "flos": 21945154884480.0,
      "grad_norm": 2.9750624090875526,
      "language_loss": 0.81059313,
      "learning_rate": 2.8024521103510723e-06,
      "loss": 0.83263755,
      "num_input_tokens_seen": 69359295,
      "step": 3223,
      "time_per_iteration": 2.6720969676971436
    },
    {
      "auxiliary_loss_clip": 0.01174236,
      "auxiliary_loss_mlp": 0.01031433,
      "balance_loss_clip": 1.04950285,
      "balance_loss_mlp": 1.02296329,
      "epoch": 0.38766307942042927,
      "flos": 21175984022400.0,
      "grad_norm": 1.8452798858271167,
      "language_loss": 0.75365776,
      "learning_rate": 2.8017385277800952e-06,
      "loss": 0.77571446,
      "num_input_tokens_seen": 69377650,
      "step": 3224,
      "time_per_iteration": 2.7115986347198486
    },
    {
      "auxiliary_loss_clip": 0.01149491,
      "auxiliary_loss_mlp": 0.01028611,
      "balance_loss_clip": 1.04794848,
      "balance_loss_mlp": 1.01974225,
      "epoch": 0.3877833223110684,
      "flos": 27417294391680.0,
      "grad_norm": 2.1161339192761917,
      "language_loss": 0.75310135,
      "learning_rate": 2.8010248235860213e-06,
      "loss": 0.77488238,
      "num_input_tokens_seen": 69397765,
      "step": 3225,
      "time_per_iteration": 2.7463183403015137
    },
    {
      "auxiliary_loss_clip": 0.01073668,
      "auxiliary_loss_mlp": 0.00755739,
      "balance_loss_clip": 1.01980603,
      "balance_loss_mlp": 1.00042903,
      "epoch": 0.38790356520170743,
      "flos": 64500019879680.0,
      "grad_norm": 0.829034908506229,
      "language_loss": 0.62792313,
      "learning_rate": 2.8003109978771192e-06,
      "loss": 0.64621723,
      "num_input_tokens_seen": 69458930,
      "step": 3226,
      "time_per_iteration": 3.360851287841797
    },
    {
      "auxiliary_loss_clip": 0.01134923,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.04260302,
      "balance_loss_mlp": 1.02097464,
      "epoch": 0.38802380809234654,
      "flos": 22345415112960.0,
      "grad_norm": 1.9455688411867902,
      "language_loss": 0.78880906,
      "learning_rate": 2.799597050761674e-06,
      "loss": 0.81045449,
      "num_input_tokens_seen": 69475135,
      "step": 3227,
      "time_per_iteration": 2.6835217475891113
    },
    {
      "auxiliary_loss_clip": 0.0119262,
      "auxiliary_loss_mlp": 0.01025488,
      "balance_loss_clip": 1.05252886,
      "balance_loss_mlp": 1.01692283,
      "epoch": 0.38814405098298566,
      "flos": 25261361199360.0,
      "grad_norm": 1.858894396097029,
      "language_loss": 0.7932533,
      "learning_rate": 2.7988829823479924e-06,
      "loss": 0.8154344,
      "num_input_tokens_seen": 69493525,
      "step": 3228,
      "time_per_iteration": 2.646496534347534
    },
    {
      "auxiliary_loss_clip": 0.0115483,
      "auxiliary_loss_mlp": 0.01031148,
      "balance_loss_clip": 1.04914474,
      "balance_loss_mlp": 1.02236819,
      "epoch": 0.3882642938736247,
      "flos": 18841180078080.0,
      "grad_norm": 1.7831991777319018,
      "language_loss": 0.64278126,
      "learning_rate": 2.7981687927443976e-06,
      "loss": 0.66464102,
      "num_input_tokens_seen": 69510325,
      "step": 3229,
      "time_per_iteration": 2.7169153690338135
    },
    {
      "auxiliary_loss_clip": 0.01173908,
      "auxiliary_loss_mlp": 0.01028938,
      "balance_loss_clip": 1.04854071,
      "balance_loss_mlp": 1.01992607,
      "epoch": 0.3883845367642638,
      "flos": 21652806090240.0,
      "grad_norm": 1.8823700329160886,
      "language_loss": 0.85529864,
      "learning_rate": 2.797454482059231e-06,
      "loss": 0.87732708,
      "num_input_tokens_seen": 69530480,
      "step": 3230,
      "time_per_iteration": 2.6146080493927
    },
    {
      "auxiliary_loss_clip": 0.01192728,
      "auxiliary_loss_mlp": 0.01026236,
      "balance_loss_clip": 1.05366254,
      "balance_loss_mlp": 1.01776028,
      "epoch": 0.3885047796549029,
      "flos": 20557530627840.0,
      "grad_norm": 1.5856909793639167,
      "language_loss": 0.84554613,
      "learning_rate": 2.7967400504008537e-06,
      "loss": 0.86773574,
      "num_input_tokens_seen": 69549780,
      "step": 3231,
      "time_per_iteration": 2.6036620140075684
    },
    {
      "auxiliary_loss_clip": 0.01041706,
      "auxiliary_loss_mlp": 0.01002217,
      "balance_loss_clip": 1.01618648,
      "balance_loss_mlp": 1.0004288,
      "epoch": 0.388625022545542,
      "flos": 64325491695360.0,
      "grad_norm": 0.7905515318842432,
      "language_loss": 0.57397246,
      "learning_rate": 2.7960254978776456e-06,
      "loss": 0.59441161,
      "num_input_tokens_seen": 69611870,
      "step": 3232,
      "time_per_iteration": 3.3132598400115967
    },
    {
      "auxiliary_loss_clip": 0.01190453,
      "auxiliary_loss_mlp": 0.01031469,
      "balance_loss_clip": 1.05273461,
      "balance_loss_mlp": 1.02288628,
      "epoch": 0.3887452654361811,
      "flos": 18113881495680.0,
      "grad_norm": 2.049306693897059,
      "language_loss": 0.81853598,
      "learning_rate": 2.7953108245980006e-06,
      "loss": 0.84075522,
      "num_input_tokens_seen": 69630385,
      "step": 3233,
      "time_per_iteration": 2.5742053985595703
    },
    {
      "auxiliary_loss_clip": 0.0115917,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.05227613,
      "balance_loss_mlp": 1.02166522,
      "epoch": 0.38886550832682015,
      "flos": 24975261371520.0,
      "grad_norm": 1.6923860301561402,
      "language_loss": 0.736404,
      "learning_rate": 2.7945960306703365e-06,
      "loss": 0.75829196,
      "num_input_tokens_seen": 69653370,
      "step": 3234,
      "time_per_iteration": 2.716172933578491
    },
    {
      "auxiliary_loss_clip": 0.01180625,
      "auxiliary_loss_mlp": 0.01028309,
      "balance_loss_clip": 1.05233169,
      "balance_loss_mlp": 1.01925457,
      "epoch": 0.38898575121745926,
      "flos": 27199496275200.0,
      "grad_norm": 1.6830132174096664,
      "language_loss": 0.6600008,
      "learning_rate": 2.7938811162030865e-06,
      "loss": 0.68209016,
      "num_input_tokens_seen": 69673635,
      "step": 3235,
      "time_per_iteration": 3.6284356117248535
    },
    {
      "auxiliary_loss_clip": 0.01174214,
      "auxiliary_loss_mlp": 0.01032502,
      "balance_loss_clip": 1.051651,
      "balance_loss_mlp": 1.02466428,
      "epoch": 0.3891059941080984,
      "flos": 28763728727040.0,
      "grad_norm": 1.6225851618359772,
      "language_loss": 0.82521915,
      "learning_rate": 2.793166081304702e-06,
      "loss": 0.84728634,
      "num_input_tokens_seen": 69694130,
      "step": 3236,
      "time_per_iteration": 2.6599864959716797
    },
    {
      "auxiliary_loss_clip": 0.01152578,
      "auxiliary_loss_mlp": 0.01028028,
      "balance_loss_clip": 1.04751873,
      "balance_loss_mlp": 1.01933181,
      "epoch": 0.38922623699873743,
      "flos": 22893447893760.0,
      "grad_norm": 1.8615951226464944,
      "language_loss": 0.82299507,
      "learning_rate": 2.7924509260836543e-06,
      "loss": 0.84480107,
      "num_input_tokens_seen": 69713255,
      "step": 3237,
      "time_per_iteration": 3.675790309906006
    },
    {
      "auxiliary_loss_clip": 0.01144001,
      "auxiliary_loss_mlp": 0.01028556,
      "balance_loss_clip": 1.04660082,
      "balance_loss_mlp": 1.01960897,
      "epoch": 0.38934647988937654,
      "flos": 19792418002560.0,
      "grad_norm": 1.5376539140713679,
      "language_loss": 0.68406224,
      "learning_rate": 2.791735650648431e-06,
      "loss": 0.7057879,
      "num_input_tokens_seen": 69732375,
      "step": 3238,
      "time_per_iteration": 2.6948530673980713
    },
    {
      "auxiliary_loss_clip": 0.01158371,
      "auxiliary_loss_mlp": 0.01026559,
      "balance_loss_clip": 1.04916346,
      "balance_loss_mlp": 1.0175705,
      "epoch": 0.38946672278001565,
      "flos": 19202081978880.0,
      "grad_norm": 2.549064714336416,
      "language_loss": 0.74365664,
      "learning_rate": 2.791020255107538e-06,
      "loss": 0.76550591,
      "num_input_tokens_seen": 69749745,
      "step": 3239,
      "time_per_iteration": 2.6820216178894043
    },
    {
      "auxiliary_loss_clip": 0.01140669,
      "auxiliary_loss_mlp": 0.01031666,
      "balance_loss_clip": 1.04441357,
      "balance_loss_mlp": 1.02315402,
      "epoch": 0.3895869656706547,
      "flos": 24936477661440.0,
      "grad_norm": 1.6124931371829117,
      "language_loss": 0.80830002,
      "learning_rate": 2.7903047395695023e-06,
      "loss": 0.83002335,
      "num_input_tokens_seen": 69769645,
      "step": 3240,
      "time_per_iteration": 2.7840681076049805
    },
    {
      "auxiliary_loss_clip": 0.01175275,
      "auxiliary_loss_mlp": 0.00764291,
      "balance_loss_clip": 1.05337143,
      "balance_loss_mlp": 1.00019097,
      "epoch": 0.3897072085612938,
      "flos": 24133622820480.0,
      "grad_norm": 4.053779956028131,
      "language_loss": 0.90030605,
      "learning_rate": 2.789589104142865e-06,
      "loss": 0.9197017,
      "num_input_tokens_seen": 69787270,
      "step": 3241,
      "time_per_iteration": 3.6048662662506104
    },
    {
      "auxiliary_loss_clip": 0.01149821,
      "auxiliary_loss_mlp": 0.01027593,
      "balance_loss_clip": 1.0494653,
      "balance_loss_mlp": 1.01872361,
      "epoch": 0.3898274514519329,
      "flos": 17166342672000.0,
      "grad_norm": 2.0240883621220638,
      "language_loss": 0.76373833,
      "learning_rate": 2.7888733489361895e-06,
      "loss": 0.78551245,
      "num_input_tokens_seen": 69805685,
      "step": 3242,
      "time_per_iteration": 3.6397271156311035
    },
    {
      "auxiliary_loss_clip": 0.01090919,
      "auxiliary_loss_mlp": 0.01002595,
      "balance_loss_clip": 1.02002406,
      "balance_loss_mlp": 1.00073576,
      "epoch": 0.389947694342572,
      "flos": 66074807952000.0,
      "grad_norm": 0.7310796590533779,
      "language_loss": 0.58682263,
      "learning_rate": 2.788157474058054e-06,
      "loss": 0.60775781,
      "num_input_tokens_seen": 69867960,
      "step": 3243,
      "time_per_iteration": 3.3321402072906494
    },
    {
      "auxiliary_loss_clip": 0.01186734,
      "auxiliary_loss_mlp": 0.01029253,
      "balance_loss_clip": 1.05219829,
      "balance_loss_mlp": 1.02083051,
      "epoch": 0.3900679372332111,
      "flos": 25740912700800.0,
      "grad_norm": 1.972669062094332,
      "language_loss": 0.70204961,
      "learning_rate": 2.7874414796170555e-06,
      "loss": 0.72420943,
      "num_input_tokens_seen": 69889450,
      "step": 3244,
      "time_per_iteration": 2.633253812789917
    },
    {
      "auxiliary_loss_clip": 0.01169299,
      "auxiliary_loss_mlp": 0.0103546,
      "balance_loss_clip": 1.04990244,
      "balance_loss_mlp": 1.02608371,
      "epoch": 0.3901881801238502,
      "flos": 11801611808640.0,
      "grad_norm": 2.238129937106705,
      "language_loss": 0.84014487,
      "learning_rate": 2.7867253657218113e-06,
      "loss": 0.86219251,
      "num_input_tokens_seen": 69903340,
      "step": 3245,
      "time_per_iteration": 2.641946792602539
    },
    {
      "auxiliary_loss_clip": 0.01158958,
      "auxiliary_loss_mlp": 0.00764047,
      "balance_loss_clip": 1.04768801,
      "balance_loss_mlp": 1.00022542,
      "epoch": 0.39030842301448926,
      "flos": 27308951994240.0,
      "grad_norm": 1.6484617553314735,
      "language_loss": 0.73365331,
      "learning_rate": 2.7860091324809544e-06,
      "loss": 0.75288332,
      "num_input_tokens_seen": 69924400,
      "step": 3246,
      "time_per_iteration": 2.7086246013641357
    },
    {
      "auxiliary_loss_clip": 0.01173083,
      "auxiliary_loss_mlp": 0.01028303,
      "balance_loss_clip": 1.05279982,
      "balance_loss_mlp": 1.02012479,
      "epoch": 0.39042866590512837,
      "flos": 27163334257920.0,
      "grad_norm": 1.654601617606051,
      "language_loss": 0.81443775,
      "learning_rate": 2.7852927800031377e-06,
      "loss": 0.83645165,
      "num_input_tokens_seen": 69944565,
      "step": 3247,
      "time_per_iteration": 2.7154603004455566
    },
    {
      "auxiliary_loss_clip": 0.01162844,
      "auxiliary_loss_mlp": 0.01032218,
      "balance_loss_clip": 1.04947972,
      "balance_loss_mlp": 1.02382588,
      "epoch": 0.3905489087957674,
      "flos": 29716115886720.0,
      "grad_norm": 1.7816544813741406,
      "language_loss": 0.82958043,
      "learning_rate": 2.7845763083970298e-06,
      "loss": 0.85153103,
      "num_input_tokens_seen": 69964965,
      "step": 3248,
      "time_per_iteration": 2.7210569381713867
    },
    {
      "auxiliary_loss_clip": 0.01166301,
      "auxiliary_loss_mlp": 0.01029715,
      "balance_loss_clip": 1.04832757,
      "balance_loss_mlp": 1.02089381,
      "epoch": 0.39066915168640653,
      "flos": 24498618871680.0,
      "grad_norm": 2.291479479280437,
      "language_loss": 0.82164752,
      "learning_rate": 2.7838597177713205e-06,
      "loss": 0.84360766,
      "num_input_tokens_seen": 69986055,
      "step": 3249,
      "time_per_iteration": 2.697658061981201
    },
    {
      "auxiliary_loss_clip": 0.01113451,
      "auxiliary_loss_mlp": 0.01038111,
      "balance_loss_clip": 1.04958999,
      "balance_loss_mlp": 1.02936745,
      "epoch": 0.39078939457704565,
      "flos": 20558572122240.0,
      "grad_norm": 2.597649224064127,
      "language_loss": 0.73806036,
      "learning_rate": 2.7831430082347143e-06,
      "loss": 0.75957596,
      "num_input_tokens_seen": 70005260,
      "step": 3250,
      "time_per_iteration": 2.8245553970336914
    },
    {
      "auxiliary_loss_clip": 0.01176939,
      "auxiliary_loss_mlp": 0.00762897,
      "balance_loss_clip": 1.05310321,
      "balance_loss_mlp": 1.0002079,
      "epoch": 0.3909096374676847,
      "flos": 22783417557120.0,
      "grad_norm": 1.9147297985926481,
      "language_loss": 0.82648611,
      "learning_rate": 2.7824261798959373e-06,
      "loss": 0.8458845,
      "num_input_tokens_seen": 70023440,
      "step": 3251,
      "time_per_iteration": 2.6061642169952393
    },
    {
      "auxiliary_loss_clip": 0.0115989,
      "auxiliary_loss_mlp": 0.01030695,
      "balance_loss_clip": 1.04606843,
      "balance_loss_mlp": 1.02232623,
      "epoch": 0.3910298803583238,
      "flos": 23003119094400.0,
      "grad_norm": 1.8895253641803071,
      "language_loss": 0.7956841,
      "learning_rate": 2.78170923286373e-06,
      "loss": 0.81758988,
      "num_input_tokens_seen": 70043040,
      "step": 3252,
      "time_per_iteration": 2.7194063663482666
    },
    {
      "auxiliary_loss_clip": 0.01101592,
      "auxiliary_loss_mlp": 0.0102609,
      "balance_loss_clip": 1.04741931,
      "balance_loss_mlp": 1.01703572,
      "epoch": 0.3911501232489629,
      "flos": 24316264500480.0,
      "grad_norm": 2.314289940810285,
      "language_loss": 0.84237635,
      "learning_rate": 2.780992167246854e-06,
      "loss": 0.86365318,
      "num_input_tokens_seen": 70060565,
      "step": 3253,
      "time_per_iteration": 2.872896671295166
    },
    {
      "auxiliary_loss_clip": 0.01070876,
      "auxiliary_loss_mlp": 0.01003425,
      "balance_loss_clip": 1.01755285,
      "balance_loss_mlp": 1.00161254,
      "epoch": 0.391270366139602,
      "flos": 60869054684160.0,
      "grad_norm": 0.9736589010466935,
      "language_loss": 0.72114909,
      "learning_rate": 2.7802749831540883e-06,
      "loss": 0.7418921,
      "num_input_tokens_seen": 70119465,
      "step": 3254,
      "time_per_iteration": 3.383840799331665
    },
    {
      "auxiliary_loss_clip": 0.01136488,
      "auxiliary_loss_mlp": 0.01027103,
      "balance_loss_clip": 1.04746914,
      "balance_loss_mlp": 1.01957512,
      "epoch": 0.3913906090302411,
      "flos": 21543494025600.0,
      "grad_norm": 1.8953867975198238,
      "language_loss": 0.81710649,
      "learning_rate": 2.7795576806942268e-06,
      "loss": 0.83874243,
      "num_input_tokens_seen": 70138270,
      "step": 3255,
      "time_per_iteration": 2.74540114402771
    },
    {
      "auxiliary_loss_clip": 0.01079663,
      "auxiliary_loss_mlp": 0.0100431,
      "balance_loss_clip": 1.0355792,
      "balance_loss_mlp": 1.00255775,
      "epoch": 0.3915108519208802,
      "flos": 49839953702400.0,
      "grad_norm": 0.7567055908940665,
      "language_loss": 0.5484755,
      "learning_rate": 2.778840259976085e-06,
      "loss": 0.5693152,
      "num_input_tokens_seen": 70193500,
      "step": 3256,
      "time_per_iteration": 3.2198386192321777
    },
    {
      "auxiliary_loss_clip": 0.0117687,
      "auxiliary_loss_mlp": 0.01032945,
      "balance_loss_clip": 1.05174327,
      "balance_loss_mlp": 1.02410555,
      "epoch": 0.39163109481151925,
      "flos": 16506447960960.0,
      "grad_norm": 2.060167716233353,
      "language_loss": 0.76723361,
      "learning_rate": 2.778122721108495e-06,
      "loss": 0.78933173,
      "num_input_tokens_seen": 70211730,
      "step": 3257,
      "time_per_iteration": 2.6169803142547607
    },
    {
      "auxiliary_loss_clip": 0.01174736,
      "auxiliary_loss_mlp": 0.01036034,
      "balance_loss_clip": 1.05420423,
      "balance_loss_mlp": 1.02755845,
      "epoch": 0.39175133770215836,
      "flos": 26067484177920.0,
      "grad_norm": 1.9165143063507595,
      "language_loss": 0.88172525,
      "learning_rate": 2.7774050642003076e-06,
      "loss": 0.90383297,
      "num_input_tokens_seen": 70232540,
      "step": 3258,
      "time_per_iteration": 2.6419460773468018
    },
    {
      "auxiliary_loss_clip": 0.01195806,
      "auxiliary_loss_mlp": 0.01032292,
      "balance_loss_clip": 1.05708456,
      "balance_loss_mlp": 1.02342868,
      "epoch": 0.3918715805927975,
      "flos": 21872076664320.0,
      "grad_norm": 2.1078726511186603,
      "language_loss": 0.93146777,
      "learning_rate": 2.7766872893603896e-06,
      "loss": 0.95374876,
      "num_input_tokens_seen": 70252515,
      "step": 3259,
      "time_per_iteration": 2.6197965145111084
    },
    {
      "auxiliary_loss_clip": 0.01179842,
      "auxiliary_loss_mlp": 0.01031802,
      "balance_loss_clip": 1.05412698,
      "balance_loss_mlp": 1.02388048,
      "epoch": 0.39199182348343653,
      "flos": 20376181837440.0,
      "grad_norm": 1.722732431557317,
      "language_loss": 0.73325956,
      "learning_rate": 2.7759693966976275e-06,
      "loss": 0.75537598,
      "num_input_tokens_seen": 70271020,
      "step": 3260,
      "time_per_iteration": 2.6392407417297363
    },
    {
      "auxiliary_loss_clip": 0.01145359,
      "auxiliary_loss_mlp": 0.01034518,
      "balance_loss_clip": 1.04826474,
      "balance_loss_mlp": 1.02498722,
      "epoch": 0.39211206637407564,
      "flos": 21683545153920.0,
      "grad_norm": 2.357233885421734,
      "language_loss": 0.85097909,
      "learning_rate": 2.7752513863209242e-06,
      "loss": 0.87277782,
      "num_input_tokens_seen": 70289600,
      "step": 3261,
      "time_per_iteration": 2.718134880065918
    },
    {
      "auxiliary_loss_clip": 0.01157029,
      "auxiliary_loss_mlp": 0.00762821,
      "balance_loss_clip": 1.05265713,
      "balance_loss_mlp": 1.00026906,
      "epoch": 0.39223230926471475,
      "flos": 21066276908160.0,
      "grad_norm": 1.7100269778413049,
      "language_loss": 0.84515667,
      "learning_rate": 2.774533258339203e-06,
      "loss": 0.86435521,
      "num_input_tokens_seen": 70307060,
      "step": 3262,
      "time_per_iteration": 3.676752805709839
    },
    {
      "auxiliary_loss_clip": 0.01131442,
      "auxiliary_loss_mlp": 0.01028285,
      "balance_loss_clip": 1.04235506,
      "balance_loss_mlp": 1.02003574,
      "epoch": 0.3923525521553538,
      "flos": 17603016312960.0,
      "grad_norm": 2.048500285926174,
      "language_loss": 0.80160481,
      "learning_rate": 2.7738150128614014e-06,
      "loss": 0.82320207,
      "num_input_tokens_seen": 70324465,
      "step": 3263,
      "time_per_iteration": 3.6646718978881836
    },
    {
      "auxiliary_loss_clip": 0.01134961,
      "auxiliary_loss_mlp": 0.01028997,
      "balance_loss_clip": 1.04738784,
      "balance_loss_mlp": 1.02065217,
      "epoch": 0.3924727950459929,
      "flos": 20558284813440.0,
      "grad_norm": 1.881881923438459,
      "language_loss": 0.89740968,
      "learning_rate": 2.7730966499964777e-06,
      "loss": 0.91904926,
      "num_input_tokens_seen": 70341415,
      "step": 3264,
      "time_per_iteration": 2.6850924491882324
    },
    {
      "auxiliary_loss_clip": 0.01189869,
      "auxiliary_loss_mlp": 0.01025672,
      "balance_loss_clip": 1.05162907,
      "balance_loss_mlp": 1.01688612,
      "epoch": 0.39259303793663197,
      "flos": 16216110328320.0,
      "grad_norm": 2.58498421946209,
      "language_loss": 0.81064636,
      "learning_rate": 2.772378169853408e-06,
      "loss": 0.83280176,
      "num_input_tokens_seen": 70358985,
      "step": 3265,
      "time_per_iteration": 2.534372568130493
    },
    {
      "auxiliary_loss_clip": 0.01147779,
      "auxiliary_loss_mlp": 0.01028639,
      "balance_loss_clip": 1.05364573,
      "balance_loss_mlp": 1.02011514,
      "epoch": 0.3927132808272711,
      "flos": 16797001075200.0,
      "grad_norm": 1.7078610096899944,
      "language_loss": 0.74596488,
      "learning_rate": 2.771659572541183e-06,
      "loss": 0.76772904,
      "num_input_tokens_seen": 70376915,
      "step": 3266,
      "time_per_iteration": 2.678567409515381
    },
    {
      "auxiliary_loss_clip": 0.01177894,
      "auxiliary_loss_mlp": 0.01028644,
      "balance_loss_clip": 1.05329347,
      "balance_loss_mlp": 1.02065158,
      "epoch": 0.3928335237179102,
      "flos": 20267228908800.0,
      "grad_norm": 2.135971499697529,
      "language_loss": 0.86896163,
      "learning_rate": 2.7709408581688143e-06,
      "loss": 0.89102697,
      "num_input_tokens_seen": 70396900,
      "step": 3267,
      "time_per_iteration": 3.5650525093078613
    },
    {
      "auxiliary_loss_clip": 0.01153876,
      "auxiliary_loss_mlp": 0.01031102,
      "balance_loss_clip": 1.05228758,
      "balance_loss_mlp": 1.02300739,
      "epoch": 0.39295376660854925,
      "flos": 24973250209920.0,
      "grad_norm": 1.665627915089905,
      "language_loss": 0.880494,
      "learning_rate": 2.7702220268453307e-06,
      "loss": 0.90234375,
      "num_input_tokens_seen": 70417260,
      "step": 3268,
      "time_per_iteration": 3.6638028621673584
    },
    {
      "auxiliary_loss_clip": 0.01162943,
      "auxiliary_loss_mlp": 0.01027821,
      "balance_loss_clip": 1.05123186,
      "balance_loss_mlp": 1.01940477,
      "epoch": 0.39307400949918836,
      "flos": 18697788984960.0,
      "grad_norm": 1.8579330019904117,
      "language_loss": 0.84816486,
      "learning_rate": 2.7695030786797785e-06,
      "loss": 0.87007248,
      "num_input_tokens_seen": 70433155,
      "step": 3269,
      "time_per_iteration": 2.659712314605713
    },
    {
      "auxiliary_loss_clip": 0.01127149,
      "auxiliary_loss_mlp": 0.01032498,
      "balance_loss_clip": 1.04657769,
      "balance_loss_mlp": 1.02387953,
      "epoch": 0.39319425238982747,
      "flos": 22415476590720.0,
      "grad_norm": 2.1651085775961056,
      "language_loss": 0.7483176,
      "learning_rate": 2.7687840137812206e-06,
      "loss": 0.76991409,
      "num_input_tokens_seen": 70451240,
      "step": 3270,
      "time_per_iteration": 2.7541627883911133
    },
    {
      "auxiliary_loss_clip": 0.01071234,
      "auxiliary_loss_mlp": 0.01003551,
      "balance_loss_clip": 1.01649165,
      "balance_loss_mlp": 1.00188184,
      "epoch": 0.3933144952804665,
      "flos": 66192954762240.0,
      "grad_norm": 0.7997843179156587,
      "language_loss": 0.62061441,
      "learning_rate": 2.7680648322587395e-06,
      "loss": 0.64136225,
      "num_input_tokens_seen": 70516115,
      "step": 3271,
      "time_per_iteration": 3.2671632766723633
    },
    {
      "auxiliary_loss_clip": 0.01189042,
      "auxiliary_loss_mlp": 0.010227,
      "balance_loss_clip": 1.0529778,
      "balance_loss_mlp": 1.01400375,
      "epoch": 0.39343473817110564,
      "flos": 15487159720320.0,
      "grad_norm": 1.84275640504053,
      "language_loss": 0.8092705,
      "learning_rate": 2.7673455342214334e-06,
      "loss": 0.83138788,
      "num_input_tokens_seen": 70533105,
      "step": 3272,
      "time_per_iteration": 2.6199252605438232
    },
    {
      "auxiliary_loss_clip": 0.01175373,
      "auxiliary_loss_mlp": 0.01026792,
      "balance_loss_clip": 1.05246079,
      "balance_loss_mlp": 1.01823878,
      "epoch": 0.39355498106174475,
      "flos": 21324905809920.0,
      "grad_norm": 1.8421667791011052,
      "language_loss": 0.7594797,
      "learning_rate": 2.7666261197784198e-06,
      "loss": 0.78150141,
      "num_input_tokens_seen": 70551920,
      "step": 3273,
      "time_per_iteration": 2.624459981918335
    },
    {
      "auxiliary_loss_clip": 0.01157154,
      "auxiliary_loss_mlp": 0.01027736,
      "balance_loss_clip": 1.0512495,
      "balance_loss_mlp": 1.0192064,
      "epoch": 0.3936752239523838,
      "flos": 13296357400320.0,
      "grad_norm": 2.0744170495955148,
      "language_loss": 0.76261717,
      "learning_rate": 2.7659065890388336e-06,
      "loss": 0.78446609,
      "num_input_tokens_seen": 70567920,
      "step": 3274,
      "time_per_iteration": 2.6840105056762695
    },
    {
      "auxiliary_loss_clip": 0.01161917,
      "auxiliary_loss_mlp": 0.01022416,
      "balance_loss_clip": 1.04861045,
      "balance_loss_mlp": 1.01446438,
      "epoch": 0.3937954668430229,
      "flos": 16800161472000.0,
      "grad_norm": 1.9589124095976442,
      "language_loss": 0.85063368,
      "learning_rate": 2.7651869421118266e-06,
      "loss": 0.87247694,
      "num_input_tokens_seen": 70584530,
      "step": 3275,
      "time_per_iteration": 2.641868829727173
    },
    {
      "auxiliary_loss_clip": 0.0117918,
      "auxiliary_loss_mlp": 0.01031072,
      "balance_loss_clip": 1.0542475,
      "balance_loss_mlp": 1.02261996,
      "epoch": 0.393915709733662,
      "flos": 21064229832960.0,
      "grad_norm": 1.6363944918259072,
      "language_loss": 0.82835627,
      "learning_rate": 2.76446717910657e-06,
      "loss": 0.85045886,
      "num_input_tokens_seen": 70605235,
      "step": 3276,
      "time_per_iteration": 2.6670329570770264
    },
    {
      "auxiliary_loss_clip": 0.01171683,
      "auxiliary_loss_mlp": 0.0102985,
      "balance_loss_clip": 1.05105853,
      "balance_loss_mlp": 1.02183914,
      "epoch": 0.3940359526243011,
      "flos": 17165265264000.0,
      "grad_norm": 2.16535933780981,
      "language_loss": 0.77144289,
      "learning_rate": 2.763747300132249e-06,
      "loss": 0.79345822,
      "num_input_tokens_seen": 70622675,
      "step": 3277,
      "time_per_iteration": 2.575369358062744
    },
    {
      "auxiliary_loss_clip": 0.01189091,
      "auxiliary_loss_mlp": 0.0103102,
      "balance_loss_clip": 1.05437779,
      "balance_loss_mlp": 1.0230335,
      "epoch": 0.3941561955149402,
      "flos": 20995856294400.0,
      "grad_norm": 1.6533014300902698,
      "language_loss": 0.87055767,
      "learning_rate": 2.7630273052980704e-06,
      "loss": 0.89275885,
      "num_input_tokens_seen": 70643265,
      "step": 3278,
      "time_per_iteration": 2.660200357437134
    },
    {
      "auxiliary_loss_clip": 0.01149011,
      "auxiliary_loss_mlp": 0.01030767,
      "balance_loss_clip": 1.04919636,
      "balance_loss_mlp": 1.02280951,
      "epoch": 0.39427643840557924,
      "flos": 18843406721280.0,
      "grad_norm": 2.1760271054542,
      "language_loss": 0.66952157,
      "learning_rate": 2.7623071947132554e-06,
      "loss": 0.69131935,
      "num_input_tokens_seen": 70660295,
      "step": 3279,
      "time_per_iteration": 2.6336703300476074
    },
    {
      "auxiliary_loss_clip": 0.01166851,
      "auxiliary_loss_mlp": 0.0103343,
      "balance_loss_clip": 1.05001855,
      "balance_loss_mlp": 1.02486467,
      "epoch": 0.39439668129621835,
      "flos": 23258659426560.0,
      "grad_norm": 2.0493208448528404,
      "language_loss": 0.7907353,
      "learning_rate": 2.7615869684870458e-06,
      "loss": 0.81273806,
      "num_input_tokens_seen": 70679605,
      "step": 3280,
      "time_per_iteration": 2.7293214797973633
    },
    {
      "auxiliary_loss_clip": 0.0117302,
      "auxiliary_loss_mlp": 0.01031923,
      "balance_loss_clip": 1.05284441,
      "balance_loss_mlp": 1.02352476,
      "epoch": 0.39451692418685746,
      "flos": 26652289507200.0,
      "grad_norm": 1.8314226317763982,
      "language_loss": 0.8490175,
      "learning_rate": 2.7608666267286986e-06,
      "loss": 0.87106693,
      "num_input_tokens_seen": 70699835,
      "step": 3281,
      "time_per_iteration": 2.721987009048462
    },
    {
      "auxiliary_loss_clip": 0.01108433,
      "auxiliary_loss_mlp": 0.01029514,
      "balance_loss_clip": 1.04144669,
      "balance_loss_mlp": 1.02019751,
      "epoch": 0.3946371670774965,
      "flos": 18258709132800.0,
      "grad_norm": 2.226008972073127,
      "language_loss": 0.8671031,
      "learning_rate": 2.760146169547489e-06,
      "loss": 0.88848257,
      "num_input_tokens_seen": 70716600,
      "step": 3282,
      "time_per_iteration": 2.871910572052002
    },
    {
      "auxiliary_loss_clip": 0.01162818,
      "auxiliary_loss_mlp": 0.01032559,
      "balance_loss_clip": 1.05334008,
      "balance_loss_mlp": 1.02439284,
      "epoch": 0.39475740996813563,
      "flos": 24206126423040.0,
      "grad_norm": 1.7900601078095661,
      "language_loss": 0.7656064,
      "learning_rate": 2.75942559705271e-06,
      "loss": 0.78756022,
      "num_input_tokens_seen": 70736335,
      "step": 3283,
      "time_per_iteration": 2.7776174545288086
    },
    {
      "auxiliary_loss_clip": 0.01173996,
      "auxiliary_loss_mlp": 0.01034297,
      "balance_loss_clip": 1.05251741,
      "balance_loss_mlp": 1.02536559,
      "epoch": 0.39487765285877474,
      "flos": 19317858491520.0,
      "grad_norm": 1.83777635418339,
      "language_loss": 0.89197129,
      "learning_rate": 2.7587049093536713e-06,
      "loss": 0.91405421,
      "num_input_tokens_seen": 70752665,
      "step": 3284,
      "time_per_iteration": 2.670316457748413
    },
    {
      "auxiliary_loss_clip": 0.01178453,
      "auxiliary_loss_mlp": 0.01032626,
      "balance_loss_clip": 1.05219042,
      "balance_loss_mlp": 1.02502656,
      "epoch": 0.3949978957494138,
      "flos": 17311744926720.0,
      "grad_norm": 1.8707022125905621,
      "language_loss": 0.80775368,
      "learning_rate": 2.757984106559701e-06,
      "loss": 0.8298645,
      "num_input_tokens_seen": 70771650,
      "step": 3285,
      "time_per_iteration": 2.6315786838531494
    },
    {
      "auxiliary_loss_clip": 0.01153391,
      "auxiliary_loss_mlp": 0.01031104,
      "balance_loss_clip": 1.05038548,
      "balance_loss_mlp": 1.02317619,
      "epoch": 0.3951181386400529,
      "flos": 36317861280000.0,
      "grad_norm": 2.188978728937141,
      "language_loss": 0.71100348,
      "learning_rate": 2.7572631887801446e-06,
      "loss": 0.73284841,
      "num_input_tokens_seen": 70793275,
      "step": 3286,
      "time_per_iteration": 2.7904465198516846
    },
    {
      "auxiliary_loss_clip": 0.01172769,
      "auxiliary_loss_mlp": 0.01029586,
      "balance_loss_clip": 1.05044866,
      "balance_loss_mlp": 1.02129507,
      "epoch": 0.395238381530692,
      "flos": 23110348170240.0,
      "grad_norm": 2.9087878593243284,
      "language_loss": 0.76917583,
      "learning_rate": 2.7565421561243654e-06,
      "loss": 0.79119939,
      "num_input_tokens_seen": 70811440,
      "step": 3287,
      "time_per_iteration": 2.6572914123535156
    },
    {
      "auxiliary_loss_clip": 0.01137617,
      "auxiliary_loss_mlp": 0.01033207,
      "balance_loss_clip": 1.04558539,
      "balance_loss_mlp": 1.02516675,
      "epoch": 0.3953586244213311,
      "flos": 24347614095360.0,
      "grad_norm": 1.9668230777160247,
      "language_loss": 0.82250762,
      "learning_rate": 2.7558210087017413e-06,
      "loss": 0.84421587,
      "num_input_tokens_seen": 70831375,
      "step": 3288,
      "time_per_iteration": 3.7082912921905518
    },
    {
      "auxiliary_loss_clip": 0.01139674,
      "auxiliary_loss_mlp": 0.01033536,
      "balance_loss_clip": 1.04968596,
      "balance_loss_mlp": 1.02549517,
      "epoch": 0.3954788673119702,
      "flos": 23440080044160.0,
      "grad_norm": 1.7895638363943807,
      "language_loss": 0.73397362,
      "learning_rate": 2.7550997466216724e-06,
      "loss": 0.75570571,
      "num_input_tokens_seen": 70849170,
      "step": 3289,
      "time_per_iteration": 3.6803553104400635
    },
    {
      "auxiliary_loss_clip": 0.01159861,
      "auxiliary_loss_mlp": 0.01028373,
      "balance_loss_clip": 1.05411792,
      "balance_loss_mlp": 1.01976037,
      "epoch": 0.3955991102026093,
      "flos": 17494063384320.0,
      "grad_norm": 1.8328788200485375,
      "language_loss": 0.81208616,
      "learning_rate": 2.7543783699935714e-06,
      "loss": 0.83396852,
      "num_input_tokens_seen": 70867200,
      "step": 3290,
      "time_per_iteration": 2.6653316020965576
    },
    {
      "auxiliary_loss_clip": 0.01173881,
      "auxiliary_loss_mlp": 0.01027185,
      "balance_loss_clip": 1.05380416,
      "balance_loss_mlp": 1.01898956,
      "epoch": 0.39571935309324835,
      "flos": 18221326053120.0,
      "grad_norm": 2.6464533646277224,
      "language_loss": 0.86718297,
      "learning_rate": 2.753656878926872e-06,
      "loss": 0.88919365,
      "num_input_tokens_seen": 70883080,
      "step": 3291,
      "time_per_iteration": 2.608905553817749
    },
    {
      "auxiliary_loss_clip": 0.01147676,
      "auxiliary_loss_mlp": 0.01031308,
      "balance_loss_clip": 1.0471493,
      "balance_loss_mlp": 1.02260602,
      "epoch": 0.39583959598388746,
      "flos": 17748813617280.0,
      "grad_norm": 1.7503547695595836,
      "language_loss": 0.74183547,
      "learning_rate": 2.752935273531023e-06,
      "loss": 0.76362532,
      "num_input_tokens_seen": 70901230,
      "step": 3292,
      "time_per_iteration": 2.7029242515563965
    },
    {
      "auxiliary_loss_clip": 0.01178577,
      "auxiliary_loss_mlp": 0.01034588,
      "balance_loss_clip": 1.05354905,
      "balance_loss_mlp": 1.02599263,
      "epoch": 0.39595983887452657,
      "flos": 19352368483200.0,
      "grad_norm": 1.7293157528522765,
      "language_loss": 0.78497589,
      "learning_rate": 2.752213553915492e-06,
      "loss": 0.80710757,
      "num_input_tokens_seen": 70919585,
      "step": 3293,
      "time_per_iteration": 3.510363817214966
    },
    {
      "auxiliary_loss_clip": 0.01061534,
      "auxiliary_loss_mlp": 0.01002257,
      "balance_loss_clip": 1.0151794,
      "balance_loss_mlp": 1.00062394,
      "epoch": 0.3960800817651656,
      "flos": 60682282940160.0,
      "grad_norm": 0.8134377214714756,
      "language_loss": 0.66127127,
      "learning_rate": 2.751491720189762e-06,
      "loss": 0.6819092,
      "num_input_tokens_seen": 70977695,
      "step": 3294,
      "time_per_iteration": 4.217949151992798
    },
    {
      "auxiliary_loss_clip": 0.01158755,
      "auxiliary_loss_mlp": 0.00763134,
      "balance_loss_clip": 1.05002201,
      "balance_loss_mlp": 1.00027895,
      "epoch": 0.39620032465580474,
      "flos": 16836718538880.0,
      "grad_norm": 2.2694241855612702,
      "language_loss": 0.91804922,
      "learning_rate": 2.7507697724633364e-06,
      "loss": 0.93726814,
      "num_input_tokens_seen": 70994455,
      "step": 3295,
      "time_per_iteration": 2.690751075744629
    },
    {
      "auxiliary_loss_clip": 0.01055583,
      "auxiliary_loss_mlp": 0.01003122,
      "balance_loss_clip": 1.02379584,
      "balance_loss_mlp": 1.00156045,
      "epoch": 0.3963205675464438,
      "flos": 69071445941760.0,
      "grad_norm": 0.7763478260805954,
      "language_loss": 0.54676288,
      "learning_rate": 2.7500477108457327e-06,
      "loss": 0.56734997,
      "num_input_tokens_seen": 71046465,
      "step": 3296,
      "time_per_iteration": 3.097503423690796
    },
    {
      "auxiliary_loss_clip": 0.01173127,
      "auxiliary_loss_mlp": 0.01032655,
      "balance_loss_clip": 1.05075109,
      "balance_loss_mlp": 1.02374411,
      "epoch": 0.3964408104370829,
      "flos": 25667439431040.0,
      "grad_norm": 1.8696170358202702,
      "language_loss": 0.81289721,
      "learning_rate": 2.7493255354464877e-06,
      "loss": 0.83495498,
      "num_input_tokens_seen": 71064275,
      "step": 3297,
      "time_per_iteration": 2.711761951446533
    },
    {
      "auxiliary_loss_clip": 0.01057849,
      "auxiliary_loss_mlp": 0.01032407,
      "balance_loss_clip": 1.03625727,
      "balance_loss_mlp": 1.02412796,
      "epoch": 0.396561053327722,
      "flos": 24277480790400.0,
      "grad_norm": 1.9823858527518616,
      "language_loss": 0.76462704,
      "learning_rate": 2.748603246375156e-06,
      "loss": 0.78552955,
      "num_input_tokens_seen": 71082290,
      "step": 3298,
      "time_per_iteration": 3.0387825965881348
    },
    {
      "auxiliary_loss_clip": 0.01189598,
      "auxiliary_loss_mlp": 0.01031896,
      "balance_loss_clip": 1.05392694,
      "balance_loss_mlp": 1.02360511,
      "epoch": 0.39668129621836107,
      "flos": 20522302364160.0,
      "grad_norm": 2.349871843356602,
      "language_loss": 0.69587123,
      "learning_rate": 2.7478808437413055e-06,
      "loss": 0.71808618,
      "num_input_tokens_seen": 71101700,
      "step": 3299,
      "time_per_iteration": 2.94031023979187
    },
    {
      "auxiliary_loss_clip": 0.01129248,
      "auxiliary_loss_mlp": 0.01029332,
      "balance_loss_clip": 1.04940367,
      "balance_loss_mlp": 1.02089787,
      "epoch": 0.3968015391090002,
      "flos": 27052585649280.0,
      "grad_norm": 2.008476693530805,
      "language_loss": 0.65646708,
      "learning_rate": 2.7471583276545263e-06,
      "loss": 0.67805278,
      "num_input_tokens_seen": 71122360,
      "step": 3300,
      "time_per_iteration": 2.7558059692382812
    },
    {
      "auxiliary_loss_clip": 0.01162537,
      "auxiliary_loss_mlp": 0.01026737,
      "balance_loss_clip": 1.04982829,
      "balance_loss_mlp": 1.0183568,
      "epoch": 0.3969217819996393,
      "flos": 12531819392640.0,
      "grad_norm": 1.7732128166686139,
      "language_loss": 0.7046622,
      "learning_rate": 2.7464356982244224e-06,
      "loss": 0.72655493,
      "num_input_tokens_seen": 71140360,
      "step": 3301,
      "time_per_iteration": 2.621948480606079
    },
    {
      "auxiliary_loss_clip": 0.01084109,
      "auxiliary_loss_mlp": 0.01005014,
      "balance_loss_clip": 1.02908111,
      "balance_loss_mlp": 1.00338066,
      "epoch": 0.39704202489027834,
      "flos": 66241399230720.0,
      "grad_norm": 0.7747660906312722,
      "language_loss": 0.61686563,
      "learning_rate": 2.745712955560617e-06,
      "loss": 0.63775682,
      "num_input_tokens_seen": 71196565,
      "step": 3302,
      "time_per_iteration": 3.197483777999878
    },
    {
      "auxiliary_loss_clip": 0.01114237,
      "auxiliary_loss_mlp": 0.01034798,
      "balance_loss_clip": 1.04474783,
      "balance_loss_mlp": 1.02623868,
      "epoch": 0.39716226778091746,
      "flos": 16982982720000.0,
      "grad_norm": 2.4999283362716835,
      "language_loss": 0.77523601,
      "learning_rate": 2.7449900997727496e-06,
      "loss": 0.79672635,
      "num_input_tokens_seen": 71214675,
      "step": 3303,
      "time_per_iteration": 2.7255303859710693
    },
    {
      "auxiliary_loss_clip": 0.01161312,
      "auxiliary_loss_mlp": 0.01030326,
      "balance_loss_clip": 1.05265915,
      "balance_loss_mlp": 1.02238107,
      "epoch": 0.39728251067155657,
      "flos": 23477139901440.0,
      "grad_norm": 2.5514167667598158,
      "language_loss": 0.84041446,
      "learning_rate": 2.744267130970476e-06,
      "loss": 0.86233079,
      "num_input_tokens_seen": 71234400,
      "step": 3304,
      "time_per_iteration": 2.7456612586975098
    },
    {
      "auxiliary_loss_clip": 0.01155175,
      "auxiliary_loss_mlp": 0.01026763,
      "balance_loss_clip": 1.05139291,
      "balance_loss_mlp": 1.01778615,
      "epoch": 0.3974027535621956,
      "flos": 20704441253760.0,
      "grad_norm": 2.5148021967809098,
      "language_loss": 0.76962405,
      "learning_rate": 2.7435440492634697e-06,
      "loss": 0.79144335,
      "num_input_tokens_seen": 71253725,
      "step": 3305,
      "time_per_iteration": 2.7264208793640137
    },
    {
      "auxiliary_loss_clip": 0.0116314,
      "auxiliary_loss_mlp": 0.01032642,
      "balance_loss_clip": 1.05099583,
      "balance_loss_mlp": 1.02259862,
      "epoch": 0.39752299645283473,
      "flos": 21543278544000.0,
      "grad_norm": 12.165097585589187,
      "language_loss": 0.67216229,
      "learning_rate": 2.7428208547614228e-06,
      "loss": 0.69412005,
      "num_input_tokens_seen": 71273220,
      "step": 3306,
      "time_per_iteration": 2.659529685974121
    },
    {
      "auxiliary_loss_clip": 0.01175735,
      "auxiliary_loss_mlp": 0.01031367,
      "balance_loss_clip": 1.05262423,
      "balance_loss_mlp": 1.02259362,
      "epoch": 0.39764323934347384,
      "flos": 19208295031680.0,
      "grad_norm": 1.8369676237465387,
      "language_loss": 0.76961166,
      "learning_rate": 2.742097547574043e-06,
      "loss": 0.79168272,
      "num_input_tokens_seen": 71291445,
      "step": 3307,
      "time_per_iteration": 2.8765435218811035
    },
    {
      "auxiliary_loss_clip": 0.01168439,
      "auxiliary_loss_mlp": 0.00763848,
      "balance_loss_clip": 1.05097914,
      "balance_loss_mlp": 1.00029302,
      "epoch": 0.3977634822341129,
      "flos": 20850202644480.0,
      "grad_norm": 1.8755128724099919,
      "language_loss": 0.77606028,
      "learning_rate": 2.7413741278110544e-06,
      "loss": 0.79538316,
      "num_input_tokens_seen": 71310135,
      "step": 3308,
      "time_per_iteration": 2.6444544792175293
    },
    {
      "auxiliary_loss_clip": 0.01166398,
      "auxiliary_loss_mlp": 0.01034785,
      "balance_loss_clip": 1.05355799,
      "balance_loss_mlp": 1.02555859,
      "epoch": 0.397883725124752,
      "flos": 39786042038400.0,
      "grad_norm": 2.040696608930875,
      "language_loss": 0.68837333,
      "learning_rate": 2.7406505955822016e-06,
      "loss": 0.7103852,
      "num_input_tokens_seen": 71331160,
      "step": 3309,
      "time_per_iteration": 2.8109352588653564
    },
    {
      "auxiliary_loss_clip": 0.0115898,
      "auxiliary_loss_mlp": 0.01033317,
      "balance_loss_clip": 1.04855728,
      "balance_loss_mlp": 1.02406669,
      "epoch": 0.39800396801539106,
      "flos": 17379507934080.0,
      "grad_norm": 5.533467068506392,
      "language_loss": 0.66017938,
      "learning_rate": 2.7399269509972415e-06,
      "loss": 0.68210244,
      "num_input_tokens_seen": 71345315,
      "step": 3310,
      "time_per_iteration": 2.645211696624756
    },
    {
      "auxiliary_loss_clip": 0.01154239,
      "auxiliary_loss_mlp": 0.01034664,
      "balance_loss_clip": 1.04563653,
      "balance_loss_mlp": 1.025455,
      "epoch": 0.3981242109060302,
      "flos": 19202764337280.0,
      "grad_norm": 2.423686333899025,
      "language_loss": 0.85175246,
      "learning_rate": 2.7392031941659514e-06,
      "loss": 0.87364155,
      "num_input_tokens_seen": 71363160,
      "step": 3311,
      "time_per_iteration": 2.664438009262085
    },
    {
      "auxiliary_loss_clip": 0.01163789,
      "auxiliary_loss_mlp": 0.01030214,
      "balance_loss_clip": 1.05502772,
      "balance_loss_mlp": 1.02180362,
      "epoch": 0.3982444537966693,
      "flos": 24565124903040.0,
      "grad_norm": 1.7576411331503077,
      "language_loss": 0.85898459,
      "learning_rate": 2.7384793251981244e-06,
      "loss": 0.8809247,
      "num_input_tokens_seen": 71382145,
      "step": 3312,
      "time_per_iteration": 2.7371463775634766
    },
    {
      "auxiliary_loss_clip": 0.01181101,
      "auxiliary_loss_mlp": 0.01027609,
      "balance_loss_clip": 1.05138075,
      "balance_loss_mlp": 1.01884079,
      "epoch": 0.39836469668730834,
      "flos": 26213856099840.0,
      "grad_norm": 2.2134529035006585,
      "language_loss": 0.80922896,
      "learning_rate": 2.737755344203571e-06,
      "loss": 0.83131599,
      "num_input_tokens_seen": 71402095,
      "step": 3313,
      "time_per_iteration": 2.6667544841766357
    },
    {
      "auxiliary_loss_clip": 0.01180943,
      "auxiliary_loss_mlp": 0.01038909,
      "balance_loss_clip": 1.05660915,
      "balance_loss_mlp": 1.02968788,
      "epoch": 0.39848493957794745,
      "flos": 27636134002560.0,
      "grad_norm": 1.70234152637495,
      "language_loss": 0.79997671,
      "learning_rate": 2.7370312512921186e-06,
      "loss": 0.82217526,
      "num_input_tokens_seen": 71423875,
      "step": 3314,
      "time_per_iteration": 4.052127122879028
    },
    {
      "auxiliary_loss_clip": 0.01165812,
      "auxiliary_loss_mlp": 0.01034841,
      "balance_loss_clip": 1.04987502,
      "balance_loss_mlp": 1.02529788,
      "epoch": 0.39860518246858656,
      "flos": 12239326944000.0,
      "grad_norm": 3.578503387245907,
      "language_loss": 0.77089137,
      "learning_rate": 2.736307046573611e-06,
      "loss": 0.79289794,
      "num_input_tokens_seen": 71439745,
      "step": 3315,
      "time_per_iteration": 3.555344343185425
    },
    {
      "auxiliary_loss_clip": 0.01189885,
      "auxiliary_loss_mlp": 0.01032491,
      "balance_loss_clip": 1.05355489,
      "balance_loss_mlp": 1.02408075,
      "epoch": 0.3987254253592256,
      "flos": 22379135005440.0,
      "grad_norm": 1.7943814777909188,
      "language_loss": 0.82075483,
      "learning_rate": 2.73558273015791e-06,
      "loss": 0.8429786,
      "num_input_tokens_seen": 71459575,
      "step": 3316,
      "time_per_iteration": 2.607576608657837
    },
    {
      "auxiliary_loss_clip": 0.01195831,
      "auxiliary_loss_mlp": 0.01031418,
      "balance_loss_clip": 1.05743909,
      "balance_loss_mlp": 1.0221858,
      "epoch": 0.3988456682498647,
      "flos": 23514020190720.0,
      "grad_norm": 2.441538984034888,
      "language_loss": 0.70620215,
      "learning_rate": 2.734858302154894e-06,
      "loss": 0.72847468,
      "num_input_tokens_seen": 71481075,
      "step": 3317,
      "time_per_iteration": 2.6018290519714355
    },
    {
      "auxiliary_loss_clip": 0.01160172,
      "auxiliary_loss_mlp": 0.01028004,
      "balance_loss_clip": 1.05379367,
      "balance_loss_mlp": 1.01918244,
      "epoch": 0.39896591114050384,
      "flos": 19208761908480.0,
      "grad_norm": 1.984612921237406,
      "language_loss": 0.7665956,
      "learning_rate": 2.734133762674457e-06,
      "loss": 0.78847736,
      "num_input_tokens_seen": 71500665,
      "step": 3318,
      "time_per_iteration": 2.6281869411468506
    },
    {
      "auxiliary_loss_clip": 0.01164232,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.05305111,
      "balance_loss_mlp": 1.0164547,
      "epoch": 0.3990861540311429,
      "flos": 28401031146240.0,
      "grad_norm": 2.4561909681602865,
      "language_loss": 0.70768166,
      "learning_rate": 2.7334091118265124e-06,
      "loss": 0.72957432,
      "num_input_tokens_seen": 71522560,
      "step": 3319,
      "time_per_iteration": 3.636202812194824
    },
    {
      "auxiliary_loss_clip": 0.01080012,
      "auxiliary_loss_mlp": 0.01001407,
      "balance_loss_clip": 1.01838899,
      "balance_loss_mlp": 0.99945182,
      "epoch": 0.399206396921782,
      "flos": 61758563086080.0,
      "grad_norm": 0.6805481396843899,
      "language_loss": 0.57817316,
      "learning_rate": 2.732684349720989e-06,
      "loss": 0.59898734,
      "num_input_tokens_seen": 71590520,
      "step": 3320,
      "time_per_iteration": 4.1919169425964355
    },
    {
      "auxiliary_loss_clip": 0.01153523,
      "auxiliary_loss_mlp": 0.01033542,
      "balance_loss_clip": 1.04948258,
      "balance_loss_mlp": 1.02439845,
      "epoch": 0.3993266398124211,
      "flos": 28074567409920.0,
      "grad_norm": 3.829545141629055,
      "language_loss": 0.75401735,
      "learning_rate": 2.7319594764678318e-06,
      "loss": 0.77588809,
      "num_input_tokens_seen": 71612620,
      "step": 3321,
      "time_per_iteration": 2.7326695919036865
    },
    {
      "auxiliary_loss_clip": 0.01134226,
      "auxiliary_loss_mlp": 0.010286,
      "balance_loss_clip": 1.0496093,
      "balance_loss_mlp": 1.01936126,
      "epoch": 0.39944688270306017,
      "flos": 23225083188480.0,
      "grad_norm": 1.8305349570287492,
      "language_loss": 0.82958949,
      "learning_rate": 2.7312344921770044e-06,
      "loss": 0.85121775,
      "num_input_tokens_seen": 71634320,
      "step": 3322,
      "time_per_iteration": 2.802638530731201
    },
    {
      "auxiliary_loss_clip": 0.01161665,
      "auxiliary_loss_mlp": 0.01029842,
      "balance_loss_clip": 1.04775381,
      "balance_loss_mlp": 1.02120531,
      "epoch": 0.3995671255936993,
      "flos": 19390433921280.0,
      "grad_norm": 2.115997871326739,
      "language_loss": 0.78557765,
      "learning_rate": 2.7305093969584857e-06,
      "loss": 0.80749273,
      "num_input_tokens_seen": 71653145,
      "step": 3323,
      "time_per_iteration": 2.6603786945343018
    },
    {
      "auxiliary_loss_clip": 0.01169971,
      "auxiliary_loss_mlp": 0.0103663,
      "balance_loss_clip": 1.05057502,
      "balance_loss_mlp": 1.02776074,
      "epoch": 0.3996873684843384,
      "flos": 23842638743040.0,
      "grad_norm": 1.9888715428305184,
      "language_loss": 0.80015278,
      "learning_rate": 2.729784190922272e-06,
      "loss": 0.82221878,
      "num_input_tokens_seen": 71674580,
      "step": 3324,
      "time_per_iteration": 2.6845834255218506
    },
    {
      "auxiliary_loss_clip": 0.01065909,
      "auxiliary_loss_mlp": 0.01000343,
      "balance_loss_clip": 1.01783121,
      "balance_loss_mlp": 0.99845934,
      "epoch": 0.39980761137497745,
      "flos": 66576877280640.0,
      "grad_norm": 0.9776976738978921,
      "language_loss": 0.57219929,
      "learning_rate": 2.729058874178378e-06,
      "loss": 0.59286183,
      "num_input_tokens_seen": 71745260,
      "step": 3325,
      "time_per_iteration": 3.371251106262207
    },
    {
      "auxiliary_loss_clip": 0.01168735,
      "auxiliary_loss_mlp": 0.01029463,
      "balance_loss_clip": 1.05478597,
      "balance_loss_mlp": 1.02034318,
      "epoch": 0.39992785426561656,
      "flos": 28549162834560.0,
      "grad_norm": 1.716591218333367,
      "language_loss": 0.69266146,
      "learning_rate": 2.7283334468368315e-06,
      "loss": 0.71464348,
      "num_input_tokens_seen": 71766540,
      "step": 3326,
      "time_per_iteration": 2.822084903717041
    },
    {
      "auxiliary_loss_clip": 0.01086083,
      "auxiliary_loss_mlp": 0.01039381,
      "balance_loss_clip": 1.03837895,
      "balance_loss_mlp": 1.03011239,
      "epoch": 0.4000480971562556,
      "flos": 15049408671360.0,
      "grad_norm": 2.098367929150586,
      "language_loss": 0.72689539,
      "learning_rate": 2.72760790900768e-06,
      "loss": 0.74815005,
      "num_input_tokens_seen": 71783125,
      "step": 3327,
      "time_per_iteration": 2.915095806121826
    },
    {
      "auxiliary_loss_clip": 0.01196851,
      "auxiliary_loss_mlp": 0.01028429,
      "balance_loss_clip": 1.05870819,
      "balance_loss_mlp": 1.01929784,
      "epoch": 0.4001683400468947,
      "flos": 23915609222400.0,
      "grad_norm": 1.9378903106590362,
      "language_loss": 0.78853273,
      "learning_rate": 2.7268822608009875e-06,
      "loss": 0.81078553,
      "num_input_tokens_seen": 71802500,
      "step": 3328,
      "time_per_iteration": 3.021928548812866
    },
    {
      "auxiliary_loss_clip": 0.01154705,
      "auxiliary_loss_mlp": 0.01030667,
      "balance_loss_clip": 1.05227137,
      "balance_loss_mlp": 1.02184582,
      "epoch": 0.40028858293753383,
      "flos": 24352677912960.0,
      "grad_norm": 1.769176814042345,
      "language_loss": 0.78424346,
      "learning_rate": 2.726156502326834e-06,
      "loss": 0.80609715,
      "num_input_tokens_seen": 71823800,
      "step": 3329,
      "time_per_iteration": 2.926999092102051
    },
    {
      "auxiliary_loss_clip": 0.01047138,
      "auxiliary_loss_mlp": 0.01013151,
      "balance_loss_clip": 1.02750325,
      "balance_loss_mlp": 1.01145864,
      "epoch": 0.4004088258281729,
      "flos": 66787025800320.0,
      "grad_norm": 0.7189636731023401,
      "language_loss": 0.60252976,
      "learning_rate": 2.725430633695316e-06,
      "loss": 0.62313271,
      "num_input_tokens_seen": 71886880,
      "step": 3330,
      "time_per_iteration": 3.39568829536438
    },
    {
      "auxiliary_loss_clip": 0.0108778,
      "auxiliary_loss_mlp": 0.0100236,
      "balance_loss_clip": 1.01690853,
      "balance_loss_mlp": 1.00045288,
      "epoch": 0.400529068718812,
      "flos": 58598386473600.0,
      "grad_norm": 0.8851323340201467,
      "language_loss": 0.57932782,
      "learning_rate": 2.7247046550165485e-06,
      "loss": 0.6002292,
      "num_input_tokens_seen": 71939005,
      "step": 3331,
      "time_per_iteration": 3.0423264503479004
    },
    {
      "auxiliary_loss_clip": 0.01195216,
      "auxiliary_loss_mlp": 0.01036351,
      "balance_loss_clip": 1.05819702,
      "balance_loss_mlp": 1.02706468,
      "epoch": 0.4006493116094511,
      "flos": 25377460934400.0,
      "grad_norm": 1.5004161641865668,
      "language_loss": 0.75982219,
      "learning_rate": 2.7239785664006606e-06,
      "loss": 0.78213787,
      "num_input_tokens_seen": 71962545,
      "step": 3332,
      "time_per_iteration": 2.649212121963501
    },
    {
      "auxiliary_loss_clip": 0.01078314,
      "auxiliary_loss_mlp": 0.01002087,
      "balance_loss_clip": 1.01594186,
      "balance_loss_mlp": 1.00019193,
      "epoch": 0.40076955450009016,
      "flos": 60280729822080.0,
      "grad_norm": 0.7750669154745256,
      "language_loss": 0.61813343,
      "learning_rate": 2.7232523679578002e-06,
      "loss": 0.63893747,
      "num_input_tokens_seen": 72025625,
      "step": 3333,
      "time_per_iteration": 3.2458696365356445
    },
    {
      "auxiliary_loss_clip": 0.01180633,
      "auxiliary_loss_mlp": 0.01029494,
      "balance_loss_clip": 1.05719852,
      "balance_loss_mlp": 1.02073801,
      "epoch": 0.4008897973907293,
      "flos": 16617268396800.0,
      "grad_norm": 2.1596398148786204,
      "language_loss": 0.79151446,
      "learning_rate": 2.7225260597981295e-06,
      "loss": 0.81361568,
      "num_input_tokens_seen": 72043330,
      "step": 3334,
      "time_per_iteration": 2.6059205532073975
    },
    {
      "auxiliary_loss_clip": 0.01147344,
      "auxiliary_loss_mlp": 0.00764893,
      "balance_loss_clip": 1.05241692,
      "balance_loss_mlp": 1.00026309,
      "epoch": 0.4010100402813684,
      "flos": 15377344865280.0,
      "grad_norm": 2.278370680455405,
      "language_loss": 0.78582919,
      "learning_rate": 2.721799642031831e-06,
      "loss": 0.80495155,
      "num_input_tokens_seen": 72059500,
      "step": 3335,
      "time_per_iteration": 2.6479475498199463
    },
    {
      "auxiliary_loss_clip": 0.01167957,
      "auxiliary_loss_mlp": 0.01031517,
      "balance_loss_clip": 1.04987609,
      "balance_loss_mlp": 1.02223039,
      "epoch": 0.40113028317200744,
      "flos": 13298835438720.0,
      "grad_norm": 2.046134127025573,
      "language_loss": 0.77734417,
      "learning_rate": 2.721073114769101e-06,
      "loss": 0.79933894,
      "num_input_tokens_seen": 72077175,
      "step": 3336,
      "time_per_iteration": 2.6756248474121094
    },
    {
      "auxiliary_loss_clip": 0.01147524,
      "auxiliary_loss_mlp": 0.01035808,
      "balance_loss_clip": 1.05191016,
      "balance_loss_mlp": 1.0269388,
      "epoch": 0.40125052606264655,
      "flos": 20668027841280.0,
      "grad_norm": 1.9107146883881425,
      "language_loss": 0.75309014,
      "learning_rate": 2.7203464781201523e-06,
      "loss": 0.7749235,
      "num_input_tokens_seen": 72096490,
      "step": 3337,
      "time_per_iteration": 2.743034839630127
    },
    {
      "auxiliary_loss_clip": 0.01195975,
      "auxiliary_loss_mlp": 0.01030916,
      "balance_loss_clip": 1.05724621,
      "balance_loss_mlp": 1.02160621,
      "epoch": 0.40137076895328566,
      "flos": 24607679541120.0,
      "grad_norm": 2.8376406164492463,
      "language_loss": 0.78402638,
      "learning_rate": 2.719619732195215e-06,
      "loss": 0.80629534,
      "num_input_tokens_seen": 72118130,
      "step": 3338,
      "time_per_iteration": 2.8128926753997803
    },
    {
      "auxiliary_loss_clip": 0.01148213,
      "auxiliary_loss_mlp": 0.01029467,
      "balance_loss_clip": 1.04947972,
      "balance_loss_mlp": 1.02024007,
      "epoch": 0.4014910118439247,
      "flos": 24206593299840.0,
      "grad_norm": 1.4420356161670347,
      "language_loss": 0.72776377,
      "learning_rate": 2.7188928771045377e-06,
      "loss": 0.74954057,
      "num_input_tokens_seen": 72139450,
      "step": 3339,
      "time_per_iteration": 2.7620186805725098
    },
    {
      "auxiliary_loss_clip": 0.01142272,
      "auxiliary_loss_mlp": 0.01037941,
      "balance_loss_clip": 1.04638624,
      "balance_loss_mlp": 1.02852905,
      "epoch": 0.4016112547345638,
      "flos": 26725080418560.0,
      "grad_norm": 1.710719677298961,
      "language_loss": 0.80215156,
      "learning_rate": 2.7181659129583815e-06,
      "loss": 0.82395369,
      "num_input_tokens_seen": 72159040,
      "step": 3340,
      "time_per_iteration": 5.0075767040252686
    },
    {
      "auxiliary_loss_clip": 0.01152122,
      "auxiliary_loss_mlp": 0.01037673,
      "balance_loss_clip": 1.04534245,
      "balance_loss_mlp": 1.02926898,
      "epoch": 0.4017314976252029,
      "flos": 21288025520640.0,
      "grad_norm": 1.8342339970227368,
      "language_loss": 0.7542851,
      "learning_rate": 2.7174388398670276e-06,
      "loss": 0.77618307,
      "num_input_tokens_seen": 72178220,
      "step": 3341,
      "time_per_iteration": 2.7579185962677
    },
    {
      "auxiliary_loss_clip": 0.01191278,
      "auxiliary_loss_mlp": 0.01039034,
      "balance_loss_clip": 1.05147755,
      "balance_loss_mlp": 1.0296042,
      "epoch": 0.401851740515842,
      "flos": 25484690010240.0,
      "grad_norm": 2.586494650451684,
      "language_loss": 0.92139453,
      "learning_rate": 2.716711657940773e-06,
      "loss": 0.94369769,
      "num_input_tokens_seen": 72199230,
      "step": 3342,
      "time_per_iteration": 2.75146746635437
    },
    {
      "auxiliary_loss_clip": 0.01052179,
      "auxiliary_loss_mlp": 0.01004187,
      "balance_loss_clip": 1.01323485,
      "balance_loss_mlp": 1.00231504,
      "epoch": 0.4019719834064811,
      "flos": 55395334978560.0,
      "grad_norm": 0.8140787817874595,
      "language_loss": 0.56506443,
      "learning_rate": 2.7159843672899284e-06,
      "loss": 0.58562815,
      "num_input_tokens_seen": 72263430,
      "step": 3343,
      "time_per_iteration": 3.5167477130889893
    },
    {
      "auxiliary_loss_clip": 0.01180066,
      "auxiliary_loss_mlp": 0.01028252,
      "balance_loss_clip": 1.05608511,
      "balance_loss_mlp": 1.01930583,
      "epoch": 0.40209222629712016,
      "flos": 18180100218240.0,
      "grad_norm": 2.4691193874804185,
      "language_loss": 0.81230664,
      "learning_rate": 2.715256968024825e-06,
      "loss": 0.83438981,
      "num_input_tokens_seen": 72280505,
      "step": 3344,
      "time_per_iteration": 2.8045685291290283
    },
    {
      "auxiliary_loss_clip": 0.01170843,
      "auxiliary_loss_mlp": 0.0103755,
      "balance_loss_clip": 1.05308807,
      "balance_loss_mlp": 1.02850509,
      "epoch": 0.40221246918775927,
      "flos": 25961009287680.0,
      "grad_norm": 1.6619518474644552,
      "language_loss": 0.82529593,
      "learning_rate": 2.7145294602558083e-06,
      "loss": 0.8473798,
      "num_input_tokens_seen": 72301215,
      "step": 3345,
      "time_per_iteration": 3.5982537269592285
    },
    {
      "auxiliary_loss_clip": 0.0117681,
      "auxiliary_loss_mlp": 0.01027381,
      "balance_loss_clip": 1.05270886,
      "balance_loss_mlp": 1.01854789,
      "epoch": 0.4023327120783984,
      "flos": 33838912056960.0,
      "grad_norm": 7.518337118082625,
      "language_loss": 0.70958918,
      "learning_rate": 2.713801844093241e-06,
      "loss": 0.7316311,
      "num_input_tokens_seen": 72322365,
      "step": 3346,
      "time_per_iteration": 3.8035991191864014
    },
    {
      "auxiliary_loss_clip": 0.01175828,
      "auxiliary_loss_mlp": 0.010331,
      "balance_loss_clip": 1.0522089,
      "balance_loss_mlp": 1.02454615,
      "epoch": 0.40245295496903744,
      "flos": 26900252069760.0,
      "grad_norm": 32.60576870945136,
      "language_loss": 0.88112581,
      "learning_rate": 2.7130741196475014e-06,
      "loss": 0.90321511,
      "num_input_tokens_seen": 72340495,
      "step": 3347,
      "time_per_iteration": 2.677321434020996
    },
    {
      "auxiliary_loss_clip": 0.01164544,
      "auxiliary_loss_mlp": 0.0103443,
      "balance_loss_clip": 1.05300975,
      "balance_loss_mlp": 1.02460098,
      "epoch": 0.40257319785967655,
      "flos": 36902738436480.0,
      "grad_norm": 2.0108775388843068,
      "language_loss": 0.79677248,
      "learning_rate": 2.7123462870289848e-06,
      "loss": 0.8187623,
      "num_input_tokens_seen": 72360545,
      "step": 3348,
      "time_per_iteration": 2.771833658218384
    },
    {
      "auxiliary_loss_clip": 0.01163743,
      "auxiliary_loss_mlp": 0.01025773,
      "balance_loss_clip": 1.04942977,
      "balance_loss_mlp": 1.01699924,
      "epoch": 0.40269344075031566,
      "flos": 24353180703360.0,
      "grad_norm": 1.7004776901740828,
      "language_loss": 0.81289601,
      "learning_rate": 2.711618346348102e-06,
      "loss": 0.83479124,
      "num_input_tokens_seen": 72381070,
      "step": 3349,
      "time_per_iteration": 2.755070209503174
    },
    {
      "auxiliary_loss_clip": 0.01154699,
      "auxiliary_loss_mlp": 0.01030103,
      "balance_loss_clip": 1.04930127,
      "balance_loss_mlp": 1.02197933,
      "epoch": 0.4028136836409547,
      "flos": 14389657614720.0,
      "grad_norm": 1.757028205950734,
      "language_loss": 0.6378935,
      "learning_rate": 2.7108902977152825e-06,
      "loss": 0.65974152,
      "num_input_tokens_seen": 72398970,
      "step": 3350,
      "time_per_iteration": 2.6395018100738525
    },
    {
      "auxiliary_loss_clip": 0.01173524,
      "auxiliary_loss_mlp": 0.01036605,
      "balance_loss_clip": 1.05040336,
      "balance_loss_mlp": 1.02716386,
      "epoch": 0.4029339265315938,
      "flos": 26136037284480.0,
      "grad_norm": 3.0150944152273556,
      "language_loss": 0.75292552,
      "learning_rate": 2.7101621412409704e-06,
      "loss": 0.7750268,
      "num_input_tokens_seen": 72418455,
      "step": 3351,
      "time_per_iteration": 2.703063488006592
    },
    {
      "auxiliary_loss_clip": 0.01190062,
      "auxiliary_loss_mlp": 0.01033664,
      "balance_loss_clip": 1.05204475,
      "balance_loss_mlp": 1.02421069,
      "epoch": 0.40305416942223293,
      "flos": 23256325042560.0,
      "grad_norm": 1.9372006311851622,
      "language_loss": 0.85830271,
      "learning_rate": 2.7094338770356256e-06,
      "loss": 0.88053989,
      "num_input_tokens_seen": 72437540,
      "step": 3352,
      "time_per_iteration": 2.581650495529175
    },
    {
      "auxiliary_loss_clip": 0.0115904,
      "auxiliary_loss_mlp": 0.01028213,
      "balance_loss_clip": 1.05164421,
      "balance_loss_mlp": 1.01915264,
      "epoch": 0.403174412312872,
      "flos": 27089645506560.0,
      "grad_norm": 2.3727004174171764,
      "language_loss": 0.64259219,
      "learning_rate": 2.708705505209726e-06,
      "loss": 0.66446477,
      "num_input_tokens_seen": 72458315,
      "step": 3353,
      "time_per_iteration": 2.7378830909729004
    },
    {
      "auxiliary_loss_clip": 0.01122975,
      "auxiliary_loss_mlp": 0.01031302,
      "balance_loss_clip": 1.04393339,
      "balance_loss_mlp": 1.02265382,
      "epoch": 0.4032946552035111,
      "flos": 21756336065280.0,
      "grad_norm": 2.060602107262452,
      "language_loss": 0.91677594,
      "learning_rate": 2.7079770258737646e-06,
      "loss": 0.93831873,
      "num_input_tokens_seen": 72476225,
      "step": 3354,
      "time_per_iteration": 2.74843692779541
    },
    {
      "auxiliary_loss_clip": 0.01141606,
      "auxiliary_loss_mlp": 0.01024966,
      "balance_loss_clip": 1.04623973,
      "balance_loss_mlp": 1.01554227,
      "epoch": 0.4034148980941502,
      "flos": 17343956448000.0,
      "grad_norm": 2.1771557639736643,
      "language_loss": 0.75590217,
      "learning_rate": 2.707248439138251e-06,
      "loss": 0.77756786,
      "num_input_tokens_seen": 72492460,
      "step": 3355,
      "time_per_iteration": 2.766432285308838
    },
    {
      "auxiliary_loss_clip": 0.01162175,
      "auxiliary_loss_mlp": 0.01028162,
      "balance_loss_clip": 1.05625677,
      "balance_loss_mlp": 1.01897132,
      "epoch": 0.40353514098478926,
      "flos": 22017838055040.0,
      "grad_norm": 1.6372049670255207,
      "language_loss": 0.65129638,
      "learning_rate": 2.7065197451137114e-06,
      "loss": 0.67319977,
      "num_input_tokens_seen": 72513840,
      "step": 3356,
      "time_per_iteration": 2.740413188934326
    },
    {
      "auxiliary_loss_clip": 0.01161119,
      "auxiliary_loss_mlp": 0.01032338,
      "balance_loss_clip": 1.05016232,
      "balance_loss_mlp": 1.02416027,
      "epoch": 0.4036553838754284,
      "flos": 14246446089600.0,
      "grad_norm": 2.141516832158929,
      "language_loss": 0.67554617,
      "learning_rate": 2.7057909439106894e-06,
      "loss": 0.69748074,
      "num_input_tokens_seen": 72531695,
      "step": 3357,
      "time_per_iteration": 2.6938605308532715
    },
    {
      "auxiliary_loss_clip": 0.01168132,
      "auxiliary_loss_mlp": 0.00764163,
      "balance_loss_clip": 1.05079746,
      "balance_loss_mlp": 1.00029349,
      "epoch": 0.40377562676606743,
      "flos": 24790644443520.0,
      "grad_norm": 1.8912302859982433,
      "language_loss": 0.7859785,
      "learning_rate": 2.7050620356397417e-06,
      "loss": 0.80530143,
      "num_input_tokens_seen": 72550645,
      "step": 3358,
      "time_per_iteration": 2.7090680599212646
    },
    {
      "auxiliary_loss_clip": 0.01190659,
      "auxiliary_loss_mlp": 0.01031667,
      "balance_loss_clip": 1.05770409,
      "balance_loss_mlp": 1.02336454,
      "epoch": 0.40389586965670654,
      "flos": 24061226958720.0,
      "grad_norm": 1.7863046769943873,
      "language_loss": 0.72286141,
      "learning_rate": 2.7043330204114437e-06,
      "loss": 0.74508464,
      "num_input_tokens_seen": 72569355,
      "step": 3359,
      "time_per_iteration": 2.636376142501831
    },
    {
      "auxiliary_loss_clip": 0.01184721,
      "auxiliary_loss_mlp": 0.01032839,
      "balance_loss_clip": 1.05042005,
      "balance_loss_mlp": 1.02388668,
      "epoch": 0.40401611254734565,
      "flos": 16399613934720.0,
      "grad_norm": 1.8599530996429623,
      "language_loss": 0.8577776,
      "learning_rate": 2.7036038983363862e-06,
      "loss": 0.87995321,
      "num_input_tokens_seen": 72585960,
      "step": 3360,
      "time_per_iteration": 2.6341536045074463
    },
    {
      "auxiliary_loss_clip": 0.01170526,
      "auxiliary_loss_mlp": 0.01029514,
      "balance_loss_clip": 1.05127835,
      "balance_loss_mlp": 1.02185464,
      "epoch": 0.4041363554379847,
      "flos": 23988220565760.0,
      "grad_norm": 1.9301631413395646,
      "language_loss": 0.84443474,
      "learning_rate": 2.702874669525177e-06,
      "loss": 0.86643511,
      "num_input_tokens_seen": 72604440,
      "step": 3361,
      "time_per_iteration": 2.7052724361419678
    },
    {
      "auxiliary_loss_clip": 0.01149569,
      "auxiliary_loss_mlp": 0.01038703,
      "balance_loss_clip": 1.05292308,
      "balance_loss_mlp": 1.02945244,
      "epoch": 0.4042565983286238,
      "flos": 28401964899840.0,
      "grad_norm": 2.3701503356677978,
      "language_loss": 0.69578612,
      "learning_rate": 2.7021453340884394e-06,
      "loss": 0.71766883,
      "num_input_tokens_seen": 72622165,
      "step": 3362,
      "time_per_iteration": 2.7761149406433105
    },
    {
      "auxiliary_loss_clip": 0.01150716,
      "auxiliary_loss_mlp": 0.00764081,
      "balance_loss_clip": 1.04915333,
      "balance_loss_mlp": 1.00030756,
      "epoch": 0.40437684121926293,
      "flos": 17710963660800.0,
      "grad_norm": 2.6771337623613785,
      "language_loss": 0.73206699,
      "learning_rate": 2.7014158921368125e-06,
      "loss": 0.75121498,
      "num_input_tokens_seen": 72640490,
      "step": 3363,
      "time_per_iteration": 2.6142802238464355
    },
    {
      "auxiliary_loss_clip": 0.01190565,
      "auxiliary_loss_mlp": 0.0103315,
      "balance_loss_clip": 1.05473757,
      "balance_loss_mlp": 1.02410769,
      "epoch": 0.404497084109902,
      "flos": 24018959629440.0,
      "grad_norm": 2.1035454991349183,
      "language_loss": 0.85900688,
      "learning_rate": 2.700686343780953e-06,
      "loss": 0.88124406,
      "num_input_tokens_seen": 72660360,
      "step": 3364,
      "time_per_iteration": 2.667922019958496
    },
    {
      "auxiliary_loss_clip": 0.01161059,
      "auxiliary_loss_mlp": 0.01031087,
      "balance_loss_clip": 1.04934573,
      "balance_loss_mlp": 1.02266479,
      "epoch": 0.4046173270005411,
      "flos": 22929861306240.0,
      "grad_norm": 2.1521049039879734,
      "language_loss": 0.88914543,
      "learning_rate": 2.699956689131532e-06,
      "loss": 0.91106689,
      "num_input_tokens_seen": 72680345,
      "step": 3365,
      "time_per_iteration": 3.583500862121582
    },
    {
      "auxiliary_loss_clip": 0.01163903,
      "auxiliary_loss_mlp": 0.01031692,
      "balance_loss_clip": 1.05144024,
      "balance_loss_mlp": 1.0231154,
      "epoch": 0.4047375698911802,
      "flos": 20668135582080.0,
      "grad_norm": 2.1756835897602143,
      "language_loss": 0.84939182,
      "learning_rate": 2.699226928299238e-06,
      "loss": 0.87134778,
      "num_input_tokens_seen": 72698365,
      "step": 3366,
      "time_per_iteration": 3.6626839637756348
    },
    {
      "auxiliary_loss_clip": 0.01176438,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 1.05261278,
      "balance_loss_mlp": 1.02144301,
      "epoch": 0.40485781278181926,
      "flos": 28912865996160.0,
      "grad_norm": 2.5999735884616757,
      "language_loss": 0.79622382,
      "learning_rate": 2.698497061394774e-06,
      "loss": 0.81828254,
      "num_input_tokens_seen": 72716850,
      "step": 3367,
      "time_per_iteration": 2.698781728744507
    },
    {
      "auxiliary_loss_clip": 0.01154656,
      "auxiliary_loss_mlp": 0.00763424,
      "balance_loss_clip": 1.05391335,
      "balance_loss_mlp": 1.00035322,
      "epoch": 0.40497805567245837,
      "flos": 23148377694720.0,
      "grad_norm": 1.643268987352253,
      "language_loss": 0.80848855,
      "learning_rate": 2.6977670885288627e-06,
      "loss": 0.82766938,
      "num_input_tokens_seen": 72738250,
      "step": 3368,
      "time_per_iteration": 2.7300407886505127
    },
    {
      "auxiliary_loss_clip": 0.0114866,
      "auxiliary_loss_mlp": 0.01031353,
      "balance_loss_clip": 1.04760242,
      "balance_loss_mlp": 1.0229249,
      "epoch": 0.4050982985630975,
      "flos": 16289404030080.0,
      "grad_norm": 2.180802719252512,
      "language_loss": 0.75093657,
      "learning_rate": 2.6970370098122378e-06,
      "loss": 0.77273673,
      "num_input_tokens_seen": 72755235,
      "step": 3369,
      "time_per_iteration": 2.7568423748016357
    },
    {
      "auxiliary_loss_clip": 0.0118753,
      "auxiliary_loss_mlp": 0.01033202,
      "balance_loss_clip": 1.05186391,
      "balance_loss_mlp": 1.02507782,
      "epoch": 0.40521854145373654,
      "flos": 34459484353920.0,
      "grad_norm": 1.8653736204988296,
      "language_loss": 0.86856151,
      "learning_rate": 2.6963068253556535e-06,
      "loss": 0.89076877,
      "num_input_tokens_seen": 72776620,
      "step": 3370,
      "time_per_iteration": 3.896428346633911
    },
    {
      "auxiliary_loss_clip": 0.01180595,
      "auxiliary_loss_mlp": 0.01033158,
      "balance_loss_clip": 1.04958153,
      "balance_loss_mlp": 1.02418184,
      "epoch": 0.40533878434437565,
      "flos": 25331099454720.0,
      "grad_norm": 1.9435232898736479,
      "language_loss": 0.85872042,
      "learning_rate": 2.6955765352698763e-06,
      "loss": 0.88085788,
      "num_input_tokens_seen": 72796765,
      "step": 3371,
      "time_per_iteration": 3.650041103363037
    },
    {
      "auxiliary_loss_clip": 0.01191239,
      "auxiliary_loss_mlp": 0.01030349,
      "balance_loss_clip": 1.05096948,
      "balance_loss_mlp": 1.02173603,
      "epoch": 0.40545902723501476,
      "flos": 15012061505280.0,
      "grad_norm": 6.573987432495985,
      "language_loss": 0.73648334,
      "learning_rate": 2.6948461396656923e-06,
      "loss": 0.75869918,
      "num_input_tokens_seen": 72814175,
      "step": 3372,
      "time_per_iteration": 2.617652177810669
    },
    {
      "auxiliary_loss_clip": 0.01183671,
      "auxiliary_loss_mlp": 0.01033315,
      "balance_loss_clip": 1.0555706,
      "balance_loss_mlp": 1.02473736,
      "epoch": 0.4055792701256538,
      "flos": 25521103422720.0,
      "grad_norm": 2.157067615043618,
      "language_loss": 0.74362409,
      "learning_rate": 2.6941156386539013e-06,
      "loss": 0.76579392,
      "num_input_tokens_seen": 72834125,
      "step": 3373,
      "time_per_iteration": 2.6585164070129395
    },
    {
      "auxiliary_loss_clip": 0.01158761,
      "auxiliary_loss_mlp": 0.01029305,
      "balance_loss_clip": 1.0537219,
      "balance_loss_mlp": 1.02074027,
      "epoch": 0.4056995130162929,
      "flos": 19574583972480.0,
      "grad_norm": 2.8341682038193814,
      "language_loss": 0.80816978,
      "learning_rate": 2.6933850323453203e-06,
      "loss": 0.83005047,
      "num_input_tokens_seen": 72852570,
      "step": 3374,
      "time_per_iteration": 2.7157886028289795
    },
    {
      "auxiliary_loss_clip": 0.01192006,
      "auxiliary_loss_mlp": 0.01028852,
      "balance_loss_clip": 1.05635679,
      "balance_loss_mlp": 1.02115703,
      "epoch": 0.405819755906932,
      "flos": 15413794191360.0,
      "grad_norm": 1.9688582017360996,
      "language_loss": 0.74926436,
      "learning_rate": 2.6926543208507806e-06,
      "loss": 0.77147293,
      "num_input_tokens_seen": 72871250,
      "step": 3375,
      "time_per_iteration": 2.566619873046875
    },
    {
      "auxiliary_loss_clip": 0.01176093,
      "auxiliary_loss_mlp": 0.01028997,
      "balance_loss_clip": 1.05370927,
      "balance_loss_mlp": 1.02033067,
      "epoch": 0.4059399987975711,
      "flos": 21433930565760.0,
      "grad_norm": 2.1715544405746026,
      "language_loss": 0.80001694,
      "learning_rate": 2.6919235042811316e-06,
      "loss": 0.82206786,
      "num_input_tokens_seen": 72890035,
      "step": 3376,
      "time_per_iteration": 2.639587640762329
    },
    {
      "auxiliary_loss_clip": 0.01146142,
      "auxiliary_loss_mlp": 0.01035294,
      "balance_loss_clip": 1.04911065,
      "balance_loss_mlp": 1.02622223,
      "epoch": 0.4060602416882102,
      "flos": 25556942217600.0,
      "grad_norm": 2.1011870123449117,
      "language_loss": 0.76397562,
      "learning_rate": 2.691192582747237e-06,
      "loss": 0.78578997,
      "num_input_tokens_seen": 72909665,
      "step": 3377,
      "time_per_iteration": 2.7537338733673096
    },
    {
      "auxiliary_loss_clip": 0.01189236,
      "auxiliary_loss_mlp": 0.01029426,
      "balance_loss_clip": 1.05297899,
      "balance_loss_mlp": 1.02008581,
      "epoch": 0.40618048457884925,
      "flos": 23766759262080.0,
      "grad_norm": 1.8954163057110602,
      "language_loss": 0.74007678,
      "learning_rate": 2.6904615563599765e-06,
      "loss": 0.76226336,
      "num_input_tokens_seen": 72929465,
      "step": 3378,
      "time_per_iteration": 2.5806188583374023
    },
    {
      "auxiliary_loss_clip": 0.0114409,
      "auxiliary_loss_mlp": 0.01031018,
      "balance_loss_clip": 1.04802203,
      "balance_loss_mlp": 1.02266765,
      "epoch": 0.40630072746948837,
      "flos": 17639681120640.0,
      "grad_norm": 1.8291239729083915,
      "language_loss": 0.8340255,
      "learning_rate": 2.6897304252302477e-06,
      "loss": 0.85577655,
      "num_input_tokens_seen": 72946785,
      "step": 3379,
      "time_per_iteration": 2.713282585144043
    },
    {
      "auxiliary_loss_clip": 0.01060916,
      "auxiliary_loss_mlp": 0.01002812,
      "balance_loss_clip": 1.02419114,
      "balance_loss_mlp": 1.00098789,
      "epoch": 0.4064209703601275,
      "flos": 60836053063680.0,
      "grad_norm": 0.7901973507070418,
      "language_loss": 0.54813409,
      "learning_rate": 2.688999189468962e-06,
      "loss": 0.56877136,
      "num_input_tokens_seen": 73003215,
      "step": 3380,
      "time_per_iteration": 3.1721577644348145
    },
    {
      "auxiliary_loss_clip": 0.01174488,
      "auxiliary_loss_mlp": 0.01034014,
      "balance_loss_clip": 1.05347431,
      "balance_loss_mlp": 1.02543128,
      "epoch": 0.40654121325076653,
      "flos": 24024346669440.0,
      "grad_norm": 2.4687682464984353,
      "language_loss": 0.76413774,
      "learning_rate": 2.6882678491870464e-06,
      "loss": 0.78622282,
      "num_input_tokens_seen": 73023650,
      "step": 3381,
      "time_per_iteration": 2.7907159328460693
    },
    {
      "auxiliary_loss_clip": 0.01180341,
      "auxiliary_loss_mlp": 0.01029639,
      "balance_loss_clip": 1.05288553,
      "balance_loss_mlp": 1.02079356,
      "epoch": 0.40666145614140564,
      "flos": 27344252085120.0,
      "grad_norm": 1.717766393200777,
      "language_loss": 0.71349537,
      "learning_rate": 2.6875364044954453e-06,
      "loss": 0.73559523,
      "num_input_tokens_seen": 73043880,
      "step": 3382,
      "time_per_iteration": 2.712151288986206
    },
    {
      "auxiliary_loss_clip": 0.01156196,
      "auxiliary_loss_mlp": 0.01030155,
      "balance_loss_clip": 1.04575157,
      "balance_loss_mlp": 1.02151787,
      "epoch": 0.40678169903204475,
      "flos": 26176724415360.0,
      "grad_norm": 1.9748049474413685,
      "language_loss": 0.82353741,
      "learning_rate": 2.6868048555051185e-06,
      "loss": 0.84540093,
      "num_input_tokens_seen": 73065410,
      "step": 3383,
      "time_per_iteration": 2.7227067947387695
    },
    {
      "auxiliary_loss_clip": 0.0116666,
      "auxiliary_loss_mlp": 0.01031654,
      "balance_loss_clip": 1.04834521,
      "balance_loss_mlp": 1.02242756,
      "epoch": 0.4069019419226838,
      "flos": 28622420622720.0,
      "grad_norm": 2.3852997327647443,
      "language_loss": 0.85655761,
      "learning_rate": 2.686073202327041e-06,
      "loss": 0.87854081,
      "num_input_tokens_seen": 73084410,
      "step": 3384,
      "time_per_iteration": 2.7458460330963135
    },
    {
      "auxiliary_loss_clip": 0.01150117,
      "auxiliary_loss_mlp": 0.01025547,
      "balance_loss_clip": 1.04647613,
      "balance_loss_mlp": 1.01732755,
      "epoch": 0.4070221848133229,
      "flos": 25229006023680.0,
      "grad_norm": 1.6875833223374777,
      "language_loss": 0.73317641,
      "learning_rate": 2.6853414450722043e-06,
      "loss": 0.75493306,
      "num_input_tokens_seen": 73104075,
      "step": 3385,
      "time_per_iteration": 2.7055723667144775
    },
    {
      "auxiliary_loss_clip": 0.01171297,
      "auxiliary_loss_mlp": 0.01026374,
      "balance_loss_clip": 1.05050516,
      "balance_loss_mlp": 1.0186311,
      "epoch": 0.40714242770396203,
      "flos": 18405224709120.0,
      "grad_norm": 1.7306728182637594,
      "language_loss": 0.85444999,
      "learning_rate": 2.684609583851616e-06,
      "loss": 0.8764267,
      "num_input_tokens_seen": 73122250,
      "step": 3386,
      "time_per_iteration": 2.6221206188201904
    },
    {
      "auxiliary_loss_clip": 0.01127479,
      "auxiliary_loss_mlp": 0.01031461,
      "balance_loss_clip": 1.04606438,
      "balance_loss_mlp": 1.02276444,
      "epoch": 0.4072626705946011,
      "flos": 30228920403840.0,
      "grad_norm": 1.8117655803515722,
      "language_loss": 0.80765688,
      "learning_rate": 2.683877618776297e-06,
      "loss": 0.82924634,
      "num_input_tokens_seen": 73144505,
      "step": 3387,
      "time_per_iteration": 2.812366247177124
    },
    {
      "auxiliary_loss_clip": 0.01153505,
      "auxiliary_loss_mlp": 0.01034041,
      "balance_loss_clip": 1.04623628,
      "balance_loss_mlp": 1.02531481,
      "epoch": 0.4073829134852402,
      "flos": 21834549930240.0,
      "grad_norm": 2.604487577553351,
      "language_loss": 0.74287868,
      "learning_rate": 2.6831455499572876e-06,
      "loss": 0.76475412,
      "num_input_tokens_seen": 73162440,
      "step": 3388,
      "time_per_iteration": 2.657925605773926
    },
    {
      "auxiliary_loss_clip": 0.01188589,
      "auxiliary_loss_mlp": 0.01025997,
      "balance_loss_clip": 1.05174351,
      "balance_loss_mlp": 1.01787877,
      "epoch": 0.40750315637587925,
      "flos": 25260211964160.0,
      "grad_norm": 2.045994883049423,
      "language_loss": 0.77815062,
      "learning_rate": 2.682413377505641e-06,
      "loss": 0.80029655,
      "num_input_tokens_seen": 73181245,
      "step": 3389,
      "time_per_iteration": 2.6371636390686035
    },
    {
      "auxiliary_loss_clip": 0.01173587,
      "auxiliary_loss_mlp": 0.01027175,
      "balance_loss_clip": 1.04953229,
      "balance_loss_mlp": 1.01847839,
      "epoch": 0.40762339926651836,
      "flos": 19712767593600.0,
      "grad_norm": 1.8464700676363104,
      "language_loss": 0.76867771,
      "learning_rate": 2.6816811015324284e-06,
      "loss": 0.79068536,
      "num_input_tokens_seen": 73199295,
      "step": 3390,
      "time_per_iteration": 2.6720380783081055
    },
    {
      "auxiliary_loss_clip": 0.0109427,
      "auxiliary_loss_mlp": 0.01001607,
      "balance_loss_clip": 1.02310586,
      "balance_loss_mlp": 0.99989033,
      "epoch": 0.40774364215715747,
      "flos": 71449307314560.0,
      "grad_norm": 0.7237512255007107,
      "language_loss": 0.56643641,
      "learning_rate": 2.6809487221487343e-06,
      "loss": 0.58739519,
      "num_input_tokens_seen": 73258780,
      "step": 3391,
      "time_per_iteration": 4.084848165512085
    },
    {
      "auxiliary_loss_clip": 0.01164057,
      "auxiliary_loss_mlp": 0.0103385,
      "balance_loss_clip": 1.04781866,
      "balance_loss_mlp": 1.02515996,
      "epoch": 0.4078638850477965,
      "flos": 15084134144640.0,
      "grad_norm": 2.7868833274048543,
      "language_loss": 0.82000959,
      "learning_rate": 2.6802162394656605e-06,
      "loss": 0.84198868,
      "num_input_tokens_seen": 73275490,
      "step": 3392,
      "time_per_iteration": 2.577679395675659
    },
    {
      "auxiliary_loss_clip": 0.01154632,
      "auxiliary_loss_mlp": 0.01030057,
      "balance_loss_clip": 1.04547858,
      "balance_loss_mlp": 1.02170622,
      "epoch": 0.40798412793843564,
      "flos": 23842890138240.0,
      "grad_norm": 2.8351010719381717,
      "language_loss": 0.71510053,
      "learning_rate": 2.679483653594324e-06,
      "loss": 0.73694742,
      "num_input_tokens_seen": 73297260,
      "step": 3393,
      "time_per_iteration": 3.7427818775177
    },
    {
      "auxiliary_loss_clip": 0.01175253,
      "auxiliary_loss_mlp": 0.01031035,
      "balance_loss_clip": 1.05026102,
      "balance_loss_mlp": 1.02255917,
      "epoch": 0.40810437082907475,
      "flos": 21065774117760.0,
      "grad_norm": 2.565379910889756,
      "language_loss": 0.76393056,
      "learning_rate": 2.678750964645857e-06,
      "loss": 0.78599346,
      "num_input_tokens_seen": 73316340,
      "step": 3394,
      "time_per_iteration": 2.5611231327056885
    },
    {
      "auxiliary_loss_clip": 0.01178643,
      "auxiliary_loss_mlp": 0.01026788,
      "balance_loss_clip": 1.05636787,
      "balance_loss_mlp": 1.01800239,
      "epoch": 0.4082246137197138,
      "flos": 11321377948800.0,
      "grad_norm": 2.4046338615515728,
      "language_loss": 0.8404566,
      "learning_rate": 2.6780181727314094e-06,
      "loss": 0.86251092,
      "num_input_tokens_seen": 73331245,
      "step": 3395,
      "time_per_iteration": 2.5958919525146484
    },
    {
      "auxiliary_loss_clip": 0.0114637,
      "auxiliary_loss_mlp": 0.0076348,
      "balance_loss_clip": 1.04717457,
      "balance_loss_mlp": 1.00032616,
      "epoch": 0.4083448566103529,
      "flos": 19062569554560.0,
      "grad_norm": 1.9166557968971487,
      "language_loss": 0.78379416,
      "learning_rate": 2.6772852779621435e-06,
      "loss": 0.80289268,
      "num_input_tokens_seen": 73349105,
      "step": 3396,
      "time_per_iteration": 2.666388988494873
    },
    {
      "auxiliary_loss_clip": 0.01170029,
      "auxiliary_loss_mlp": 0.00762873,
      "balance_loss_clip": 1.05314028,
      "balance_loss_mlp": 1.000229,
      "epoch": 0.408465099500992,
      "flos": 23550254035200.0,
      "grad_norm": 1.876502527092022,
      "language_loss": 0.86291069,
      "learning_rate": 2.676552280449239e-06,
      "loss": 0.8822397,
      "num_input_tokens_seen": 73368990,
      "step": 3397,
      "time_per_iteration": 3.498594284057617
    },
    {
      "auxiliary_loss_clip": 0.01164568,
      "auxiliary_loss_mlp": 0.01028409,
      "balance_loss_clip": 1.04902732,
      "balance_loss_mlp": 1.02010596,
      "epoch": 0.4085853423916311,
      "flos": 12750012558720.0,
      "grad_norm": 2.3577291925068025,
      "language_loss": 0.75840855,
      "learning_rate": 2.6758191803038917e-06,
      "loss": 0.78033829,
      "num_input_tokens_seen": 73387485,
      "step": 3398,
      "time_per_iteration": 3.5802838802337646
    },
    {
      "auxiliary_loss_clip": 0.01110703,
      "auxiliary_loss_mlp": 0.01039361,
      "balance_loss_clip": 1.04559207,
      "balance_loss_mlp": 1.02989578,
      "epoch": 0.4087055852822702,
      "flos": 24353072962560.0,
      "grad_norm": 1.7800567481382366,
      "language_loss": 0.82922393,
      "learning_rate": 2.6750859776373125e-06,
      "loss": 0.85072458,
      "num_input_tokens_seen": 73406940,
      "step": 3399,
      "time_per_iteration": 2.7886383533477783
    },
    {
      "auxiliary_loss_clip": 0.01041,
      "auxiliary_loss_mlp": 0.01003054,
      "balance_loss_clip": 1.01945066,
      "balance_loss_mlp": 1.00109899,
      "epoch": 0.4088258281729093,
      "flos": 66387950720640.0,
      "grad_norm": 0.7667163126199344,
      "language_loss": 0.60375249,
      "learning_rate": 2.674352672560727e-06,
      "loss": 0.62419307,
      "num_input_tokens_seen": 73468385,
      "step": 3400,
      "time_per_iteration": 3.346703052520752
    },
    {
      "auxiliary_loss_clip": 0.01143639,
      "auxiliary_loss_mlp": 0.01026997,
      "balance_loss_clip": 1.04673028,
      "balance_loss_mlp": 1.01853943,
      "epoch": 0.40894607106354836,
      "flos": 20449260057600.0,
      "grad_norm": 1.724481869885827,
      "language_loss": 0.76783741,
      "learning_rate": 2.673619265185377e-06,
      "loss": 0.78954375,
      "num_input_tokens_seen": 73488225,
      "step": 3401,
      "time_per_iteration": 2.812732458114624
    },
    {
      "auxiliary_loss_clip": 0.01178329,
      "auxiliary_loss_mlp": 0.01034346,
      "balance_loss_clip": 1.05157769,
      "balance_loss_mlp": 1.02542281,
      "epoch": 0.40906631395418747,
      "flos": 27053627143680.0,
      "grad_norm": 1.7658095550985171,
      "language_loss": 0.77715123,
      "learning_rate": 2.672885755622521e-06,
      "loss": 0.79927796,
      "num_input_tokens_seen": 73510640,
      "step": 3402,
      "time_per_iteration": 2.6845688819885254
    },
    {
      "auxiliary_loss_clip": 0.01128467,
      "auxiliary_loss_mlp": 0.01025091,
      "balance_loss_clip": 1.04575968,
      "balance_loss_mlp": 1.01691866,
      "epoch": 0.4091865568448266,
      "flos": 25484151306240.0,
      "grad_norm": 9.118597132695399,
      "language_loss": 0.70786166,
      "learning_rate": 2.67215214398343e-06,
      "loss": 0.72939724,
      "num_input_tokens_seen": 73530655,
      "step": 3403,
      "time_per_iteration": 2.7989044189453125
    },
    {
      "auxiliary_loss_clip": 0.01135172,
      "auxiliary_loss_mlp": 0.01033915,
      "balance_loss_clip": 1.04608154,
      "balance_loss_mlp": 1.02506328,
      "epoch": 0.40930679973546563,
      "flos": 28657864368000.0,
      "grad_norm": 2.1375372352918096,
      "language_loss": 0.78425789,
      "learning_rate": 2.671418430379393e-06,
      "loss": 0.80594879,
      "num_input_tokens_seen": 73549340,
      "step": 3404,
      "time_per_iteration": 2.807591676712036
    },
    {
      "auxiliary_loss_clip": 0.01188453,
      "auxiliary_loss_mlp": 0.01026563,
      "balance_loss_clip": 1.05236602,
      "balance_loss_mlp": 1.01821184,
      "epoch": 0.40942704262610474,
      "flos": 20886292834560.0,
      "grad_norm": 1.7651770783118657,
      "language_loss": 0.83525592,
      "learning_rate": 2.670684614921715e-06,
      "loss": 0.85740602,
      "num_input_tokens_seen": 73568315,
      "step": 3405,
      "time_per_iteration": 2.5892527103424072
    },
    {
      "auxiliary_loss_clip": 0.01160564,
      "auxiliary_loss_mlp": 0.01031307,
      "balance_loss_clip": 1.04776287,
      "balance_loss_mlp": 1.02302814,
      "epoch": 0.4095472855167438,
      "flos": 21618080616960.0,
      "grad_norm": 2.1932933106323684,
      "language_loss": 0.69119108,
      "learning_rate": 2.6699506977217128e-06,
      "loss": 0.71310985,
      "num_input_tokens_seen": 73588490,
      "step": 3406,
      "time_per_iteration": 2.6777000427246094
    },
    {
      "auxiliary_loss_clip": 0.01173577,
      "auxiliary_loss_mlp": 0.01033464,
      "balance_loss_clip": 1.05414391,
      "balance_loss_mlp": 1.02510154,
      "epoch": 0.4096675284073829,
      "flos": 27926112499200.0,
      "grad_norm": 2.1453274378353124,
      "language_loss": 0.70135438,
      "learning_rate": 2.6692166788907233e-06,
      "loss": 0.72342479,
      "num_input_tokens_seen": 73608685,
      "step": 3407,
      "time_per_iteration": 2.732755184173584
    },
    {
      "auxiliary_loss_clip": 0.01160297,
      "auxiliary_loss_mlp": 0.01031703,
      "balance_loss_clip": 1.04830933,
      "balance_loss_mlp": 1.02328086,
      "epoch": 0.409787771298022,
      "flos": 19206607092480.0,
      "grad_norm": 1.9065886196249275,
      "language_loss": 0.77071106,
      "learning_rate": 2.6684825585400957e-06,
      "loss": 0.79263109,
      "num_input_tokens_seen": 73627630,
      "step": 3408,
      "time_per_iteration": 2.6374309062957764
    },
    {
      "auxiliary_loss_clip": 0.01060496,
      "auxiliary_loss_mlp": 0.01001674,
      "balance_loss_clip": 1.01616001,
      "balance_loss_mlp": 0.99992162,
      "epoch": 0.4099080141886611,
      "flos": 59269234832640.0,
      "grad_norm": 0.8119079647868245,
      "language_loss": 0.65156496,
      "learning_rate": 2.6677483367811947e-06,
      "loss": 0.67218673,
      "num_input_tokens_seen": 73687670,
      "step": 3409,
      "time_per_iteration": 3.340127944946289
    },
    {
      "auxiliary_loss_clip": 0.01174298,
      "auxiliary_loss_mlp": 0.01034407,
      "balance_loss_clip": 1.04949701,
      "balance_loss_mlp": 1.02584195,
      "epoch": 0.4100282570793002,
      "flos": 21906443001600.0,
      "grad_norm": 2.111189588988882,
      "language_loss": 0.75537926,
      "learning_rate": 2.6670140137254028e-06,
      "loss": 0.7774663,
      "num_input_tokens_seen": 73707145,
      "step": 3410,
      "time_per_iteration": 2.6366915702819824
    },
    {
      "auxiliary_loss_clip": 0.0112816,
      "auxiliary_loss_mlp": 0.01032881,
      "balance_loss_clip": 1.04612589,
      "balance_loss_mlp": 1.02464986,
      "epoch": 0.4101484999699393,
      "flos": 18551596631040.0,
      "grad_norm": 2.6902346197187903,
      "language_loss": 0.89712447,
      "learning_rate": 2.666279589484115e-06,
      "loss": 0.91873491,
      "num_input_tokens_seen": 73725045,
      "step": 3411,
      "time_per_iteration": 2.761784076690674
    },
    {
      "auxiliary_loss_clip": 0.01131548,
      "auxiliary_loss_mlp": 0.010273,
      "balance_loss_clip": 1.04573631,
      "balance_loss_mlp": 1.01892591,
      "epoch": 0.41026874286057835,
      "flos": 19094529680640.0,
      "grad_norm": 1.8478108182069197,
      "language_loss": 0.81199461,
      "learning_rate": 2.6655450641687435e-06,
      "loss": 0.83358312,
      "num_input_tokens_seen": 73742610,
      "step": 3412,
      "time_per_iteration": 2.832249164581299
    },
    {
      "auxiliary_loss_clip": 0.01189631,
      "auxiliary_loss_mlp": 0.01023616,
      "balance_loss_clip": 1.05518603,
      "balance_loss_mlp": 1.0157783,
      "epoch": 0.41038898575121746,
      "flos": 31209568588800.0,
      "grad_norm": 1.6503818140706048,
      "language_loss": 0.69280785,
      "learning_rate": 2.664810437890715e-06,
      "loss": 0.71494031,
      "num_input_tokens_seen": 73764280,
      "step": 3413,
      "time_per_iteration": 2.678687810897827
    },
    {
      "auxiliary_loss_clip": 0.01107998,
      "auxiliary_loss_mlp": 0.01032607,
      "balance_loss_clip": 1.04749167,
      "balance_loss_mlp": 1.02412486,
      "epoch": 0.41050922864185657,
      "flos": 14355865895040.0,
      "grad_norm": 2.142994583912524,
      "language_loss": 0.79584241,
      "learning_rate": 2.6640757107614714e-06,
      "loss": 0.81724846,
      "num_input_tokens_seen": 73782375,
      "step": 3414,
      "time_per_iteration": 2.7275562286376953
    },
    {
      "auxiliary_loss_clip": 0.01140217,
      "auxiliary_loss_mlp": 0.01031789,
      "balance_loss_clip": 1.04913783,
      "balance_loss_mlp": 1.02294993,
      "epoch": 0.4106294715324956,
      "flos": 30956290813440.0,
      "grad_norm": 2.3005841437079138,
      "language_loss": 0.69445038,
      "learning_rate": 2.6633408828924697e-06,
      "loss": 0.71617043,
      "num_input_tokens_seen": 73801240,
      "step": 3415,
      "time_per_iteration": 2.7826969623565674
    },
    {
      "auxiliary_loss_clip": 0.01156685,
      "auxiliary_loss_mlp": 0.01031555,
      "balance_loss_clip": 1.05076468,
      "balance_loss_mlp": 1.02272177,
      "epoch": 0.41074971442313474,
      "flos": 24457321209600.0,
      "grad_norm": 1.7275473696677672,
      "language_loss": 0.70420516,
      "learning_rate": 2.662605954395185e-06,
      "loss": 0.72608757,
      "num_input_tokens_seen": 73821200,
      "step": 3416,
      "time_per_iteration": 2.7817084789276123
    },
    {
      "auxiliary_loss_clip": 0.01176923,
      "auxiliary_loss_mlp": 0.01035977,
      "balance_loss_clip": 1.0513072,
      "balance_loss_mlp": 1.02798438,
      "epoch": 0.41086995731377385,
      "flos": 21542991235200.0,
      "grad_norm": 1.7247591756642005,
      "language_loss": 0.8394137,
      "learning_rate": 2.6618709253811027e-06,
      "loss": 0.8615427,
      "num_input_tokens_seen": 73840655,
      "step": 3417,
      "time_per_iteration": 3.6339950561523438
    },
    {
      "auxiliary_loss_clip": 0.01186513,
      "auxiliary_loss_mlp": 0.01026841,
      "balance_loss_clip": 1.05413246,
      "balance_loss_mlp": 1.01882458,
      "epoch": 0.4109902002044129,
      "flos": 20702753314560.0,
      "grad_norm": 9.627631549087427,
      "language_loss": 0.87844926,
      "learning_rate": 2.6611357959617277e-06,
      "loss": 0.90058279,
      "num_input_tokens_seen": 73860275,
      "step": 3418,
      "time_per_iteration": 3.645164728164673
    },
    {
      "auxiliary_loss_clip": 0.0113967,
      "auxiliary_loss_mlp": 0.01030765,
      "balance_loss_clip": 1.04807806,
      "balance_loss_mlp": 1.02261674,
      "epoch": 0.411110443095052,
      "flos": 18179992477440.0,
      "grad_norm": 1.8815169648959387,
      "language_loss": 0.90942311,
      "learning_rate": 2.660400566248578e-06,
      "loss": 0.93112743,
      "num_input_tokens_seen": 73878400,
      "step": 3419,
      "time_per_iteration": 2.7315635681152344
    },
    {
      "auxiliary_loss_clip": 0.01147863,
      "auxiliary_loss_mlp": 0.01034107,
      "balance_loss_clip": 1.04883707,
      "balance_loss_mlp": 1.02489841,
      "epoch": 0.41123068598569107,
      "flos": 14575244209920.0,
      "grad_norm": 2.535322785383011,
      "language_loss": 0.66755366,
      "learning_rate": 2.6596652363531876e-06,
      "loss": 0.68937343,
      "num_input_tokens_seen": 73894275,
      "step": 3420,
      "time_per_iteration": 2.653235673904419
    },
    {
      "auxiliary_loss_clip": 0.01188696,
      "auxiliary_loss_mlp": 0.01027683,
      "balance_loss_clip": 1.05423999,
      "balance_loss_mlp": 1.01937461,
      "epoch": 0.4113509288763302,
      "flos": 21177995184000.0,
      "grad_norm": 1.5481938353352083,
      "language_loss": 0.78199625,
      "learning_rate": 2.6589298063871055e-06,
      "loss": 0.80416006,
      "num_input_tokens_seen": 73914450,
      "step": 3421,
      "time_per_iteration": 2.6544947624206543
    },
    {
      "auxiliary_loss_clip": 0.01187973,
      "auxiliary_loss_mlp": 0.01030046,
      "balance_loss_clip": 1.05358195,
      "balance_loss_mlp": 1.02161813,
      "epoch": 0.4114711717669693,
      "flos": 18442212739200.0,
      "grad_norm": 1.8756455144250808,
      "language_loss": 0.69622803,
      "learning_rate": 2.658194276461895e-06,
      "loss": 0.71840823,
      "num_input_tokens_seen": 73932375,
      "step": 3422,
      "time_per_iteration": 3.4783692359924316
    },
    {
      "auxiliary_loss_clip": 0.01159108,
      "auxiliary_loss_mlp": 0.01036787,
      "balance_loss_clip": 1.04697871,
      "balance_loss_mlp": 1.02750611,
      "epoch": 0.41159141465760835,
      "flos": 27233395735680.0,
      "grad_norm": 1.8774703509976973,
      "language_loss": 0.67067403,
      "learning_rate": 2.6574586466891368e-06,
      "loss": 0.69263303,
      "num_input_tokens_seen": 73952850,
      "step": 3423,
      "time_per_iteration": 2.6711585521698
    },
    {
      "auxiliary_loss_clip": 0.01159157,
      "auxiliary_loss_mlp": 0.00763132,
      "balance_loss_clip": 1.04932582,
      "balance_loss_mlp": 1.00032043,
      "epoch": 0.41171165754824746,
      "flos": 20006876154240.0,
      "grad_norm": 1.9063783893889557,
      "language_loss": 0.64746243,
      "learning_rate": 2.6567229171804247e-06,
      "loss": 0.66668534,
      "num_input_tokens_seen": 73970735,
      "step": 3424,
      "time_per_iteration": 3.6242713928222656
    },
    {
      "auxiliary_loss_clip": 0.01153877,
      "auxiliary_loss_mlp": 0.01034942,
      "balance_loss_clip": 1.04698634,
      "balance_loss_mlp": 1.0261507,
      "epoch": 0.41183190043888657,
      "flos": 18004318035840.0,
      "grad_norm": 2.3429901222118192,
      "language_loss": 0.87766331,
      "learning_rate": 2.655987088047368e-06,
      "loss": 0.89955151,
      "num_input_tokens_seen": 73989080,
      "step": 3425,
      "time_per_iteration": 2.69061017036438
    },
    {
      "auxiliary_loss_clip": 0.01157239,
      "auxiliary_loss_mlp": 0.01038127,
      "balance_loss_clip": 1.04982352,
      "balance_loss_mlp": 1.02918005,
      "epoch": 0.4119521433295256,
      "flos": 27163370171520.0,
      "grad_norm": 2.1768971254464815,
      "language_loss": 0.78794837,
      "learning_rate": 2.6552511594015912e-06,
      "loss": 0.80990195,
      "num_input_tokens_seen": 74009470,
      "step": 3426,
      "time_per_iteration": 2.732583522796631
    },
    {
      "auxiliary_loss_clip": 0.01156598,
      "auxiliary_loss_mlp": 0.0103578,
      "balance_loss_clip": 1.04580045,
      "balance_loss_mlp": 1.02629709,
      "epoch": 0.41207238622016473,
      "flos": 15122020014720.0,
      "grad_norm": 2.010757128051907,
      "language_loss": 0.85543442,
      "learning_rate": 2.654515131354735e-06,
      "loss": 0.8773582,
      "num_input_tokens_seen": 74027735,
      "step": 3427,
      "time_per_iteration": 2.6264939308166504
    },
    {
      "auxiliary_loss_clip": 0.01149401,
      "auxiliary_loss_mlp": 0.01032513,
      "balance_loss_clip": 1.05148411,
      "balance_loss_mlp": 1.0243535,
      "epoch": 0.41219262911080384,
      "flos": 27052872958080.0,
      "grad_norm": 2.0251542465815975,
      "language_loss": 0.84853733,
      "learning_rate": 2.653779004018453e-06,
      "loss": 0.87035656,
      "num_input_tokens_seen": 74048300,
      "step": 3428,
      "time_per_iteration": 2.723256826400757
    },
    {
      "auxiliary_loss_clip": 0.01155386,
      "auxiliary_loss_mlp": 0.01030851,
      "balance_loss_clip": 1.04994297,
      "balance_loss_mlp": 1.02312016,
      "epoch": 0.4123128720014429,
      "flos": 24686360282880.0,
      "grad_norm": 1.7867335010547336,
      "language_loss": 0.82270563,
      "learning_rate": 2.653042777504417e-06,
      "loss": 0.84456789,
      "num_input_tokens_seen": 74070890,
      "step": 3429,
      "time_per_iteration": 2.7638304233551025
    },
    {
      "auxiliary_loss_clip": 0.01166645,
      "auxiliary_loss_mlp": 0.01037387,
      "balance_loss_clip": 1.04943812,
      "balance_loss_mlp": 1.02847624,
      "epoch": 0.412433114892082,
      "flos": 26244774731520.0,
      "grad_norm": 1.928449727108071,
      "language_loss": 0.80010736,
      "learning_rate": 2.6523064519243105e-06,
      "loss": 0.82214761,
      "num_input_tokens_seen": 74090460,
      "step": 3430,
      "time_per_iteration": 2.6562678813934326
    },
    {
      "auxiliary_loss_clip": 0.01175962,
      "auxiliary_loss_mlp": 0.01043032,
      "balance_loss_clip": 1.05391335,
      "balance_loss_mlp": 1.0333643,
      "epoch": 0.4125533577827211,
      "flos": 21361031913600.0,
      "grad_norm": 2.2784621248748294,
      "language_loss": 0.78951699,
      "learning_rate": 2.6515700273898333e-06,
      "loss": 0.8117069,
      "num_input_tokens_seen": 74108335,
      "step": 3431,
      "time_per_iteration": 2.652026414871216
    },
    {
      "auxiliary_loss_clip": 0.01152367,
      "auxiliary_loss_mlp": 0.01027504,
      "balance_loss_clip": 1.05094683,
      "balance_loss_mlp": 1.01835489,
      "epoch": 0.4126736006733602,
      "flos": 26067556005120.0,
      "grad_norm": 2.3361639286209557,
      "language_loss": 0.6936447,
      "learning_rate": 2.6508335040127018e-06,
      "loss": 0.71544337,
      "num_input_tokens_seen": 74128030,
      "step": 3432,
      "time_per_iteration": 2.657735586166382
    },
    {
      "auxiliary_loss_clip": 0.01179628,
      "auxiliary_loss_mlp": 0.01036122,
      "balance_loss_clip": 1.05417943,
      "balance_loss_mlp": 1.02709758,
      "epoch": 0.4127938435639993,
      "flos": 25666146541440.0,
      "grad_norm": 1.6860907290193277,
      "language_loss": 0.77411485,
      "learning_rate": 2.6500968819046446e-06,
      "loss": 0.7962724,
      "num_input_tokens_seen": 74148330,
      "step": 3433,
      "time_per_iteration": 2.6907362937927246
    },
    {
      "auxiliary_loss_clip": 0.01132461,
      "auxiliary_loss_mlp": 0.01031432,
      "balance_loss_clip": 1.04422295,
      "balance_loss_mlp": 1.02268207,
      "epoch": 0.4129140864546384,
      "flos": 17995914253440.0,
      "grad_norm": 2.7964847820670755,
      "language_loss": 0.58542073,
      "learning_rate": 2.649360161177408e-06,
      "loss": 0.60705966,
      "num_input_tokens_seen": 74163390,
      "step": 3434,
      "time_per_iteration": 2.6369788646698
    },
    {
      "auxiliary_loss_clip": 0.01182763,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.05294418,
      "balance_loss_mlp": 1.02006888,
      "epoch": 0.41303432934527745,
      "flos": 23732895715200.0,
      "grad_norm": 1.9960570044819903,
      "language_loss": 0.73900819,
      "learning_rate": 2.6486233419427504e-06,
      "loss": 0.76112092,
      "num_input_tokens_seen": 74183205,
      "step": 3435,
      "time_per_iteration": 2.679516077041626
    },
    {
      "auxiliary_loss_clip": 0.01139199,
      "auxiliary_loss_mlp": 0.01037654,
      "balance_loss_clip": 1.05006468,
      "balance_loss_mlp": 1.02847493,
      "epoch": 0.41315457223591656,
      "flos": 19755286318080.0,
      "grad_norm": 2.389838805316941,
      "language_loss": 0.75197947,
      "learning_rate": 2.6478864243124484e-06,
      "loss": 0.77374804,
      "num_input_tokens_seen": 74202870,
      "step": 3436,
      "time_per_iteration": 2.688663959503174
    },
    {
      "auxiliary_loss_clip": 0.01177733,
      "auxiliary_loss_mlp": 0.01026966,
      "balance_loss_clip": 1.05141926,
      "balance_loss_mlp": 1.01827526,
      "epoch": 0.4132748151265556,
      "flos": 20923316778240.0,
      "grad_norm": 2.138538675287534,
      "language_loss": 0.84963012,
      "learning_rate": 2.6471494083982903e-06,
      "loss": 0.87167716,
      "num_input_tokens_seen": 74222255,
      "step": 3437,
      "time_per_iteration": 2.70694637298584
    },
    {
      "auxiliary_loss_clip": 0.01148864,
      "auxiliary_loss_mlp": 0.01028724,
      "balance_loss_clip": 1.0478797,
      "balance_loss_mlp": 1.02014685,
      "epoch": 0.4133950580171947,
      "flos": 32232520016640.0,
      "grad_norm": 2.175602593059648,
      "language_loss": 0.74804032,
      "learning_rate": 2.6464122943120818e-06,
      "loss": 0.76981616,
      "num_input_tokens_seen": 74242480,
      "step": 3438,
      "time_per_iteration": 2.7653820514678955
    },
    {
      "auxiliary_loss_clip": 0.01144864,
      "auxiliary_loss_mlp": 0.01029601,
      "balance_loss_clip": 1.05083561,
      "balance_loss_mlp": 1.02124763,
      "epoch": 0.41351530090783384,
      "flos": 23292487059840.0,
      "grad_norm": 3.058443775271033,
      "language_loss": 0.82441491,
      "learning_rate": 2.645675082165642e-06,
      "loss": 0.84615958,
      "num_input_tokens_seen": 74258690,
      "step": 3439,
      "time_per_iteration": 2.7158288955688477
    },
    {
      "auxiliary_loss_clip": 0.01160485,
      "auxiliary_loss_mlp": 0.01028682,
      "balance_loss_clip": 1.0523243,
      "balance_loss_mlp": 1.02030766,
      "epoch": 0.4136355437984729,
      "flos": 25593571111680.0,
      "grad_norm": 2.104474401664527,
      "language_loss": 0.75467438,
      "learning_rate": 2.644937772070806e-06,
      "loss": 0.77656603,
      "num_input_tokens_seen": 74277135,
      "step": 3440,
      "time_per_iteration": 2.67889404296875
    },
    {
      "auxiliary_loss_clip": 0.01190622,
      "auxiliary_loss_mlp": 0.01031125,
      "balance_loss_clip": 1.05495977,
      "balance_loss_mlp": 1.02301908,
      "epoch": 0.413755786689112,
      "flos": 19828615933440.0,
      "grad_norm": 2.4726996750138355,
      "language_loss": 0.83096242,
      "learning_rate": 2.6442003641394225e-06,
      "loss": 0.85317987,
      "num_input_tokens_seen": 74294730,
      "step": 3441,
      "time_per_iteration": 2.6235013008117676
    },
    {
      "auxiliary_loss_clip": 0.0115458,
      "auxiliary_loss_mlp": 0.01026654,
      "balance_loss_clip": 1.04601216,
      "balance_loss_mlp": 1.01875067,
      "epoch": 0.4138760295797511,
      "flos": 26870446759680.0,
      "grad_norm": 1.6189487590758203,
      "language_loss": 0.84077823,
      "learning_rate": 2.643462858483356e-06,
      "loss": 0.86259055,
      "num_input_tokens_seen": 74315015,
      "step": 3442,
      "time_per_iteration": 2.704502820968628
    },
    {
      "auxiliary_loss_clip": 0.01126226,
      "auxiliary_loss_mlp": 0.01028058,
      "balance_loss_clip": 1.04583788,
      "balance_loss_mlp": 1.01907015,
      "epoch": 0.41399627247039017,
      "flos": 16399254798720.0,
      "grad_norm": 1.9322128268780854,
      "language_loss": 0.72560322,
      "learning_rate": 2.6427252552144856e-06,
      "loss": 0.74714601,
      "num_input_tokens_seen": 74333665,
      "step": 3443,
      "time_per_iteration": 2.726768732070923
    },
    {
      "auxiliary_loss_clip": 0.01189674,
      "auxiliary_loss_mlp": 0.01029521,
      "balance_loss_clip": 1.05364478,
      "balance_loss_mlp": 1.02059269,
      "epoch": 0.4141165153610293,
      "flos": 22930220442240.0,
      "grad_norm": 2.579008041045323,
      "language_loss": 0.75161958,
      "learning_rate": 2.6419875544447044e-06,
      "loss": 0.77381152,
      "num_input_tokens_seen": 74355065,
      "step": 3444,
      "time_per_iteration": 3.53420352935791
    },
    {
      "auxiliary_loss_clip": 0.01189621,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.05340052,
      "balance_loss_mlp": 1.01714563,
      "epoch": 0.4142367582516684,
      "flos": 25192556697600.0,
      "grad_norm": 2.9982187054761726,
      "language_loss": 0.71592963,
      "learning_rate": 2.6412497562859218e-06,
      "loss": 0.73807806,
      "num_input_tokens_seen": 74376345,
      "step": 3445,
      "time_per_iteration": 3.648343801498413
    },
    {
      "auxiliary_loss_clip": 0.0117731,
      "auxiliary_loss_mlp": 0.01035007,
      "balance_loss_clip": 1.05011392,
      "balance_loss_mlp": 1.02638245,
      "epoch": 0.41435700114230745,
      "flos": 21690476478720.0,
      "grad_norm": 2.5015572733997655,
      "language_loss": 0.76007223,
      "learning_rate": 2.6405118608500617e-06,
      "loss": 0.78219545,
      "num_input_tokens_seen": 74395170,
      "step": 3446,
      "time_per_iteration": 2.6316254138946533
    },
    {
      "auxiliary_loss_clip": 0.0114216,
      "auxiliary_loss_mlp": 0.01029205,
      "balance_loss_clip": 1.05453718,
      "balance_loss_mlp": 1.02112222,
      "epoch": 0.41447724403294656,
      "flos": 25995160143360.0,
      "grad_norm": 1.6768617700699564,
      "language_loss": 0.81488323,
      "learning_rate": 2.6397738682490613e-06,
      "loss": 0.83659685,
      "num_input_tokens_seen": 74416070,
      "step": 3447,
      "time_per_iteration": 2.8883109092712402
    },
    {
      "auxiliary_loss_clip": 0.01189902,
      "auxiliary_loss_mlp": 0.01029853,
      "balance_loss_clip": 1.0550462,
      "balance_loss_mlp": 1.02186573,
      "epoch": 0.41459748692358567,
      "flos": 18259678800000.0,
      "grad_norm": 1.8246867863719942,
      "language_loss": 0.75117499,
      "learning_rate": 2.6390357785948734e-06,
      "loss": 0.77337253,
      "num_input_tokens_seen": 74433185,
      "step": 3448,
      "time_per_iteration": 2.6066389083862305
    },
    {
      "auxiliary_loss_clip": 0.01173877,
      "auxiliary_loss_mlp": 0.0103092,
      "balance_loss_clip": 1.05240607,
      "balance_loss_mlp": 1.02224195,
      "epoch": 0.4147177298142247,
      "flos": 24168456034560.0,
      "grad_norm": 2.3559172640764974,
      "language_loss": 0.79929084,
      "learning_rate": 2.6382975919994667e-06,
      "loss": 0.82133883,
      "num_input_tokens_seen": 74453760,
      "step": 3449,
      "time_per_iteration": 4.54530143737793
    },
    {
      "auxiliary_loss_clip": 0.01160255,
      "auxiliary_loss_mlp": 0.01028137,
      "balance_loss_clip": 1.04877543,
      "balance_loss_mlp": 1.02031076,
      "epoch": 0.41483797270486383,
      "flos": 20084659056000.0,
      "grad_norm": 1.6146889275249017,
      "language_loss": 0.73408115,
      "learning_rate": 2.637559308574822e-06,
      "loss": 0.75596511,
      "num_input_tokens_seen": 74473505,
      "step": 3450,
      "time_per_iteration": 2.7554574012756348
    },
    {
      "auxiliary_loss_clip": 0.01188289,
      "auxiliary_loss_mlp": 0.01026786,
      "balance_loss_clip": 1.05348134,
      "balance_loss_mlp": 1.01839423,
      "epoch": 0.4149582155955029,
      "flos": 30081040110720.0,
      "grad_norm": 2.0144794436663287,
      "language_loss": 0.71425462,
      "learning_rate": 2.6368209284329376e-06,
      "loss": 0.73640537,
      "num_input_tokens_seen": 74494135,
      "step": 3451,
      "time_per_iteration": 2.678292751312256
    },
    {
      "auxiliary_loss_clip": 0.01171444,
      "auxiliary_loss_mlp": 0.01034709,
      "balance_loss_clip": 1.0485301,
      "balance_loss_mlp": 1.02603054,
      "epoch": 0.415078458486142,
      "flos": 16764394504320.0,
      "grad_norm": 1.9455408929824647,
      "language_loss": 0.75845838,
      "learning_rate": 2.636082451685825e-06,
      "loss": 0.78051996,
      "num_input_tokens_seen": 74512335,
      "step": 3452,
      "time_per_iteration": 2.6549808979034424
    },
    {
      "auxiliary_loss_clip": 0.01162168,
      "auxiliary_loss_mlp": 0.01034355,
      "balance_loss_clip": 1.05311465,
      "balance_loss_mlp": 1.02614737,
      "epoch": 0.4151987013767811,
      "flos": 26033692458240.0,
      "grad_norm": 1.627794785055567,
      "language_loss": 0.86395073,
      "learning_rate": 2.6353438784455094e-06,
      "loss": 0.88591594,
      "num_input_tokens_seen": 74535620,
      "step": 3453,
      "time_per_iteration": 2.7135393619537354
    },
    {
      "auxiliary_loss_clip": 0.01158393,
      "auxiliary_loss_mlp": 0.01034859,
      "balance_loss_clip": 1.05048299,
      "balance_loss_mlp": 1.02594197,
      "epoch": 0.41531894426742016,
      "flos": 24608002763520.0,
      "grad_norm": 2.180556666069311,
      "language_loss": 0.71795481,
      "learning_rate": 2.6346052088240326e-06,
      "loss": 0.7398873,
      "num_input_tokens_seen": 74555140,
      "step": 3454,
      "time_per_iteration": 2.734668254852295
    },
    {
      "auxiliary_loss_clip": 0.01158351,
      "auxiliary_loss_mlp": 0.01031171,
      "balance_loss_clip": 1.0481509,
      "balance_loss_mlp": 1.02276707,
      "epoch": 0.4154391871580593,
      "flos": 14975791747200.0,
      "grad_norm": 3.321985415273994,
      "language_loss": 0.77109122,
      "learning_rate": 2.63386644293345e-06,
      "loss": 0.79298639,
      "num_input_tokens_seen": 74571485,
      "step": 3455,
      "time_per_iteration": 2.6947431564331055
    },
    {
      "auxiliary_loss_clip": 0.01139859,
      "auxiliary_loss_mlp": 0.01028305,
      "balance_loss_clip": 1.04425478,
      "balance_loss_mlp": 1.02037716,
      "epoch": 0.4155594300486984,
      "flos": 14647173194880.0,
      "grad_norm": 2.9130662080399192,
      "language_loss": 0.82818758,
      "learning_rate": 2.633127580885833e-06,
      "loss": 0.84986913,
      "num_input_tokens_seen": 74585985,
      "step": 3456,
      "time_per_iteration": 2.6883790493011475
    },
    {
      "auxiliary_loss_clip": 0.01187186,
      "auxiliary_loss_mlp": 0.01029569,
      "balance_loss_clip": 1.0551945,
      "balance_loss_mlp": 1.02131939,
      "epoch": 0.41567967293933744,
      "flos": 29497276275840.0,
      "grad_norm": 1.947079291832593,
      "language_loss": 0.6511538,
      "learning_rate": 2.632388622793265e-06,
      "loss": 0.67332137,
      "num_input_tokens_seen": 74605140,
      "step": 3457,
      "time_per_iteration": 2.7083117961883545
    },
    {
      "auxiliary_loss_clip": 0.01175095,
      "auxiliary_loss_mlp": 0.01028957,
      "balance_loss_clip": 1.05440736,
      "balance_loss_mlp": 1.02164674,
      "epoch": 0.41579991582997655,
      "flos": 19238387650560.0,
      "grad_norm": 1.9189855756383594,
      "language_loss": 0.68408209,
      "learning_rate": 2.6316495687678457e-06,
      "loss": 0.70612258,
      "num_input_tokens_seen": 74623790,
      "step": 3458,
      "time_per_iteration": 2.7322962284088135
    },
    {
      "auxiliary_loss_clip": 0.01124004,
      "auxiliary_loss_mlp": 0.01030428,
      "balance_loss_clip": 1.04584885,
      "balance_loss_mlp": 1.02231002,
      "epoch": 0.41592015872061566,
      "flos": 24462061804800.0,
      "grad_norm": 2.490931793322291,
      "language_loss": 0.7655586,
      "learning_rate": 2.6309104189216887e-06,
      "loss": 0.78710288,
      "num_input_tokens_seen": 74641355,
      "step": 3459,
      "time_per_iteration": 2.8138561248779297
    },
    {
      "auxiliary_loss_clip": 0.01132751,
      "auxiliary_loss_mlp": 0.00763215,
      "balance_loss_clip": 1.046525,
      "balance_loss_mlp": 1.00033283,
      "epoch": 0.4160404016112547,
      "flos": 20775651966720.0,
      "grad_norm": 2.106232315719616,
      "language_loss": 0.75054574,
      "learning_rate": 2.630171173366923e-06,
      "loss": 0.76950538,
      "num_input_tokens_seen": 74657155,
      "step": 3460,
      "time_per_iteration": 2.7280468940734863
    },
    {
      "auxiliary_loss_clip": 0.01127874,
      "auxiliary_loss_mlp": 0.01027814,
      "balance_loss_clip": 1.0462085,
      "balance_loss_mlp": 1.01920068,
      "epoch": 0.41616064450189383,
      "flos": 13916462820480.0,
      "grad_norm": 2.2364271667394515,
      "language_loss": 0.74526012,
      "learning_rate": 2.629431832215691e-06,
      "loss": 0.76681703,
      "num_input_tokens_seen": 74671960,
      "step": 3461,
      "time_per_iteration": 2.7445452213287354
    },
    {
      "auxiliary_loss_clip": 0.01152852,
      "auxiliary_loss_mlp": 0.01025522,
      "balance_loss_clip": 1.05014396,
      "balance_loss_mlp": 1.01715934,
      "epoch": 0.41628088739253294,
      "flos": 20010826650240.0,
      "grad_norm": 1.675373474709286,
      "language_loss": 0.87255937,
      "learning_rate": 2.628692395580151e-06,
      "loss": 0.89434308,
      "num_input_tokens_seen": 74692050,
      "step": 3462,
      "time_per_iteration": 2.7479617595672607
    },
    {
      "auxiliary_loss_clip": 0.01095877,
      "auxiliary_loss_mlp": 0.01032229,
      "balance_loss_clip": 1.04014421,
      "balance_loss_mlp": 1.02424765,
      "epoch": 0.416401130283172,
      "flos": 29168801377920.0,
      "grad_norm": 2.1329157290533804,
      "language_loss": 0.79811943,
      "learning_rate": 2.6279528635724747e-06,
      "loss": 0.81940043,
      "num_input_tokens_seen": 74712205,
      "step": 3463,
      "time_per_iteration": 2.8461639881134033
    },
    {
      "auxiliary_loss_clip": 0.01170079,
      "auxiliary_loss_mlp": 0.01025059,
      "balance_loss_clip": 1.04846191,
      "balance_loss_mlp": 1.01681626,
      "epoch": 0.4165213731738111,
      "flos": 16246813478400.0,
      "grad_norm": 3.2682228932403024,
      "language_loss": 0.7883687,
      "learning_rate": 2.627213236304848e-06,
      "loss": 0.81032008,
      "num_input_tokens_seen": 74729005,
      "step": 3464,
      "time_per_iteration": 2.6360480785369873
    },
    {
      "auxiliary_loss_clip": 0.01174149,
      "auxiliary_loss_mlp": 0.0102961,
      "balance_loss_clip": 1.05086446,
      "balance_loss_mlp": 1.02193832,
      "epoch": 0.4166416160644502,
      "flos": 33765438787200.0,
      "grad_norm": 2.0498646495773487,
      "language_loss": 0.71076667,
      "learning_rate": 2.626473513889472e-06,
      "loss": 0.7328043,
      "num_input_tokens_seen": 74751385,
      "step": 3465,
      "time_per_iteration": 2.7906296253204346
    },
    {
      "auxiliary_loss_clip": 0.01164371,
      "auxiliary_loss_mlp": 0.01037318,
      "balance_loss_clip": 1.04854512,
      "balance_loss_mlp": 1.02938485,
      "epoch": 0.41676185895508927,
      "flos": 20917498775040.0,
      "grad_norm": 1.933758726291422,
      "language_loss": 0.8285774,
      "learning_rate": 2.625733696438562e-06,
      "loss": 0.85059428,
      "num_input_tokens_seen": 74768890,
      "step": 3466,
      "time_per_iteration": 2.6378533840179443
    },
    {
      "auxiliary_loss_clip": 0.01153293,
      "auxiliary_loss_mlp": 0.01030275,
      "balance_loss_clip": 1.04778707,
      "balance_loss_mlp": 1.02151942,
      "epoch": 0.4168821018457284,
      "flos": 18406122549120.0,
      "grad_norm": 1.6893305854417422,
      "language_loss": 0.75210321,
      "learning_rate": 2.6249937840643476e-06,
      "loss": 0.77393889,
      "num_input_tokens_seen": 74787195,
      "step": 3467,
      "time_per_iteration": 2.6766085624694824
    },
    {
      "auxiliary_loss_clip": 0.01188731,
      "auxiliary_loss_mlp": 0.00762918,
      "balance_loss_clip": 1.05548835,
      "balance_loss_mlp": 1.00046515,
      "epoch": 0.41700234473636744,
      "flos": 18698399516160.0,
      "grad_norm": 1.7083835791490265,
      "language_loss": 0.67110729,
      "learning_rate": 2.6242537768790733e-06,
      "loss": 0.69062376,
      "num_input_tokens_seen": 74806350,
      "step": 3468,
      "time_per_iteration": 2.5978121757507324
    },
    {
      "auxiliary_loss_clip": 0.01170644,
      "auxiliary_loss_mlp": 0.01028851,
      "balance_loss_clip": 1.05083823,
      "balance_loss_mlp": 1.02039337,
      "epoch": 0.41712258762700655,
      "flos": 31033283616000.0,
      "grad_norm": 1.783627595316444,
      "language_loss": 0.6866684,
      "learning_rate": 2.6235136749949975e-06,
      "loss": 0.70866334,
      "num_input_tokens_seen": 74829800,
      "step": 3469,
      "time_per_iteration": 3.630089282989502
    },
    {
      "auxiliary_loss_clip": 0.01184778,
      "auxiliary_loss_mlp": 0.01031123,
      "balance_loss_clip": 1.0513401,
      "balance_loss_mlp": 1.02324963,
      "epoch": 0.41724283051764566,
      "flos": 35914763877120.0,
      "grad_norm": 2.1910644875299967,
      "language_loss": 0.61547709,
      "learning_rate": 2.6227734785243924e-06,
      "loss": 0.63763613,
      "num_input_tokens_seen": 74849760,
      "step": 3470,
      "time_per_iteration": 2.8189282417297363
    },
    {
      "auxiliary_loss_clip": 0.01105629,
      "auxiliary_loss_mlp": 0.01027283,
      "balance_loss_clip": 1.04039872,
      "balance_loss_mlp": 1.01957631,
      "epoch": 0.4173630734082847,
      "flos": 25333649320320.0,
      "grad_norm": 1.7649569000693905,
      "language_loss": 0.79392701,
      "learning_rate": 2.6220331875795466e-06,
      "loss": 0.81525618,
      "num_input_tokens_seen": 74869110,
      "step": 3471,
      "time_per_iteration": 3.8060641288757324
    },
    {
      "auxiliary_loss_clip": 0.01169917,
      "auxiliary_loss_mlp": 0.01031863,
      "balance_loss_clip": 1.05213511,
      "balance_loss_mlp": 1.02310693,
      "epoch": 0.4174833162989238,
      "flos": 26685398868480.0,
      "grad_norm": 1.6794432390180742,
      "language_loss": 0.75377905,
      "learning_rate": 2.62129280227276e-06,
      "loss": 0.77579677,
      "num_input_tokens_seen": 74889110,
      "step": 3472,
      "time_per_iteration": 2.725999355316162
    },
    {
      "auxiliary_loss_clip": 0.01176128,
      "auxiliary_loss_mlp": 0.01038253,
      "balance_loss_clip": 1.05119348,
      "balance_loss_mlp": 1.03007507,
      "epoch": 0.41760355918956293,
      "flos": 74739584010240.0,
      "grad_norm": 2.1118188719547533,
      "language_loss": 0.68561637,
      "learning_rate": 2.62055232271635e-06,
      "loss": 0.7077601,
      "num_input_tokens_seen": 74916260,
      "step": 3473,
      "time_per_iteration": 3.062546968460083
    },
    {
      "auxiliary_loss_clip": 0.01135494,
      "auxiliary_loss_mlp": 0.01029302,
      "balance_loss_clip": 1.04556906,
      "balance_loss_mlp": 1.02083778,
      "epoch": 0.417723802080202,
      "flos": 14317513148160.0,
      "grad_norm": 2.0822162067017143,
      "language_loss": 0.87909168,
      "learning_rate": 2.619811749022646e-06,
      "loss": 0.90073967,
      "num_input_tokens_seen": 74931570,
      "step": 3474,
      "time_per_iteration": 2.65370512008667
    },
    {
      "auxiliary_loss_clip": 0.01175488,
      "auxiliary_loss_mlp": 0.01036272,
      "balance_loss_clip": 1.05412126,
      "balance_loss_mlp": 1.02676523,
      "epoch": 0.4178440449708411,
      "flos": 14643797316480.0,
      "grad_norm": 2.188612991979792,
      "language_loss": 0.71374488,
      "learning_rate": 2.6190710813039917e-06,
      "loss": 0.73586243,
      "num_input_tokens_seen": 74944695,
      "step": 3475,
      "time_per_iteration": 4.6343934535980225
    },
    {
      "auxiliary_loss_clip": 0.01122787,
      "auxiliary_loss_mlp": 0.00763388,
      "balance_loss_clip": 1.04175401,
      "balance_loss_mlp": 1.00026035,
      "epoch": 0.4179642878614802,
      "flos": 21507296094720.0,
      "grad_norm": 2.2090507424219212,
      "language_loss": 0.84153306,
      "learning_rate": 2.618330319672747e-06,
      "loss": 0.86039484,
      "num_input_tokens_seen": 74964115,
      "step": 3476,
      "time_per_iteration": 2.8584110736846924
    },
    {
      "auxiliary_loss_clip": 0.01188214,
      "auxiliary_loss_mlp": 0.01025883,
      "balance_loss_clip": 1.05357289,
      "balance_loss_mlp": 1.01797938,
      "epoch": 0.41808453075211927,
      "flos": 18441997257600.0,
      "grad_norm": 2.1960193070222442,
      "language_loss": 0.92070234,
      "learning_rate": 2.617589464241284e-06,
      "loss": 0.94284326,
      "num_input_tokens_seen": 74978515,
      "step": 3477,
      "time_per_iteration": 2.555351734161377
    },
    {
      "auxiliary_loss_clip": 0.01146963,
      "auxiliary_loss_mlp": 0.01031346,
      "balance_loss_clip": 1.04870176,
      "balance_loss_mlp": 1.02399659,
      "epoch": 0.4182047736427584,
      "flos": 20301020628480.0,
      "grad_norm": 2.615546260840198,
      "language_loss": 0.74643987,
      "learning_rate": 2.6168485151219914e-06,
      "loss": 0.76822299,
      "num_input_tokens_seen": 74998135,
      "step": 3478,
      "time_per_iteration": 2.7152771949768066
    },
    {
      "auxiliary_loss_clip": 0.01174292,
      "auxiliary_loss_mlp": 0.01028034,
      "balance_loss_clip": 1.05454421,
      "balance_loss_mlp": 1.02016056,
      "epoch": 0.4183250165333975,
      "flos": 18876623823360.0,
      "grad_norm": 2.9550968260440276,
      "language_loss": 0.71574187,
      "learning_rate": 2.616107472427269e-06,
      "loss": 0.73776507,
      "num_input_tokens_seen": 75012830,
      "step": 3479,
      "time_per_iteration": 2.5872716903686523
    },
    {
      "auxiliary_loss_clip": 0.01175523,
      "auxiliary_loss_mlp": 0.01035587,
      "balance_loss_clip": 1.04994488,
      "balance_loss_mlp": 1.02678943,
      "epoch": 0.41844525942403654,
      "flos": 17740050698880.0,
      "grad_norm": 2.5237533087946393,
      "language_loss": 0.76637131,
      "learning_rate": 2.615366336269533e-06,
      "loss": 0.78848237,
      "num_input_tokens_seen": 75026495,
      "step": 3480,
      "time_per_iteration": 2.6310482025146484
    },
    {
      "auxiliary_loss_clip": 0.01189773,
      "auxiliary_loss_mlp": 0.01028406,
      "balance_loss_clip": 1.05272388,
      "balance_loss_mlp": 1.0196383,
      "epoch": 0.41856550231467565,
      "flos": 18361377181440.0,
      "grad_norm": 2.274528555565257,
      "language_loss": 0.80402422,
      "learning_rate": 2.6146251067612126e-06,
      "loss": 0.82620597,
      "num_input_tokens_seen": 75041970,
      "step": 3481,
      "time_per_iteration": 2.553116798400879
    },
    {
      "auxiliary_loss_clip": 0.01171477,
      "auxiliary_loss_mlp": 0.01029699,
      "balance_loss_clip": 1.05303407,
      "balance_loss_mlp": 1.02143216,
      "epoch": 0.41868574520531476,
      "flos": 22781801445120.0,
      "grad_norm": 1.7801494720379165,
      "language_loss": 0.83029985,
      "learning_rate": 2.6138837840147525e-06,
      "loss": 0.85231155,
      "num_input_tokens_seen": 75061005,
      "step": 3482,
      "time_per_iteration": 2.6493980884552
    },
    {
      "auxiliary_loss_clip": 0.01140433,
      "auxiliary_loss_mlp": 0.01027716,
      "balance_loss_clip": 1.04650402,
      "balance_loss_mlp": 1.01972294,
      "epoch": 0.4188059880959538,
      "flos": 13699167494400.0,
      "grad_norm": 1.9407853367226497,
      "language_loss": 0.76713479,
      "learning_rate": 2.6131423681426103e-06,
      "loss": 0.78881633,
      "num_input_tokens_seen": 75076920,
      "step": 3483,
      "time_per_iteration": 2.6533124446868896
    },
    {
      "auxiliary_loss_clip": 0.01186302,
      "auxiliary_loss_mlp": 0.01030471,
      "balance_loss_clip": 1.05354261,
      "balance_loss_mlp": 1.02307749,
      "epoch": 0.41892623098659293,
      "flos": 37818281220480.0,
      "grad_norm": 2.1163683964241353,
      "language_loss": 0.73243648,
      "learning_rate": 2.6124008592572587e-06,
      "loss": 0.75460422,
      "num_input_tokens_seen": 75100905,
      "step": 3484,
      "time_per_iteration": 2.77797794342041
    },
    {
      "auxiliary_loss_clip": 0.01187856,
      "auxiliary_loss_mlp": 0.01026029,
      "balance_loss_clip": 1.05148411,
      "balance_loss_mlp": 1.01734471,
      "epoch": 0.419046473877232,
      "flos": 23258874908160.0,
      "grad_norm": 2.2359405730952493,
      "language_loss": 0.82259643,
      "learning_rate": 2.6116592574711835e-06,
      "loss": 0.84473526,
      "num_input_tokens_seen": 75119205,
      "step": 3485,
      "time_per_iteration": 2.592000961303711
    },
    {
      "auxiliary_loss_clip": 0.01189677,
      "auxiliary_loss_mlp": 0.01029901,
      "balance_loss_clip": 1.05362463,
      "balance_loss_mlp": 1.02197337,
      "epoch": 0.4191667167678711,
      "flos": 20741034234240.0,
      "grad_norm": 1.9063379827170264,
      "language_loss": 0.84463942,
      "learning_rate": 2.6109175628968853e-06,
      "loss": 0.86683524,
      "num_input_tokens_seen": 75138970,
      "step": 3486,
      "time_per_iteration": 2.6076834201812744
    },
    {
      "auxiliary_loss_clip": 0.0115966,
      "auxiliary_loss_mlp": 0.01027507,
      "balance_loss_clip": 1.04631937,
      "balance_loss_mlp": 1.01897204,
      "epoch": 0.4192869596585102,
      "flos": 23586416052480.0,
      "grad_norm": 2.0422084935262825,
      "language_loss": 0.82944775,
      "learning_rate": 2.610175775646878e-06,
      "loss": 0.85131943,
      "num_input_tokens_seen": 75157550,
      "step": 3487,
      "time_per_iteration": 2.6416242122650146
    },
    {
      "auxiliary_loss_clip": 0.01154696,
      "auxiliary_loss_mlp": 0.01028914,
      "balance_loss_clip": 1.04753566,
      "balance_loss_mlp": 1.01993155,
      "epoch": 0.41940720254914926,
      "flos": 25081269384960.0,
      "grad_norm": 2.1825446265148267,
      "language_loss": 0.73407805,
      "learning_rate": 2.6094338958336907e-06,
      "loss": 0.75591415,
      "num_input_tokens_seen": 75176220,
      "step": 3488,
      "time_per_iteration": 2.6953697204589844
    },
    {
      "auxiliary_loss_clip": 0.01161148,
      "auxiliary_loss_mlp": 0.01026544,
      "balance_loss_clip": 1.05236435,
      "balance_loss_mlp": 1.01881897,
      "epoch": 0.41952744543978837,
      "flos": 15554132628480.0,
      "grad_norm": 1.9888104546797019,
      "language_loss": 0.82260889,
      "learning_rate": 2.608691923569867e-06,
      "loss": 0.84448576,
      "num_input_tokens_seen": 75193095,
      "step": 3489,
      "time_per_iteration": 2.6333351135253906
    },
    {
      "auxiliary_loss_clip": 0.01177685,
      "auxiliary_loss_mlp": 0.01026675,
      "balance_loss_clip": 1.0548265,
      "balance_loss_mlp": 1.01877761,
      "epoch": 0.4196476883304275,
      "flos": 24644775312000.0,
      "grad_norm": 1.641227764346916,
      "language_loss": 0.75737971,
      "learning_rate": 2.6079498589679616e-06,
      "loss": 0.7794233,
      "num_input_tokens_seen": 75214185,
      "step": 3490,
      "time_per_iteration": 2.6646971702575684
    },
    {
      "auxiliary_loss_clip": 0.01112296,
      "auxiliary_loss_mlp": 0.01031169,
      "balance_loss_clip": 1.04285717,
      "balance_loss_mlp": 1.02150691,
      "epoch": 0.41976793122106654,
      "flos": 24531333183360.0,
      "grad_norm": 2.3863812840644694,
      "language_loss": 0.76185155,
      "learning_rate": 2.6072077021405465e-06,
      "loss": 0.78328621,
      "num_input_tokens_seen": 75233020,
      "step": 3491,
      "time_per_iteration": 2.826784133911133
    },
    {
      "auxiliary_loss_clip": 0.01150974,
      "auxiliary_loss_mlp": 0.01029215,
      "balance_loss_clip": 1.04692221,
      "balance_loss_mlp": 1.02142477,
      "epoch": 0.41988817411170565,
      "flos": 21175301664000.0,
      "grad_norm": 1.7252455326851732,
      "language_loss": 0.69625974,
      "learning_rate": 2.6064654532002054e-06,
      "loss": 0.71806169,
      "num_input_tokens_seen": 75252030,
      "step": 3492,
      "time_per_iteration": 2.7395827770233154
    },
    {
      "auxiliary_loss_clip": 0.0118685,
      "auxiliary_loss_mlp": 0.01026123,
      "balance_loss_clip": 1.05379832,
      "balance_loss_mlp": 1.01853502,
      "epoch": 0.42000841700234476,
      "flos": 31649402626560.0,
      "grad_norm": 1.5116171081009089,
      "language_loss": 0.75760198,
      "learning_rate": 2.6057231122595375e-06,
      "loss": 0.77973169,
      "num_input_tokens_seen": 75273340,
      "step": 3493,
      "time_per_iteration": 2.7768354415893555
    },
    {
      "auxiliary_loss_clip": 0.01160533,
      "auxiliary_loss_mlp": 0.01034287,
      "balance_loss_clip": 1.04978406,
      "balance_loss_mlp": 1.0265379,
      "epoch": 0.4201286598929838,
      "flos": 21281525159040.0,
      "grad_norm": 1.6038393927783618,
      "language_loss": 0.73202062,
      "learning_rate": 2.604980679431154e-06,
      "loss": 0.75396883,
      "num_input_tokens_seen": 75291580,
      "step": 3494,
      "time_per_iteration": 2.668672800064087
    },
    {
      "auxiliary_loss_clip": 0.01175076,
      "auxiliary_loss_mlp": 0.01024212,
      "balance_loss_clip": 1.05017197,
      "balance_loss_mlp": 1.01603436,
      "epoch": 0.4202489027836229,
      "flos": 18546532813440.0,
      "grad_norm": 2.0588178169971685,
      "language_loss": 0.74831045,
      "learning_rate": 2.604238154827684e-06,
      "loss": 0.77030331,
      "num_input_tokens_seen": 75308205,
      "step": 3495,
      "time_per_iteration": 3.58237361907959
    },
    {
      "auxiliary_loss_clip": 0.01177106,
      "auxiliary_loss_mlp": 0.01031122,
      "balance_loss_clip": 1.05388117,
      "balance_loss_mlp": 1.02288449,
      "epoch": 0.42036914567426203,
      "flos": 19317643009920.0,
      "grad_norm": 2.4956957114922527,
      "language_loss": 0.7276969,
      "learning_rate": 2.6034955385617656e-06,
      "loss": 0.74977911,
      "num_input_tokens_seen": 75326535,
      "step": 3496,
      "time_per_iteration": 2.630383014678955
    },
    {
      "auxiliary_loss_clip": 0.01052933,
      "auxiliary_loss_mlp": 0.01000887,
      "balance_loss_clip": 1.01504242,
      "balance_loss_mlp": 0.99913424,
      "epoch": 0.4204893885649011,
      "flos": 67842942935040.0,
      "grad_norm": 0.7177584716073458,
      "language_loss": 0.61709106,
      "learning_rate": 2.6027528307460544e-06,
      "loss": 0.63762927,
      "num_input_tokens_seen": 75390540,
      "step": 3497,
      "time_per_iteration": 4.434598445892334
    },
    {
      "auxiliary_loss_clip": 0.01189314,
      "auxiliary_loss_mlp": 0.01028516,
      "balance_loss_clip": 1.05415344,
      "balance_loss_mlp": 1.0206244,
      "epoch": 0.4206096314555402,
      "flos": 21908777385600.0,
      "grad_norm": 1.7670282637300714,
      "language_loss": 0.86396283,
      "learning_rate": 2.602010031493217e-06,
      "loss": 0.88614118,
      "num_input_tokens_seen": 75408770,
      "step": 3498,
      "time_per_iteration": 2.7410974502563477
    },
    {
      "auxiliary_loss_clip": 0.01142297,
      "auxiliary_loss_mlp": 0.01021143,
      "balance_loss_clip": 1.04945958,
      "balance_loss_mlp": 1.01298916,
      "epoch": 0.42072987434617926,
      "flos": 29278185269760.0,
      "grad_norm": 1.8401753803207481,
      "language_loss": 0.87083703,
      "learning_rate": 2.6012671409159367e-06,
      "loss": 0.89247143,
      "num_input_tokens_seen": 75430105,
      "step": 3499,
      "time_per_iteration": 2.738534450531006
    },
    {
      "auxiliary_loss_clip": 0.01156722,
      "auxiliary_loss_mlp": 0.01027277,
      "balance_loss_clip": 1.05145383,
      "balance_loss_mlp": 1.01914716,
      "epoch": 0.42085011723681837,
      "flos": 27600726170880.0,
      "grad_norm": 2.0494679149966957,
      "language_loss": 0.81925535,
      "learning_rate": 2.6005241591269097e-06,
      "loss": 0.84109539,
      "num_input_tokens_seen": 75449475,
      "step": 3500,
      "time_per_iteration": 3.636561155319214
    },
    {
      "auxiliary_loss_clip": 0.01142542,
      "auxiliary_loss_mlp": 0.01027207,
      "balance_loss_clip": 1.05289173,
      "balance_loss_mlp": 1.0197866,
      "epoch": 0.4209703601274575,
      "flos": 27818632028160.0,
      "grad_norm": 1.8098474620214282,
      "language_loss": 0.79843616,
      "learning_rate": 2.5997810862388454e-06,
      "loss": 0.82013363,
      "num_input_tokens_seen": 75469315,
      "step": 3501,
      "time_per_iteration": 3.6834137439727783
    },
    {
      "auxiliary_loss_clip": 0.01157735,
      "auxiliary_loss_mlp": 0.01026645,
      "balance_loss_clip": 1.04861534,
      "balance_loss_mlp": 1.01840723,
      "epoch": 0.42109060301809653,
      "flos": 27525529048320.0,
      "grad_norm": 2.7986332192261187,
      "language_loss": 0.75737619,
      "learning_rate": 2.599037922364467e-06,
      "loss": 0.77922004,
      "num_input_tokens_seen": 75488215,
      "step": 3502,
      "time_per_iteration": 2.781593084335327
    },
    {
      "auxiliary_loss_clip": 0.01139763,
      "auxiliary_loss_mlp": 0.01030728,
      "balance_loss_clip": 1.04922843,
      "balance_loss_mlp": 1.02229357,
      "epoch": 0.42121084590873564,
      "flos": 29314275459840.0,
      "grad_norm": 2.0934604262163,
      "language_loss": 0.75530541,
      "learning_rate": 2.5982946676165112e-06,
      "loss": 0.77701032,
      "num_input_tokens_seen": 75507985,
      "step": 3503,
      "time_per_iteration": 2.7563836574554443
    },
    {
      "auxiliary_loss_clip": 0.01077297,
      "auxiliary_loss_mlp": 0.01003586,
      "balance_loss_clip": 1.04001045,
      "balance_loss_mlp": 1.00147617,
      "epoch": 0.42133108879937475,
      "flos": 67398835178880.0,
      "grad_norm": 0.7268753768007374,
      "language_loss": 0.57545578,
      "learning_rate": 2.5975513221077313e-06,
      "loss": 0.5962646,
      "num_input_tokens_seen": 75571955,
      "step": 3504,
      "time_per_iteration": 3.3643743991851807
    },
    {
      "auxiliary_loss_clip": 0.01150325,
      "auxiliary_loss_mlp": 0.01034391,
      "balance_loss_clip": 1.04816842,
      "balance_loss_mlp": 1.02583838,
      "epoch": 0.4214513316900138,
      "flos": 23106038538240.0,
      "grad_norm": 2.168303480937892,
      "language_loss": 0.88535792,
      "learning_rate": 2.5968078859508897e-06,
      "loss": 0.90720505,
      "num_input_tokens_seen": 75589155,
      "step": 3505,
      "time_per_iteration": 2.6907010078430176
    },
    {
      "auxiliary_loss_clip": 0.01174405,
      "auxiliary_loss_mlp": 0.01023612,
      "balance_loss_clip": 1.0533042,
      "balance_loss_mlp": 1.01576757,
      "epoch": 0.4215715745806529,
      "flos": 15336190857600.0,
      "grad_norm": 2.0701894595746007,
      "language_loss": 0.79835999,
      "learning_rate": 2.5960643592587673e-06,
      "loss": 0.8203401,
      "num_input_tokens_seen": 75606565,
      "step": 3506,
      "time_per_iteration": 2.6331839561462402
    },
    {
      "auxiliary_loss_clip": 0.01146599,
      "auxiliary_loss_mlp": 0.01028045,
      "balance_loss_clip": 1.04839158,
      "balance_loss_mlp": 1.02005839,
      "epoch": 0.42169181747129203,
      "flos": 22127257860480.0,
      "grad_norm": 1.701310473139929,
      "language_loss": 0.81319064,
      "learning_rate": 2.5953207421441553e-06,
      "loss": 0.83493704,
      "num_input_tokens_seen": 75625165,
      "step": 3507,
      "time_per_iteration": 2.7252910137176514
    },
    {
      "auxiliary_loss_clip": 0.01147524,
      "auxiliary_loss_mlp": 0.01027042,
      "balance_loss_clip": 1.05007088,
      "balance_loss_mlp": 1.01851845,
      "epoch": 0.4218120603619311,
      "flos": 22630724841600.0,
      "grad_norm": 2.1119677359501736,
      "language_loss": 0.75638521,
      "learning_rate": 2.5945770347198603e-06,
      "loss": 0.77813083,
      "num_input_tokens_seen": 75643320,
      "step": 3508,
      "time_per_iteration": 2.747039794921875
    },
    {
      "auxiliary_loss_clip": 0.01153387,
      "auxiliary_loss_mlp": 0.01026239,
      "balance_loss_clip": 1.04603457,
      "balance_loss_mlp": 1.01843107,
      "epoch": 0.4219323032525702,
      "flos": 19682818629120.0,
      "grad_norm": 1.7252196732033775,
      "language_loss": 0.8218528,
      "learning_rate": 2.593833237098701e-06,
      "loss": 0.84364903,
      "num_input_tokens_seen": 75660920,
      "step": 3509,
      "time_per_iteration": 2.6713287830352783
    },
    {
      "auxiliary_loss_clip": 0.01169511,
      "auxiliary_loss_mlp": 0.01033037,
      "balance_loss_clip": 1.04710472,
      "balance_loss_mlp": 1.02414417,
      "epoch": 0.4220525461432093,
      "flos": 30190747224960.0,
      "grad_norm": 2.03285972268686,
      "language_loss": 0.63008821,
      "learning_rate": 2.593089349393512e-06,
      "loss": 0.65211368,
      "num_input_tokens_seen": 75681410,
      "step": 3510,
      "time_per_iteration": 2.6857268810272217
    },
    {
      "auxiliary_loss_clip": 0.01171709,
      "auxiliary_loss_mlp": 0.01029901,
      "balance_loss_clip": 1.05200231,
      "balance_loss_mlp": 1.02182508,
      "epoch": 0.42217278903384836,
      "flos": 24315941278080.0,
      "grad_norm": 2.4187369202665825,
      "language_loss": 0.83614099,
      "learning_rate": 2.592345371717141e-06,
      "loss": 0.85815716,
      "num_input_tokens_seen": 75700940,
      "step": 3511,
      "time_per_iteration": 2.650291919708252
    },
    {
      "auxiliary_loss_clip": 0.01174054,
      "auxiliary_loss_mlp": 0.01034091,
      "balance_loss_clip": 1.05531311,
      "balance_loss_mlp": 1.02570152,
      "epoch": 0.42229303192448747,
      "flos": 17092474352640.0,
      "grad_norm": 2.315202723197227,
      "language_loss": 0.71931577,
      "learning_rate": 2.591601304182448e-06,
      "loss": 0.7413972,
      "num_input_tokens_seen": 75718910,
      "step": 3512,
      "time_per_iteration": 2.604853630065918
    },
    {
      "auxiliary_loss_clip": 0.01159532,
      "auxiliary_loss_mlp": 0.01025344,
      "balance_loss_clip": 1.05367434,
      "balance_loss_mlp": 1.01747024,
      "epoch": 0.4224132748151266,
      "flos": 22784530878720.0,
      "grad_norm": 2.172189887692531,
      "language_loss": 0.79566717,
      "learning_rate": 2.5908571469023067e-06,
      "loss": 0.81751591,
      "num_input_tokens_seen": 75738395,
      "step": 3513,
      "time_per_iteration": 2.700212001800537
    },
    {
      "auxiliary_loss_clip": 0.0118637,
      "auxiliary_loss_mlp": 0.01029794,
      "balance_loss_clip": 1.05340123,
      "balance_loss_mlp": 1.0218612,
      "epoch": 0.42253351770576564,
      "flos": 17819090576640.0,
      "grad_norm": 2.3710727306675223,
      "language_loss": 0.75804257,
      "learning_rate": 2.5901128999896067e-06,
      "loss": 0.78020418,
      "num_input_tokens_seen": 75753825,
      "step": 3514,
      "time_per_iteration": 2.605175495147705
    },
    {
      "auxiliary_loss_clip": 0.01171554,
      "auxiliary_loss_mlp": 0.01030213,
      "balance_loss_clip": 1.0517422,
      "balance_loss_mlp": 1.02228522,
      "epoch": 0.42265376059640475,
      "flos": 28512390286080.0,
      "grad_norm": 1.6271151910872301,
      "language_loss": 0.6834954,
      "learning_rate": 2.5893685635572487e-06,
      "loss": 0.70551306,
      "num_input_tokens_seen": 75774675,
      "step": 3515,
      "time_per_iteration": 2.740384101867676
    },
    {
      "auxiliary_loss_clip": 0.01154322,
      "auxiliary_loss_mlp": 0.01030359,
      "balance_loss_clip": 1.0503813,
      "balance_loss_mlp": 1.02193642,
      "epoch": 0.4227740034870438,
      "flos": 16253349753600.0,
      "grad_norm": 1.990422708851614,
      "language_loss": 0.69204509,
      "learning_rate": 2.5886241377181483e-06,
      "loss": 0.71389186,
      "num_input_tokens_seen": 75793545,
      "step": 3516,
      "time_per_iteration": 2.632050037384033
    },
    {
      "auxiliary_loss_clip": 0.01173856,
      "auxiliary_loss_mlp": 0.01029002,
      "balance_loss_clip": 1.051126,
      "balance_loss_mlp": 1.01945949,
      "epoch": 0.4228942463776829,
      "flos": 25295691623040.0,
      "grad_norm": 1.8878280754387362,
      "language_loss": 0.8141073,
      "learning_rate": 2.587879622585234e-06,
      "loss": 0.83613586,
      "num_input_tokens_seen": 75812145,
      "step": 3517,
      "time_per_iteration": 2.663898229598999
    },
    {
      "auxiliary_loss_clip": 0.01172611,
      "auxiliary_loss_mlp": 0.01033496,
      "balance_loss_clip": 1.05435061,
      "balance_loss_mlp": 1.02534807,
      "epoch": 0.423014489268322,
      "flos": 26395779507840.0,
      "grad_norm": 2.016497279838565,
      "language_loss": 0.7591657,
      "learning_rate": 2.5871350182714486e-06,
      "loss": 0.78122675,
      "num_input_tokens_seen": 75833025,
      "step": 3518,
      "time_per_iteration": 2.6465697288513184
    },
    {
      "auxiliary_loss_clip": 0.01184665,
      "auxiliary_loss_mlp": 0.01030853,
      "balance_loss_clip": 1.05198216,
      "balance_loss_mlp": 1.02252007,
      "epoch": 0.4231347321589611,
      "flos": 17274002711040.0,
      "grad_norm": 2.233988058457752,
      "language_loss": 0.80219162,
      "learning_rate": 2.586390324889748e-06,
      "loss": 0.82434678,
      "num_input_tokens_seen": 75848925,
      "step": 3519,
      "time_per_iteration": 2.605771064758301
    },
    {
      "auxiliary_loss_clip": 0.01171054,
      "auxiliary_loss_mlp": 0.01028994,
      "balance_loss_clip": 1.05175471,
      "balance_loss_mlp": 1.02066112,
      "epoch": 0.4232549750496002,
      "flos": 22999635475200.0,
      "grad_norm": 1.7517251440479846,
      "language_loss": 0.67360592,
      "learning_rate": 2.5856455425531003e-06,
      "loss": 0.69560641,
      "num_input_tokens_seen": 75870400,
      "step": 3520,
      "time_per_iteration": 2.687532424926758
    },
    {
      "auxiliary_loss_clip": 0.01172392,
      "auxiliary_loss_mlp": 0.01032566,
      "balance_loss_clip": 1.05458355,
      "balance_loss_mlp": 1.02413177,
      "epoch": 0.4233752179402393,
      "flos": 21248343970560.0,
      "grad_norm": 1.7734261134982823,
      "language_loss": 0.81168109,
      "learning_rate": 2.5849006713744902e-06,
      "loss": 0.8337307,
      "num_input_tokens_seen": 75889195,
      "step": 3521,
      "time_per_iteration": 3.628653049468994
    },
    {
      "auxiliary_loss_clip": 0.01157862,
      "auxiliary_loss_mlp": 0.01029396,
      "balance_loss_clip": 1.0495131,
      "balance_loss_mlp": 1.02053261,
      "epoch": 0.42349546083087836,
      "flos": 20704297599360.0,
      "grad_norm": 70.52847032222567,
      "language_loss": 0.72594148,
      "learning_rate": 2.5841557114669135e-06,
      "loss": 0.74781406,
      "num_input_tokens_seen": 75906055,
      "step": 3522,
      "time_per_iteration": 2.6959068775177
    },
    {
      "auxiliary_loss_clip": 0.01189467,
      "auxiliary_loss_mlp": 0.01027906,
      "balance_loss_clip": 1.05168617,
      "balance_loss_mlp": 1.01874435,
      "epoch": 0.42361570372151747,
      "flos": 18585065128320.0,
      "grad_norm": 2.5780370557755052,
      "language_loss": 0.67121387,
      "learning_rate": 2.58341066294338e-06,
      "loss": 0.69338757,
      "num_input_tokens_seen": 75922720,
      "step": 3523,
      "time_per_iteration": 3.5693626403808594
    },
    {
      "auxiliary_loss_clip": 0.0113286,
      "auxiliary_loss_mlp": 0.00763236,
      "balance_loss_clip": 1.04697978,
      "balance_loss_mlp": 1.00042415,
      "epoch": 0.4237359466121566,
      "flos": 20959478795520.0,
      "grad_norm": 2.2172959979199196,
      "language_loss": 0.85188544,
      "learning_rate": 2.5826655259169124e-06,
      "loss": 0.87084639,
      "num_input_tokens_seen": 75941375,
      "step": 3524,
      "time_per_iteration": 2.7979917526245117
    },
    {
      "auxiliary_loss_clip": 0.01189337,
      "auxiliary_loss_mlp": 0.01029614,
      "balance_loss_clip": 1.05458462,
      "balance_loss_mlp": 1.02183008,
      "epoch": 0.42385618950279563,
      "flos": 18038181582720.0,
      "grad_norm": 1.891978018177093,
      "language_loss": 0.90713,
      "learning_rate": 2.5819203005005475e-06,
      "loss": 0.9293195,
      "num_input_tokens_seen": 75958710,
      "step": 3525,
      "time_per_iteration": 2.6336874961853027
    },
    {
      "auxiliary_loss_clip": 0.01153773,
      "auxiliary_loss_mlp": 0.01034421,
      "balance_loss_clip": 1.0497334,
      "balance_loss_mlp": 1.02537251,
      "epoch": 0.42397643239343474,
      "flos": 23769129559680.0,
      "grad_norm": 2.008509497961551,
      "language_loss": 0.78706998,
      "learning_rate": 2.581174986807336e-06,
      "loss": 0.80895191,
      "num_input_tokens_seen": 75978945,
      "step": 3526,
      "time_per_iteration": 3.6497247219085693
    },
    {
      "auxiliary_loss_clip": 0.01163513,
      "auxiliary_loss_mlp": 0.00763516,
      "balance_loss_clip": 1.04884744,
      "balance_loss_mlp": 1.00043666,
      "epoch": 0.42409667528407385,
      "flos": 16545088016640.0,
      "grad_norm": 2.6213761965992424,
      "language_loss": 0.91247493,
      "learning_rate": 2.580429584950341e-06,
      "loss": 0.93174517,
      "num_input_tokens_seen": 75994695,
      "step": 3527,
      "time_per_iteration": 3.5901148319244385
    },
    {
      "auxiliary_loss_clip": 0.01150493,
      "auxiliary_loss_mlp": 0.0103147,
      "balance_loss_clip": 1.04871988,
      "balance_loss_mlp": 1.02249932,
      "epoch": 0.4242169181747129,
      "flos": 16034186920320.0,
      "grad_norm": 2.1535403463354386,
      "language_loss": 0.66553438,
      "learning_rate": 2.5796840950426397e-06,
      "loss": 0.68735397,
      "num_input_tokens_seen": 76011780,
      "step": 3528,
      "time_per_iteration": 2.6876461505889893
    },
    {
      "auxiliary_loss_clip": 0.01166296,
      "auxiliary_loss_mlp": 0.01034893,
      "balance_loss_clip": 1.05135202,
      "balance_loss_mlp": 1.02561307,
      "epoch": 0.424337161065352,
      "flos": 20084012611200.0,
      "grad_norm": 4.935629946808246,
      "language_loss": 0.65883648,
      "learning_rate": 2.578938517197322e-06,
      "loss": 0.68084842,
      "num_input_tokens_seen": 76029875,
      "step": 3529,
      "time_per_iteration": 2.6843271255493164
    },
    {
      "auxiliary_loss_clip": 0.01152535,
      "auxiliary_loss_mlp": 0.01031701,
      "balance_loss_clip": 1.04934788,
      "balance_loss_mlp": 1.02263546,
      "epoch": 0.4244574039559911,
      "flos": 23878369797120.0,
      "grad_norm": 4.447605780148659,
      "language_loss": 0.62831652,
      "learning_rate": 2.5781928515274916e-06,
      "loss": 0.65015888,
      "num_input_tokens_seen": 76048595,
      "step": 3530,
      "time_per_iteration": 2.7046899795532227
    },
    {
      "auxiliary_loss_clip": 0.01176804,
      "auxiliary_loss_mlp": 0.01027614,
      "balance_loss_clip": 1.05309367,
      "balance_loss_mlp": 1.01909649,
      "epoch": 0.4245776468466302,
      "flos": 17565920542080.0,
      "grad_norm": 1.8152004737302525,
      "language_loss": 0.67537308,
      "learning_rate": 2.577447098146265e-06,
      "loss": 0.69741726,
      "num_input_tokens_seen": 76065770,
      "step": 3531,
      "time_per_iteration": 2.7420899868011475
    },
    {
      "auxiliary_loss_clip": 0.01146901,
      "auxiliary_loss_mlp": 0.0102884,
      "balance_loss_clip": 1.04956615,
      "balance_loss_mlp": 1.02025723,
      "epoch": 0.4246978897372693,
      "flos": 27776256958080.0,
      "grad_norm": 1.660756420769618,
      "language_loss": 0.78907013,
      "learning_rate": 2.5767012571667724e-06,
      "loss": 0.81082755,
      "num_input_tokens_seen": 76085250,
      "step": 3532,
      "time_per_iteration": 2.7533302307128906
    },
    {
      "auxiliary_loss_clip": 0.01174867,
      "auxiliary_loss_mlp": 0.01032389,
      "balance_loss_clip": 1.05071115,
      "balance_loss_mlp": 1.02321029,
      "epoch": 0.42481813262790835,
      "flos": 15596615439360.0,
      "grad_norm": 1.9610806697253336,
      "language_loss": 0.68545687,
      "learning_rate": 2.5759553287021587e-06,
      "loss": 0.70752943,
      "num_input_tokens_seen": 76103580,
      "step": 3533,
      "time_per_iteration": 2.601832389831543
    },
    {
      "auxiliary_loss_clip": 0.01157346,
      "auxiliary_loss_mlp": 0.01035406,
      "balance_loss_clip": 1.0500288,
      "balance_loss_mlp": 1.02696574,
      "epoch": 0.42493837551854746,
      "flos": 23951088881280.0,
      "grad_norm": 1.9468276654194652,
      "language_loss": 0.77546251,
      "learning_rate": 2.5752093128655786e-06,
      "loss": 0.79739004,
      "num_input_tokens_seen": 76121825,
      "step": 3534,
      "time_per_iteration": 2.7032108306884766
    },
    {
      "auxiliary_loss_clip": 0.01148012,
      "auxiliary_loss_mlp": 0.01028197,
      "balance_loss_clip": 1.04532337,
      "balance_loss_mlp": 1.01973939,
      "epoch": 0.4250586184091866,
      "flos": 20813466009600.0,
      "grad_norm": 1.9666882887240578,
      "language_loss": 0.73791575,
      "learning_rate": 2.574463209770204e-06,
      "loss": 0.75967789,
      "num_input_tokens_seen": 76141140,
      "step": 3535,
      "time_per_iteration": 2.657381296157837
    },
    {
      "auxiliary_loss_clip": 0.01140672,
      "auxiliary_loss_mlp": 0.01027391,
      "balance_loss_clip": 1.04491913,
      "balance_loss_mlp": 1.01868856,
      "epoch": 0.42517886129982563,
      "flos": 30371018607360.0,
      "grad_norm": 1.6636445369566226,
      "language_loss": 0.79535794,
      "learning_rate": 2.5737170195292165e-06,
      "loss": 0.81703854,
      "num_input_tokens_seen": 76164475,
      "step": 3536,
      "time_per_iteration": 2.8733224868774414
    },
    {
      "auxiliary_loss_clip": 0.01141179,
      "auxiliary_loss_mlp": 0.0103291,
      "balance_loss_clip": 1.04515755,
      "balance_loss_mlp": 1.02428484,
      "epoch": 0.42529910419046474,
      "flos": 20080636732800.0,
      "grad_norm": 1.6810804670594863,
      "language_loss": 0.77865154,
      "learning_rate": 2.572970742255814e-06,
      "loss": 0.80039239,
      "num_input_tokens_seen": 76182965,
      "step": 3537,
      "time_per_iteration": 2.678542137145996
    },
    {
      "auxiliary_loss_clip": 0.01171809,
      "auxiliary_loss_mlp": 0.01028587,
      "balance_loss_clip": 1.0527215,
      "balance_loss_mlp": 1.02006996,
      "epoch": 0.42541934708110385,
      "flos": 22632448694400.0,
      "grad_norm": 1.837817260773353,
      "language_loss": 0.81435907,
      "learning_rate": 2.5722243780632046e-06,
      "loss": 0.83636308,
      "num_input_tokens_seen": 76201230,
      "step": 3538,
      "time_per_iteration": 2.7220025062561035
    },
    {
      "auxiliary_loss_clip": 0.01047706,
      "auxiliary_loss_mlp": 0.01006834,
      "balance_loss_clip": 1.01787877,
      "balance_loss_mlp": 1.00490308,
      "epoch": 0.4255395899717429,
      "flos": 66200676186240.0,
      "grad_norm": 0.7773195658321704,
      "language_loss": 0.60387391,
      "learning_rate": 2.5714779270646125e-06,
      "loss": 0.62441933,
      "num_input_tokens_seen": 76262000,
      "step": 3539,
      "time_per_iteration": 3.227846145629883
    },
    {
      "auxiliary_loss_clip": 0.01159643,
      "auxiliary_loss_mlp": 0.00764367,
      "balance_loss_clip": 1.05209899,
      "balance_loss_mlp": 1.00037956,
      "epoch": 0.425659832862382,
      "flos": 17931814433280.0,
      "grad_norm": 2.466787777941296,
      "language_loss": 0.78031385,
      "learning_rate": 2.5707313893732735e-06,
      "loss": 0.79955399,
      "num_input_tokens_seen": 76280540,
      "step": 3540,
      "time_per_iteration": 2.7084479331970215
    },
    {
      "auxiliary_loss_clip": 0.0109356,
      "auxiliary_loss_mlp": 0.01027867,
      "balance_loss_clip": 1.03899407,
      "balance_loss_mlp": 1.01888418,
      "epoch": 0.4257800757530211,
      "flos": 24022550989440.0,
      "grad_norm": 1.7943008791281314,
      "language_loss": 0.76943958,
      "learning_rate": 2.5699847651024364e-06,
      "loss": 0.79065382,
      "num_input_tokens_seen": 76301180,
      "step": 3541,
      "time_per_iteration": 2.831658363342285
    },
    {
      "auxiliary_loss_clip": 0.0117091,
      "auxiliary_loss_mlp": 0.01033001,
      "balance_loss_clip": 1.05277729,
      "balance_loss_mlp": 1.02466202,
      "epoch": 0.4259003186436602,
      "flos": 23696015425920.0,
      "grad_norm": 2.6743016238093817,
      "language_loss": 0.76482153,
      "learning_rate": 2.5692380543653627e-06,
      "loss": 0.78686064,
      "num_input_tokens_seen": 76319335,
      "step": 3542,
      "time_per_iteration": 2.6723670959472656
    },
    {
      "auxiliary_loss_clip": 0.01178004,
      "auxiliary_loss_mlp": 0.00763787,
      "balance_loss_clip": 1.05272245,
      "balance_loss_mlp": 1.00039136,
      "epoch": 0.4260205615342993,
      "flos": 15259772672640.0,
      "grad_norm": 2.0710600411036673,
      "language_loss": 0.70150197,
      "learning_rate": 2.5684912572753293e-06,
      "loss": 0.72091991,
      "num_input_tokens_seen": 76335010,
      "step": 3543,
      "time_per_iteration": 2.6652090549468994
    },
    {
      "auxiliary_loss_clip": 0.01184597,
      "auxiliary_loss_mlp": 0.01030376,
      "balance_loss_clip": 1.0529393,
      "balance_loss_mlp": 1.02281761,
      "epoch": 0.4261408044249384,
      "flos": 30665306736000.0,
      "grad_norm": 1.9845208613039869,
      "language_loss": 0.8450948,
      "learning_rate": 2.5677443739456245e-06,
      "loss": 0.86724454,
      "num_input_tokens_seen": 76356670,
      "step": 3544,
      "time_per_iteration": 2.6831905841827393
    },
    {
      "auxiliary_loss_clip": 0.01162873,
      "auxiliary_loss_mlp": 0.01032205,
      "balance_loss_clip": 1.05262196,
      "balance_loss_mlp": 1.02387273,
      "epoch": 0.42626104731557746,
      "flos": 23257905240960.0,
      "grad_norm": 2.2970676957567013,
      "language_loss": 0.7964263,
      "learning_rate": 2.5669974044895495e-06,
      "loss": 0.81837702,
      "num_input_tokens_seen": 76373065,
      "step": 3545,
      "time_per_iteration": 2.720111608505249
    },
    {
      "auxiliary_loss_clip": 0.01151173,
      "auxiliary_loss_mlp": 0.0102622,
      "balance_loss_clip": 1.04832041,
      "balance_loss_mlp": 1.01762509,
      "epoch": 0.42638129020621657,
      "flos": 25884770670720.0,
      "grad_norm": 1.6333459186891055,
      "language_loss": 0.79005802,
      "learning_rate": 2.5662503490204187e-06,
      "loss": 0.81183195,
      "num_input_tokens_seen": 76393230,
      "step": 3546,
      "time_per_iteration": 2.720024585723877
    },
    {
      "auxiliary_loss_clip": 0.01155036,
      "auxiliary_loss_mlp": 0.01038337,
      "balance_loss_clip": 1.04686451,
      "balance_loss_mlp": 1.02929497,
      "epoch": 0.4265015330968556,
      "flos": 26502362138880.0,
      "grad_norm": 1.9890829846609375,
      "language_loss": 0.76438498,
      "learning_rate": 2.5655032076515603e-06,
      "loss": 0.78631872,
      "num_input_tokens_seen": 76412555,
      "step": 3547,
      "time_per_iteration": 3.6662683486938477
    },
    {
      "auxiliary_loss_clip": 0.01160781,
      "auxiliary_loss_mlp": 0.01028355,
      "balance_loss_clip": 1.0507586,
      "balance_loss_mlp": 1.02006412,
      "epoch": 0.42662177598749473,
      "flos": 24389522288640.0,
      "grad_norm": 5.143813064632022,
      "language_loss": 0.82132494,
      "learning_rate": 2.5647559804963155e-06,
      "loss": 0.8432163,
      "num_input_tokens_seen": 76432485,
      "step": 3548,
      "time_per_iteration": 2.7028098106384277
    },
    {
      "auxiliary_loss_clip": 0.01137883,
      "auxiliary_loss_mlp": 0.01027937,
      "balance_loss_clip": 1.0490104,
      "balance_loss_mlp": 1.01945519,
      "epoch": 0.42674201887813384,
      "flos": 23148629089920.0,
      "grad_norm": 1.9756868275064754,
      "language_loss": 0.78851688,
      "learning_rate": 2.5640086676680364e-06,
      "loss": 0.81017506,
      "num_input_tokens_seen": 76453980,
      "step": 3549,
      "time_per_iteration": 3.7275803089141846
    },
    {
      "auxiliary_loss_clip": 0.0117586,
      "auxiliary_loss_mlp": 0.01030967,
      "balance_loss_clip": 1.0529722,
      "balance_loss_mlp": 1.02266431,
      "epoch": 0.4268622617687729,
      "flos": 21689614552320.0,
      "grad_norm": 2.7465713268694962,
      "language_loss": 0.80935514,
      "learning_rate": 2.5632612692800923e-06,
      "loss": 0.8314234,
      "num_input_tokens_seen": 76473045,
      "step": 3550,
      "time_per_iteration": 2.6946353912353516
    },
    {
      "auxiliary_loss_clip": 0.01147176,
      "auxiliary_loss_mlp": 0.01032943,
      "balance_loss_clip": 1.04845333,
      "balance_loss_mlp": 1.02335227,
      "epoch": 0.426982504659412,
      "flos": 23440151871360.0,
      "grad_norm": 1.9843526577873574,
      "language_loss": 0.7560339,
      "learning_rate": 2.5625137854458603e-06,
      "loss": 0.77783507,
      "num_input_tokens_seen": 76492060,
      "step": 3551,
      "time_per_iteration": 3.680642604827881
    },
    {
      "auxiliary_loss_clip": 0.01162801,
      "auxiliary_loss_mlp": 0.01032481,
      "balance_loss_clip": 1.04974842,
      "balance_loss_mlp": 1.02463675,
      "epoch": 0.4271027475500511,
      "flos": 18916556768640.0,
      "grad_norm": 1.9859393310219438,
      "language_loss": 0.80291712,
      "learning_rate": 2.561766216278735e-06,
      "loss": 0.82486993,
      "num_input_tokens_seen": 76509655,
      "step": 3552,
      "time_per_iteration": 2.732646942138672
    },
    {
      "auxiliary_loss_clip": 0.01131413,
      "auxiliary_loss_mlp": 0.01031998,
      "balance_loss_clip": 1.048612,
      "balance_loss_mlp": 1.02365971,
      "epoch": 0.4272229904406902,
      "flos": 26870554500480.0,
      "grad_norm": 2.0006516303086372,
      "language_loss": 0.81056547,
      "learning_rate": 2.561018561892121e-06,
      "loss": 0.83219957,
      "num_input_tokens_seen": 76528795,
      "step": 3553,
      "time_per_iteration": 3.69384503364563
    },
    {
      "auxiliary_loss_clip": 0.01155289,
      "auxiliary_loss_mlp": 0.01027849,
      "balance_loss_clip": 1.04602098,
      "balance_loss_mlp": 1.01908755,
      "epoch": 0.4273432333313293,
      "flos": 23951376190080.0,
      "grad_norm": 2.435051292531486,
      "language_loss": 0.76852381,
      "learning_rate": 2.5602708223994363e-06,
      "loss": 0.79035521,
      "num_input_tokens_seen": 76550660,
      "step": 3554,
      "time_per_iteration": 2.715123414993286
    },
    {
      "auxiliary_loss_clip": 0.01143261,
      "auxiliary_loss_mlp": 0.01030623,
      "balance_loss_clip": 1.04293823,
      "balance_loss_mlp": 1.02245116,
      "epoch": 0.4274634762219684,
      "flos": 29570354496000.0,
      "grad_norm": 2.375513861930382,
      "language_loss": 0.67875224,
      "learning_rate": 2.559522997914115e-06,
      "loss": 0.70049107,
      "num_input_tokens_seen": 76570240,
      "step": 3555,
      "time_per_iteration": 2.778697967529297
    },
    {
      "auxiliary_loss_clip": 0.01186809,
      "auxiliary_loss_mlp": 0.0103172,
      "balance_loss_clip": 1.05515075,
      "balance_loss_mlp": 1.02392375,
      "epoch": 0.42758371911260745,
      "flos": 21434146047360.0,
      "grad_norm": 1.9781009858662122,
      "language_loss": 0.84300494,
      "learning_rate": 2.558775088549599e-06,
      "loss": 0.86519027,
      "num_input_tokens_seen": 76589820,
      "step": 3556,
      "time_per_iteration": 2.6392624378204346
    },
    {
      "auxiliary_loss_clip": 0.01181319,
      "auxiliary_loss_mlp": 0.01029057,
      "balance_loss_clip": 1.05251098,
      "balance_loss_mlp": 1.01981831,
      "epoch": 0.42770396200324656,
      "flos": 14752822072320.0,
      "grad_norm": 2.2405724634686264,
      "language_loss": 0.66221523,
      "learning_rate": 2.5580270944193467e-06,
      "loss": 0.68431896,
      "num_input_tokens_seen": 76606640,
      "step": 3557,
      "time_per_iteration": 2.6725265979766846
    },
    {
      "auxiliary_loss_clip": 0.01086473,
      "auxiliary_loss_mlp": 0.01004789,
      "balance_loss_clip": 1.0178144,
      "balance_loss_mlp": 1.00288188,
      "epoch": 0.4278242048938857,
      "flos": 70654712601600.0,
      "grad_norm": 0.7433794170768325,
      "language_loss": 0.55488724,
      "learning_rate": 2.557279015636827e-06,
      "loss": 0.57579988,
      "num_input_tokens_seen": 76667050,
      "step": 3558,
      "time_per_iteration": 3.233607530593872
    },
    {
      "auxiliary_loss_clip": 0.01071592,
      "auxiliary_loss_mlp": 0.01002804,
      "balance_loss_clip": 1.01730716,
      "balance_loss_mlp": 1.00106359,
      "epoch": 0.42794444778452473,
      "flos": 69366165033600.0,
      "grad_norm": 0.8209970267353723,
      "language_loss": 0.61218166,
      "learning_rate": 2.5565308523155245e-06,
      "loss": 0.63292563,
      "num_input_tokens_seen": 76726650,
      "step": 3559,
      "time_per_iteration": 3.1589066982269287
    },
    {
      "auxiliary_loss_clip": 0.01126032,
      "auxiliary_loss_mlp": 0.01037976,
      "balance_loss_clip": 1.04764533,
      "balance_loss_mlp": 1.02968502,
      "epoch": 0.42806469067516384,
      "flos": 18215328481920.0,
      "grad_norm": 2.378983481962055,
      "language_loss": 0.81835186,
      "learning_rate": 2.5557826045689336e-06,
      "loss": 0.83999193,
      "num_input_tokens_seen": 76742890,
      "step": 3560,
      "time_per_iteration": 2.764399290084839
    },
    {
      "auxiliary_loss_clip": 0.01065005,
      "auxiliary_loss_mlp": 0.01002882,
      "balance_loss_clip": 1.02848685,
      "balance_loss_mlp": 1.00091541,
      "epoch": 0.4281849335658029,
      "flos": 54535814432640.0,
      "grad_norm": 0.8267305275907741,
      "language_loss": 0.58783567,
      "learning_rate": 2.5550342725105643e-06,
      "loss": 0.60851455,
      "num_input_tokens_seen": 76801055,
      "step": 3561,
      "time_per_iteration": 3.387869358062744
    },
    {
      "auxiliary_loss_clip": 0.01175699,
      "auxiliary_loss_mlp": 0.01032895,
      "balance_loss_clip": 1.05599821,
      "balance_loss_mlp": 1.02392459,
      "epoch": 0.428305176456442,
      "flos": 17274828723840.0,
      "grad_norm": 1.9924327099530497,
      "language_loss": 0.81119561,
      "learning_rate": 2.554285856253937e-06,
      "loss": 0.83328152,
      "num_input_tokens_seen": 76819890,
      "step": 3562,
      "time_per_iteration": 2.8317251205444336
    },
    {
      "auxiliary_loss_clip": 0.01156939,
      "auxiliary_loss_mlp": 0.01038118,
      "balance_loss_clip": 1.05233812,
      "balance_loss_mlp": 1.02985692,
      "epoch": 0.4284254193470811,
      "flos": 26359509749760.0,
      "grad_norm": 1.6545504500514931,
      "language_loss": 0.77698159,
      "learning_rate": 2.5535373559125855e-06,
      "loss": 0.79893214,
      "num_input_tokens_seen": 76840255,
      "step": 3563,
      "time_per_iteration": 2.7970194816589355
    },
    {
      "auxiliary_loss_clip": 0.01098715,
      "auxiliary_loss_mlp": 0.01034425,
      "balance_loss_clip": 1.04153204,
      "balance_loss_mlp": 1.02570808,
      "epoch": 0.42854566223772017,
      "flos": 29714248379520.0,
      "grad_norm": 1.5468399878867374,
      "language_loss": 0.82216614,
      "learning_rate": 2.552788771600057e-06,
      "loss": 0.84349751,
      "num_input_tokens_seen": 76860565,
      "step": 3564,
      "time_per_iteration": 2.96913743019104
    },
    {
      "auxiliary_loss_clip": 0.01149718,
      "auxiliary_loss_mlp": 0.0103536,
      "balance_loss_clip": 1.05157089,
      "balance_loss_mlp": 1.02677655,
      "epoch": 0.4286659051283593,
      "flos": 22018161277440.0,
      "grad_norm": 1.7453123792323384,
      "language_loss": 0.82063746,
      "learning_rate": 2.5520401034299118e-06,
      "loss": 0.84248823,
      "num_input_tokens_seen": 76878325,
      "step": 3565,
      "time_per_iteration": 2.7962448596954346
    },
    {
      "auxiliary_loss_clip": 0.01175334,
      "auxiliary_loss_mlp": 0.01031061,
      "balance_loss_clip": 1.05113995,
      "balance_loss_mlp": 1.02116644,
      "epoch": 0.4287861480189984,
      "flos": 13334422838400.0,
      "grad_norm": 1.9092666056164034,
      "language_loss": 0.87961483,
      "learning_rate": 2.551291351515722e-06,
      "loss": 0.90167874,
      "num_input_tokens_seen": 76895340,
      "step": 3566,
      "time_per_iteration": 2.6054606437683105
    },
    {
      "auxiliary_loss_clip": 0.01140629,
      "auxiliary_loss_mlp": 0.00763838,
      "balance_loss_clip": 1.04466927,
      "balance_loss_mlp": 1.00044119,
      "epoch": 0.42890639090963745,
      "flos": 26651535321600.0,
      "grad_norm": 1.7577367914418303,
      "language_loss": 0.86000705,
      "learning_rate": 2.5505425159710726e-06,
      "loss": 0.87905174,
      "num_input_tokens_seen": 76915150,
      "step": 3567,
      "time_per_iteration": 2.7730746269226074
    },
    {
      "auxiliary_loss_clip": 0.01166721,
      "auxiliary_loss_mlp": 0.00764057,
      "balance_loss_clip": 1.05074382,
      "balance_loss_mlp": 1.00041628,
      "epoch": 0.42902663380027656,
      "flos": 24055768091520.0,
      "grad_norm": 1.7881097299079334,
      "language_loss": 0.83138978,
      "learning_rate": 2.549793596909561e-06,
      "loss": 0.85069752,
      "num_input_tokens_seen": 76933770,
      "step": 3568,
      "time_per_iteration": 2.7250311374664307
    },
    {
      "auxiliary_loss_clip": 0.01155435,
      "auxiliary_loss_mlp": 0.01026673,
      "balance_loss_clip": 1.04982626,
      "balance_loss_mlp": 1.01844203,
      "epoch": 0.42914687669091567,
      "flos": 15632561975040.0,
      "grad_norm": 2.1716195801292075,
      "language_loss": 0.65919113,
      "learning_rate": 2.5490445944447976e-06,
      "loss": 0.68101227,
      "num_input_tokens_seen": 76952265,
      "step": 3569,
      "time_per_iteration": 2.6952881813049316
    },
    {
      "auxiliary_loss_clip": 0.01175845,
      "auxiliary_loss_mlp": 0.01032663,
      "balance_loss_clip": 1.05286682,
      "balance_loss_mlp": 1.02402616,
      "epoch": 0.4292671195815547,
      "flos": 31467802440960.0,
      "grad_norm": 2.0790128373576175,
      "language_loss": 0.6547606,
      "learning_rate": 2.548295508690406e-06,
      "loss": 0.67684567,
      "num_input_tokens_seen": 76973560,
      "step": 3570,
      "time_per_iteration": 2.703143358230591
    },
    {
      "auxiliary_loss_clip": 0.01176208,
      "auxiliary_loss_mlp": 0.01033841,
      "balance_loss_clip": 1.05001926,
      "balance_loss_mlp": 1.02520967,
      "epoch": 0.42938736247219383,
      "flos": 30257756046720.0,
      "grad_norm": 2.015925765642274,
      "language_loss": 0.76393878,
      "learning_rate": 2.5475463397600217e-06,
      "loss": 0.78603923,
      "num_input_tokens_seen": 76993640,
      "step": 3571,
      "time_per_iteration": 2.7500104904174805
    },
    {
      "auxiliary_loss_clip": 0.01190487,
      "auxiliary_loss_mlp": 0.01031278,
      "balance_loss_clip": 1.05398476,
      "balance_loss_mlp": 1.02262902,
      "epoch": 0.42950760536283294,
      "flos": 29349683291520.0,
      "grad_norm": 2.154189343392535,
      "language_loss": 0.77808809,
      "learning_rate": 2.546797087767293e-06,
      "loss": 0.80030572,
      "num_input_tokens_seen": 77013765,
      "step": 3572,
      "time_per_iteration": 2.734323263168335
    },
    {
      "auxiliary_loss_clip": 0.01128853,
      "auxiliary_loss_mlp": 0.01030406,
      "balance_loss_clip": 1.04753816,
      "balance_loss_mlp": 1.0218643,
      "epoch": 0.429627848253472,
      "flos": 26869943969280.0,
      "grad_norm": 1.668514765294981,
      "language_loss": 0.87296593,
      "learning_rate": 2.546047752825881e-06,
      "loss": 0.89455849,
      "num_input_tokens_seen": 77034370,
      "step": 3573,
      "time_per_iteration": 3.6749446392059326
    },
    {
      "auxiliary_loss_clip": 0.01136434,
      "auxiliary_loss_mlp": 0.01023452,
      "balance_loss_clip": 1.04759145,
      "balance_loss_mlp": 1.01478004,
      "epoch": 0.4297480911441111,
      "flos": 13881270470400.0,
      "grad_norm": 2.070212120025745,
      "language_loss": 0.93612134,
      "learning_rate": 2.5452983350494595e-06,
      "loss": 0.95772016,
      "num_input_tokens_seen": 77049925,
      "step": 3574,
      "time_per_iteration": 2.789116144180298
    },
    {
      "auxiliary_loss_clip": 0.01178006,
      "auxiliary_loss_mlp": 0.00763498,
      "balance_loss_clip": 1.05365205,
      "balance_loss_mlp": 1.00045145,
      "epoch": 0.4298683340347502,
      "flos": 20741141975040.0,
      "grad_norm": 1.983582721311658,
      "language_loss": 0.65405214,
      "learning_rate": 2.544548834551713e-06,
      "loss": 0.67346716,
      "num_input_tokens_seen": 77068930,
      "step": 3575,
      "time_per_iteration": 3.74849009513855
    },
    {
      "auxiliary_loss_clip": 0.01142188,
      "auxiliary_loss_mlp": 0.00763972,
      "balance_loss_clip": 1.04752111,
      "balance_loss_mlp": 1.00041449,
      "epoch": 0.4299885769253893,
      "flos": 20882126856960.0,
      "grad_norm": 2.8950959525872295,
      "language_loss": 0.9490723,
      "learning_rate": 2.5437992514463424e-06,
      "loss": 0.96813393,
      "num_input_tokens_seen": 77082255,
      "step": 3576,
      "time_per_iteration": 2.699000120162964
    },
    {
      "auxiliary_loss_clip": 0.01173379,
      "auxiliary_loss_mlp": 0.01032095,
      "balance_loss_clip": 1.05099189,
      "balance_loss_mlp": 1.02296996,
      "epoch": 0.4301088198160284,
      "flos": 25484618183040.0,
      "grad_norm": 1.6838051996476395,
      "language_loss": 0.87912571,
      "learning_rate": 2.5430495858470565e-06,
      "loss": 0.90118039,
      "num_input_tokens_seen": 77101725,
      "step": 3577,
      "time_per_iteration": 2.715313673019409
    },
    {
      "auxiliary_loss_clip": 0.01171226,
      "auxiliary_loss_mlp": 0.0102945,
      "balance_loss_clip": 1.05102205,
      "balance_loss_mlp": 1.02052736,
      "epoch": 0.43022906270666744,
      "flos": 18259427404800.0,
      "grad_norm": 2.305336896513299,
      "language_loss": 0.77498603,
      "learning_rate": 2.54229983786758e-06,
      "loss": 0.79699278,
      "num_input_tokens_seen": 77119670,
      "step": 3578,
      "time_per_iteration": 4.491747856140137
    },
    {
      "auxiliary_loss_clip": 0.01162518,
      "auxiliary_loss_mlp": 0.01032136,
      "balance_loss_clip": 1.05062532,
      "balance_loss_mlp": 1.02267647,
      "epoch": 0.43034930559730655,
      "flos": 23399536567680.0,
      "grad_norm": 1.8193216065438784,
      "language_loss": 0.8542459,
      "learning_rate": 2.541550007621651e-06,
      "loss": 0.87619245,
      "num_input_tokens_seen": 77138160,
      "step": 3579,
      "time_per_iteration": 2.81290340423584
    },
    {
      "auxiliary_loss_clip": 0.01176115,
      "auxiliary_loss_mlp": 0.01027664,
      "balance_loss_clip": 1.05624771,
      "balance_loss_mlp": 1.01945662,
      "epoch": 0.43046954848794566,
      "flos": 28184382264960.0,
      "grad_norm": 2.287982713966207,
      "language_loss": 0.80618942,
      "learning_rate": 2.5408000952230156e-06,
      "loss": 0.82822722,
      "num_input_tokens_seen": 77156950,
      "step": 3580,
      "time_per_iteration": 2.6883723735809326
    },
    {
      "auxiliary_loss_clip": 0.01154805,
      "auxiliary_loss_mlp": 0.01030235,
      "balance_loss_clip": 1.04811537,
      "balance_loss_mlp": 1.02125227,
      "epoch": 0.4305897913785847,
      "flos": 28580476515840.0,
      "grad_norm": 1.9742496237279235,
      "language_loss": 0.90624464,
      "learning_rate": 2.5400501007854357e-06,
      "loss": 0.92809504,
      "num_input_tokens_seen": 77176395,
      "step": 3581,
      "time_per_iteration": 2.738523006439209
    },
    {
      "auxiliary_loss_clip": 0.01126854,
      "auxiliary_loss_mlp": 0.01025217,
      "balance_loss_clip": 1.04375935,
      "balance_loss_mlp": 1.01667547,
      "epoch": 0.43071003426922383,
      "flos": 20448721353600.0,
      "grad_norm": 2.2363266237033015,
      "language_loss": 0.75403535,
      "learning_rate": 2.539300024422685e-06,
      "loss": 0.77555609,
      "num_input_tokens_seen": 77194340,
      "step": 3582,
      "time_per_iteration": 2.789151668548584
    },
    {
      "auxiliary_loss_clip": 0.01052933,
      "auxiliary_loss_mlp": 0.0100332,
      "balance_loss_clip": 1.01751876,
      "balance_loss_mlp": 1.00161517,
      "epoch": 0.43083027715986294,
      "flos": 51997969883520.0,
      "grad_norm": 0.7924929334248949,
      "language_loss": 0.60932624,
      "learning_rate": 2.538549866248549e-06,
      "loss": 0.62988877,
      "num_input_tokens_seen": 77249320,
      "step": 3583,
      "time_per_iteration": 3.1530730724334717
    },
    {
      "auxiliary_loss_clip": 0.01177021,
      "auxiliary_loss_mlp": 0.01029392,
      "balance_loss_clip": 1.05246389,
      "balance_loss_mlp": 1.02071393,
      "epoch": 0.430950520050502,
      "flos": 16690885320960.0,
      "grad_norm": 2.5729074450994536,
      "language_loss": 0.81156611,
      "learning_rate": 2.5377996263768274e-06,
      "loss": 0.83363032,
      "num_input_tokens_seen": 77267400,
      "step": 3584,
      "time_per_iteration": 2.6495094299316406
    },
    {
      "auxiliary_loss_clip": 0.01171248,
      "auxiliary_loss_mlp": 0.01038755,
      "balance_loss_clip": 1.05107844,
      "balance_loss_mlp": 1.02997565,
      "epoch": 0.4310707629411411,
      "flos": 24608433726720.0,
      "grad_norm": 1.9486414217562429,
      "language_loss": 0.6862939,
      "learning_rate": 2.5370493049213293e-06,
      "loss": 0.70839393,
      "num_input_tokens_seen": 77287045,
      "step": 3585,
      "time_per_iteration": 2.7423524856567383
    },
    {
      "auxiliary_loss_clip": 0.01084145,
      "auxiliary_loss_mlp": 0.01029948,
      "balance_loss_clip": 1.04283309,
      "balance_loss_mlp": 1.02155566,
      "epoch": 0.4311910058317802,
      "flos": 26432983019520.0,
      "grad_norm": 1.9674277147102284,
      "language_loss": 0.79746211,
      "learning_rate": 2.536298901995878e-06,
      "loss": 0.8186031,
      "num_input_tokens_seen": 77306255,
      "step": 3586,
      "time_per_iteration": 3.07484769821167
    },
    {
      "auxiliary_loss_clip": 0.01163398,
      "auxiliary_loss_mlp": 0.01029193,
      "balance_loss_clip": 1.05138612,
      "balance_loss_mlp": 1.01993048,
      "epoch": 0.43131124872241927,
      "flos": 25155891889920.0,
      "grad_norm": 1.9803882135957005,
      "language_loss": 0.80058783,
      "learning_rate": 2.535548417714311e-06,
      "loss": 0.8225137,
      "num_input_tokens_seen": 77325555,
      "step": 3587,
      "time_per_iteration": 2.9281415939331055
    },
    {
      "auxiliary_loss_clip": 0.01178842,
      "auxiliary_loss_mlp": 0.01030134,
      "balance_loss_clip": 1.05081105,
      "balance_loss_mlp": 1.02150941,
      "epoch": 0.4314314916130584,
      "flos": 21614812479360.0,
      "grad_norm": 1.6373541943091054,
      "language_loss": 0.87224609,
      "learning_rate": 2.534797852190474e-06,
      "loss": 0.89433587,
      "num_input_tokens_seen": 77345735,
      "step": 3588,
      "time_per_iteration": 2.642416477203369
    },
    {
      "auxiliary_loss_clip": 0.01170043,
      "auxiliary_loss_mlp": 0.01034634,
      "balance_loss_clip": 1.04970396,
      "balance_loss_mlp": 1.02603889,
      "epoch": 0.4315517345036975,
      "flos": 19275016544640.0,
      "grad_norm": 2.049066740227965,
      "language_loss": 0.81519711,
      "learning_rate": 2.5340472055382283e-06,
      "loss": 0.83724391,
      "num_input_tokens_seen": 77361765,
      "step": 3589,
      "time_per_iteration": 2.6205978393554688
    },
    {
      "auxiliary_loss_clip": 0.0114527,
      "auxiliary_loss_mlp": 0.01025078,
      "balance_loss_clip": 1.04720354,
      "balance_loss_mlp": 1.01657856,
      "epoch": 0.43167197739433655,
      "flos": 24273853516800.0,
      "grad_norm": 2.1644296582411235,
      "language_loss": 0.81341338,
      "learning_rate": 2.5332964778714468e-06,
      "loss": 0.8351168,
      "num_input_tokens_seen": 77378950,
      "step": 3590,
      "time_per_iteration": 3.1484291553497314
    },
    {
      "auxiliary_loss_clip": 0.01143561,
      "auxiliary_loss_mlp": 0.01033588,
      "balance_loss_clip": 1.0506556,
      "balance_loss_mlp": 1.02498746,
      "epoch": 0.43179222028497566,
      "flos": 16867816738560.0,
      "grad_norm": 1.6278454847381083,
      "language_loss": 0.66170764,
      "learning_rate": 2.5325456693040123e-06,
      "loss": 0.68347907,
      "num_input_tokens_seen": 77396145,
      "step": 3591,
      "time_per_iteration": 2.7334723472595215
    },
    {
      "auxiliary_loss_clip": 0.01181687,
      "auxiliary_loss_mlp": 0.01027527,
      "balance_loss_clip": 1.05277085,
      "balance_loss_mlp": 1.01834846,
      "epoch": 0.43191246317561477,
      "flos": 17639214243840.0,
      "grad_norm": 2.478333963376418,
      "language_loss": 0.75274992,
      "learning_rate": 2.531794779949824e-06,
      "loss": 0.77484208,
      "num_input_tokens_seen": 77414045,
      "step": 3592,
      "time_per_iteration": 2.5936429500579834
    },
    {
      "auxiliary_loss_clip": 0.01134848,
      "auxiliary_loss_mlp": 0.01032632,
      "balance_loss_clip": 1.04709888,
      "balance_loss_mlp": 1.0245676,
      "epoch": 0.4320327060662538,
      "flos": 23878800760320.0,
      "grad_norm": 1.8413564400909865,
      "language_loss": 0.88030982,
      "learning_rate": 2.5310438099227903e-06,
      "loss": 0.90198463,
      "num_input_tokens_seen": 77431310,
      "step": 3593,
      "time_per_iteration": 2.750535249710083
    },
    {
      "auxiliary_loss_clip": 0.01075993,
      "auxiliary_loss_mlp": 0.01003302,
      "balance_loss_clip": 1.0178895,
      "balance_loss_mlp": 1.00158548,
      "epoch": 0.43215294895689293,
      "flos": 66394917959040.0,
      "grad_norm": 0.7983724054049584,
      "language_loss": 0.53287089,
      "learning_rate": 2.530292759336833e-06,
      "loss": 0.55366385,
      "num_input_tokens_seen": 77492045,
      "step": 3594,
      "time_per_iteration": 3.2955424785614014
    },
    {
      "auxiliary_loss_clip": 0.01156676,
      "auxiliary_loss_mlp": 0.01029196,
      "balance_loss_clip": 1.05027533,
      "balance_loss_mlp": 1.02018368,
      "epoch": 0.432273191847532,
      "flos": 20594267262720.0,
      "grad_norm": 2.112965963138508,
      "language_loss": 0.69893849,
      "learning_rate": 2.5295416283058855e-06,
      "loss": 0.7207973,
      "num_input_tokens_seen": 77510910,
      "step": 3595,
      "time_per_iteration": 2.6660494804382324
    },
    {
      "auxiliary_loss_clip": 0.01155464,
      "auxiliary_loss_mlp": 0.00762855,
      "balance_loss_clip": 1.04966319,
      "balance_loss_mlp": 1.00049376,
      "epoch": 0.4323934347381711,
      "flos": 19282127437440.0,
      "grad_norm": 1.5714699853993486,
      "language_loss": 0.66103268,
      "learning_rate": 2.5287904169438943e-06,
      "loss": 0.68021584,
      "num_input_tokens_seen": 77530115,
      "step": 3596,
      "time_per_iteration": 2.722686529159546
    },
    {
      "auxiliary_loss_clip": 0.01110932,
      "auxiliary_loss_mlp": 0.01029143,
      "balance_loss_clip": 1.04575348,
      "balance_loss_mlp": 1.01972604,
      "epoch": 0.4325136776288102,
      "flos": 21726315273600.0,
      "grad_norm": 2.793167884968842,
      "language_loss": 0.63825214,
      "learning_rate": 2.528039125364817e-06,
      "loss": 0.65965289,
      "num_input_tokens_seen": 77548920,
      "step": 3597,
      "time_per_iteration": 2.9336624145507812
    },
    {
      "auxiliary_loss_clip": 0.01146538,
      "auxiliary_loss_mlp": 0.01029747,
      "balance_loss_clip": 1.04821038,
      "balance_loss_mlp": 1.02062118,
      "epoch": 0.43263392051944927,
      "flos": 22340746344960.0,
      "grad_norm": 2.0117084046638443,
      "language_loss": 0.76061273,
      "learning_rate": 2.5272877536826246e-06,
      "loss": 0.78237557,
      "num_input_tokens_seen": 77567715,
      "step": 3598,
      "time_per_iteration": 3.0277509689331055
    },
    {
      "auxiliary_loss_clip": 0.01130976,
      "auxiliary_loss_mlp": 0.01032365,
      "balance_loss_clip": 1.0444454,
      "balance_loss_mlp": 1.02336419,
      "epoch": 0.4327541634100884,
      "flos": 29168406328320.0,
      "grad_norm": 1.9426299984911068,
      "language_loss": 0.70967764,
      "learning_rate": 2.5265363020112986e-06,
      "loss": 0.73131108,
      "num_input_tokens_seen": 77588035,
      "step": 3599,
      "time_per_iteration": 3.9300270080566406
    },
    {
      "auxiliary_loss_clip": 0.01175915,
      "auxiliary_loss_mlp": 0.01033987,
      "balance_loss_clip": 1.05535424,
      "balance_loss_mlp": 1.02394342,
      "epoch": 0.4328744063007275,
      "flos": 26067448264320.0,
      "grad_norm": 1.920568097569299,
      "language_loss": 0.8421188,
      "learning_rate": 2.5257847704648344e-06,
      "loss": 0.86421788,
      "num_input_tokens_seen": 77609265,
      "step": 3600,
      "time_per_iteration": 3.6557302474975586
    },
    {
      "auxiliary_loss_clip": 0.0118867,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.05420578,
      "balance_loss_mlp": 1.02135324,
      "epoch": 0.43299464919136654,
      "flos": 16581357774720.0,
      "grad_norm": 3.021355966762936,
      "language_loss": 0.761464,
      "learning_rate": 2.525033159157239e-06,
      "loss": 0.78365153,
      "num_input_tokens_seen": 77625580,
      "step": 3601,
      "time_per_iteration": 2.6542649269104004
    },
    {
      "auxiliary_loss_clip": 0.01172487,
      "auxiliary_loss_mlp": 0.01031782,
      "balance_loss_clip": 1.05185318,
      "balance_loss_mlp": 1.02287686,
      "epoch": 0.43311489208200565,
      "flos": 16107265140480.0,
      "grad_norm": 1.7839130845087798,
      "language_loss": 0.77092624,
      "learning_rate": 2.52428146820253e-06,
      "loss": 0.79296893,
      "num_input_tokens_seen": 77643835,
      "step": 3602,
      "time_per_iteration": 2.626575469970703
    },
    {
      "auxiliary_loss_clip": 0.01148991,
      "auxiliary_loss_mlp": 0.01031178,
      "balance_loss_clip": 1.05135489,
      "balance_loss_mlp": 1.02237463,
      "epoch": 0.43323513497264476,
      "flos": 22930220442240.0,
      "grad_norm": 1.7280420835321257,
      "language_loss": 0.81696415,
      "learning_rate": 2.52352969771474e-06,
      "loss": 0.83876586,
      "num_input_tokens_seen": 77663060,
      "step": 3603,
      "time_per_iteration": 2.7398667335510254
    },
    {
      "auxiliary_loss_clip": 0.01162522,
      "auxiliary_loss_mlp": 0.01035204,
      "balance_loss_clip": 1.05162907,
      "balance_loss_mlp": 1.02675223,
      "epoch": 0.4333553778632838,
      "flos": 25299031587840.0,
      "grad_norm": 2.2304841587324002,
      "language_loss": 0.88602364,
      "learning_rate": 2.5227778478079106e-06,
      "loss": 0.90800095,
      "num_input_tokens_seen": 77682470,
      "step": 3604,
      "time_per_iteration": 4.6022655963897705
    },
    {
      "auxiliary_loss_clip": 0.01169159,
      "auxiliary_loss_mlp": 0.01030859,
      "balance_loss_clip": 1.05015874,
      "balance_loss_mlp": 1.02269316,
      "epoch": 0.43347562075392293,
      "flos": 19387165783680.0,
      "grad_norm": 1.5385197407757567,
      "language_loss": 0.77083331,
      "learning_rate": 2.522025918596098e-06,
      "loss": 0.79283345,
      "num_input_tokens_seen": 77700770,
      "step": 3605,
      "time_per_iteration": 2.706249475479126
    },
    {
      "auxiliary_loss_clip": 0.01175877,
      "auxiliary_loss_mlp": 0.0103032,
      "balance_loss_clip": 1.05314636,
      "balance_loss_mlp": 1.02253604,
      "epoch": 0.43359586364456204,
      "flos": 26325969425280.0,
      "grad_norm": 1.4686010595213426,
      "language_loss": 0.65865022,
      "learning_rate": 2.521273910193368e-06,
      "loss": 0.68071216,
      "num_input_tokens_seen": 77723950,
      "step": 3606,
      "time_per_iteration": 2.8919973373413086
    },
    {
      "auxiliary_loss_clip": 0.01182039,
      "auxiliary_loss_mlp": 0.01029622,
      "balance_loss_clip": 1.05411959,
      "balance_loss_mlp": 1.02072906,
      "epoch": 0.4337161065352011,
      "flos": 15989261984640.0,
      "grad_norm": 2.698900062602605,
      "language_loss": 0.87418818,
      "learning_rate": 2.5205218227138006e-06,
      "loss": 0.89630485,
      "num_input_tokens_seen": 77736905,
      "step": 3607,
      "time_per_iteration": 2.6692521572113037
    },
    {
      "auxiliary_loss_clip": 0.01188521,
      "auxiliary_loss_mlp": 0.01030953,
      "balance_loss_clip": 1.05239415,
      "balance_loss_mlp": 1.02245986,
      "epoch": 0.4338363494258402,
      "flos": 20224710184320.0,
      "grad_norm": 1.9986375322641783,
      "language_loss": 0.7884537,
      "learning_rate": 2.519769656271486e-06,
      "loss": 0.81064844,
      "num_input_tokens_seen": 77754325,
      "step": 3608,
      "time_per_iteration": 2.598769426345825
    },
    {
      "auxiliary_loss_clip": 0.011236,
      "auxiliary_loss_mlp": 0.0102931,
      "balance_loss_clip": 1.04641795,
      "balance_loss_mlp": 1.02053082,
      "epoch": 0.43395659231647926,
      "flos": 20083904870400.0,
      "grad_norm": 2.311717771648048,
      "language_loss": 0.67433977,
      "learning_rate": 2.5190174109805285e-06,
      "loss": 0.69586891,
      "num_input_tokens_seen": 77774150,
      "step": 3609,
      "time_per_iteration": 2.7555224895477295
    },
    {
      "auxiliary_loss_clip": 0.01149969,
      "auxiliary_loss_mlp": 0.01037037,
      "balance_loss_clip": 1.04822731,
      "balance_loss_mlp": 1.02773833,
      "epoch": 0.43407683520711837,
      "flos": 19901801894400.0,
      "grad_norm": 2.192360444066044,
      "language_loss": 0.64147007,
      "learning_rate": 2.518265086955042e-06,
      "loss": 0.66334015,
      "num_input_tokens_seen": 77791870,
      "step": 3610,
      "time_per_iteration": 2.6512627601623535
    },
    {
      "auxiliary_loss_clip": 0.01187387,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.05180478,
      "balance_loss_mlp": 1.02303863,
      "epoch": 0.4341970780977575,
      "flos": 23108732058240.0,
      "grad_norm": 1.7632546807791794,
      "language_loss": 0.83763015,
      "learning_rate": 2.5175126843091534e-06,
      "loss": 0.85982263,
      "num_input_tokens_seen": 77811240,
      "step": 3611,
      "time_per_iteration": 2.7160422801971436
    },
    {
      "auxiliary_loss_clip": 0.0116531,
      "auxiliary_loss_mlp": 0.01033073,
      "balance_loss_clip": 1.05426753,
      "balance_loss_mlp": 1.02425194,
      "epoch": 0.43431732098839654,
      "flos": 37408288406400.0,
      "grad_norm": 2.3094431773180197,
      "language_loss": 0.75538152,
      "learning_rate": 2.5167602031570034e-06,
      "loss": 0.77736533,
      "num_input_tokens_seen": 77831425,
      "step": 3612,
      "time_per_iteration": 2.784390926361084
    },
    {
      "auxiliary_loss_clip": 0.01190302,
      "auxiliary_loss_mlp": 0.0102703,
      "balance_loss_clip": 1.05463409,
      "balance_loss_mlp": 1.01868546,
      "epoch": 0.43443756387903565,
      "flos": 31868206323840.0,
      "grad_norm": 1.6567404832385049,
      "language_loss": 0.73378932,
      "learning_rate": 2.51600764361274e-06,
      "loss": 0.75596267,
      "num_input_tokens_seen": 77852950,
      "step": 3613,
      "time_per_iteration": 2.717949867248535
    },
    {
      "auxiliary_loss_clip": 0.01191602,
      "auxiliary_loss_mlp": 0.01033689,
      "balance_loss_clip": 1.05484939,
      "balance_loss_mlp": 1.02468908,
      "epoch": 0.43455780676967476,
      "flos": 23477139901440.0,
      "grad_norm": 3.917447712906433,
      "language_loss": 0.78658634,
      "learning_rate": 2.5152550057905283e-06,
      "loss": 0.80883926,
      "num_input_tokens_seen": 77872840,
      "step": 3614,
      "time_per_iteration": 2.6141862869262695
    },
    {
      "auxiliary_loss_clip": 0.01177016,
      "auxiliary_loss_mlp": 0.00763787,
      "balance_loss_clip": 1.0548656,
      "balance_loss_mlp": 1.00044918,
      "epoch": 0.4346780496603138,
      "flos": 24207060176640.0,
      "grad_norm": 1.9604280709142665,
      "language_loss": 0.77136207,
      "learning_rate": 2.5145022898045415e-06,
      "loss": 0.79077011,
      "num_input_tokens_seen": 77892025,
      "step": 3615,
      "time_per_iteration": 2.6132619380950928
    },
    {
      "auxiliary_loss_clip": 0.01162322,
      "auxiliary_loss_mlp": 0.01027145,
      "balance_loss_clip": 1.04848802,
      "balance_loss_mlp": 1.01824605,
      "epoch": 0.4347982925509529,
      "flos": 17092366611840.0,
      "grad_norm": 1.9988865896241532,
      "language_loss": 0.90022373,
      "learning_rate": 2.5137494957689664e-06,
      "loss": 0.92211848,
      "num_input_tokens_seen": 77907635,
      "step": 3616,
      "time_per_iteration": 2.607708215713501
    },
    {
      "auxiliary_loss_clip": 0.01060772,
      "auxiliary_loss_mlp": 0.01004011,
      "balance_loss_clip": 1.01528203,
      "balance_loss_mlp": 1.00230587,
      "epoch": 0.43491853544159204,
      "flos": 60945544696320.0,
      "grad_norm": 0.7647456987844858,
      "language_loss": 0.57326162,
      "learning_rate": 2.5129966237980016e-06,
      "loss": 0.5939095,
      "num_input_tokens_seen": 77970630,
      "step": 3617,
      "time_per_iteration": 3.2574472427368164
    },
    {
      "auxiliary_loss_clip": 0.011468,
      "auxiliary_loss_mlp": 0.01029277,
      "balance_loss_clip": 1.04853559,
      "balance_loss_mlp": 1.02102137,
      "epoch": 0.4350387783322311,
      "flos": 21944652094080.0,
      "grad_norm": 1.8666160967155916,
      "language_loss": 0.78088498,
      "learning_rate": 2.512243674005857e-06,
      "loss": 0.8026458,
      "num_input_tokens_seen": 77989995,
      "step": 3618,
      "time_per_iteration": 2.8058979511260986
    },
    {
      "auxiliary_loss_clip": 0.0111534,
      "auxiliary_loss_mlp": 0.0103203,
      "balance_loss_clip": 1.0450995,
      "balance_loss_mlp": 1.02373326,
      "epoch": 0.4351590212228702,
      "flos": 25082705928960.0,
      "grad_norm": 2.062267515332014,
      "language_loss": 0.8613615,
      "learning_rate": 2.5114906465067537e-06,
      "loss": 0.88283521,
      "num_input_tokens_seen": 78010980,
      "step": 3619,
      "time_per_iteration": 2.809054136276245
    },
    {
      "auxiliary_loss_clip": 0.01174691,
      "auxiliary_loss_mlp": 0.01029832,
      "balance_loss_clip": 1.05060339,
      "balance_loss_mlp": 1.02142799,
      "epoch": 0.4352792641135093,
      "flos": 21506541909120.0,
      "grad_norm": 1.9836836797076929,
      "language_loss": 0.74792045,
      "learning_rate": 2.5107375414149264e-06,
      "loss": 0.76996571,
      "num_input_tokens_seen": 78030225,
      "step": 3620,
      "time_per_iteration": 2.745995044708252
    },
    {
      "auxiliary_loss_clip": 0.01119862,
      "auxiliary_loss_mlp": 0.01030194,
      "balance_loss_clip": 1.04159379,
      "balance_loss_mlp": 1.02105105,
      "epoch": 0.43539950700414837,
      "flos": 16253457494400.0,
      "grad_norm": 2.1920158282005495,
      "language_loss": 0.71927905,
      "learning_rate": 2.5099843588446197e-06,
      "loss": 0.74077964,
      "num_input_tokens_seen": 78048545,
      "step": 3621,
      "time_per_iteration": 2.7285523414611816
    },
    {
      "auxiliary_loss_clip": 0.01137535,
      "auxiliary_loss_mlp": 0.01036889,
      "balance_loss_clip": 1.04900301,
      "balance_loss_mlp": 1.02812064,
      "epoch": 0.4355197498947875,
      "flos": 16691819074560.0,
      "grad_norm": 1.6955827829356918,
      "language_loss": 0.617275,
      "learning_rate": 2.509231098910091e-06,
      "loss": 0.63901925,
      "num_input_tokens_seen": 78068415,
      "step": 3622,
      "time_per_iteration": 2.764375686645508
    },
    {
      "auxiliary_loss_clip": 0.01158795,
      "auxiliary_loss_mlp": 0.01029062,
      "balance_loss_clip": 1.05437517,
      "balance_loss_mlp": 1.02081823,
      "epoch": 0.4356399927854266,
      "flos": 16362733645440.0,
      "grad_norm": 3.6049638575166365,
      "language_loss": 0.7506777,
      "learning_rate": 2.508477761725611e-06,
      "loss": 0.7725563,
      "num_input_tokens_seen": 78086690,
      "step": 3623,
      "time_per_iteration": 2.721052885055542
    },
    {
      "auxiliary_loss_clip": 0.01175953,
      "auxiliary_loss_mlp": 0.01034877,
      "balance_loss_clip": 1.04986525,
      "balance_loss_mlp": 1.02630532,
      "epoch": 0.43576023567606564,
      "flos": 17202037812480.0,
      "grad_norm": 1.842106936216907,
      "language_loss": 0.81182772,
      "learning_rate": 2.507724347405458e-06,
      "loss": 0.83393598,
      "num_input_tokens_seen": 78104640,
      "step": 3624,
      "time_per_iteration": 2.5896763801574707
    },
    {
      "auxiliary_loss_clip": 0.0112434,
      "auxiliary_loss_mlp": 0.01027231,
      "balance_loss_clip": 1.04282761,
      "balance_loss_mlp": 1.01901793,
      "epoch": 0.43588047856670475,
      "flos": 15917656222080.0,
      "grad_norm": 2.0968851276627416,
      "language_loss": 0.82015496,
      "learning_rate": 2.5069708560639243e-06,
      "loss": 0.84167069,
      "num_input_tokens_seen": 78122550,
      "step": 3625,
      "time_per_iteration": 3.639252185821533
    },
    {
      "auxiliary_loss_clip": 0.01145596,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.04774213,
      "balance_loss_mlp": 1.01917672,
      "epoch": 0.4360007214573438,
      "flos": 23659566099840.0,
      "grad_norm": 3.5340223065839527,
      "language_loss": 0.61671197,
      "learning_rate": 2.5062172878153158e-06,
      "loss": 0.63844454,
      "num_input_tokens_seen": 78141825,
      "step": 3626,
      "time_per_iteration": 3.664763927459717
    },
    {
      "auxiliary_loss_clip": 0.01123327,
      "auxiliary_loss_mlp": 0.01031167,
      "balance_loss_clip": 1.04452574,
      "balance_loss_mlp": 1.02275634,
      "epoch": 0.4361209643479829,
      "flos": 21978767036160.0,
      "grad_norm": 1.7691054769969645,
      "language_loss": 0.86904716,
      "learning_rate": 2.505463642773947e-06,
      "loss": 0.8905921,
      "num_input_tokens_seen": 78161790,
      "step": 3627,
      "time_per_iteration": 2.8500239849090576
    },
    {
      "auxiliary_loss_clip": 0.01149151,
      "auxiliary_loss_mlp": 0.00763945,
      "balance_loss_clip": 1.050951,
      "balance_loss_mlp": 1.00036335,
      "epoch": 0.43624120723862203,
      "flos": 17420159151360.0,
      "grad_norm": 2.5462585670334215,
      "language_loss": 0.75086522,
      "learning_rate": 2.504709921054146e-06,
      "loss": 0.76999617,
      "num_input_tokens_seen": 78178605,
      "step": 3628,
      "time_per_iteration": 2.771569013595581
    },
    {
      "auxiliary_loss_clip": 0.01137486,
      "auxiliary_loss_mlp": 0.01035151,
      "balance_loss_clip": 1.04330301,
      "balance_loss_mlp": 1.02615643,
      "epoch": 0.4363614501292611,
      "flos": 17895293280000.0,
      "grad_norm": 2.5031655976005633,
      "language_loss": 0.83599633,
      "learning_rate": 2.50395612277025e-06,
      "loss": 0.8577227,
      "num_input_tokens_seen": 78194460,
      "step": 3629,
      "time_per_iteration": 2.750523805618286
    },
    {
      "auxiliary_loss_clip": 0.01163549,
      "auxiliary_loss_mlp": 0.01034063,
      "balance_loss_clip": 1.04959452,
      "balance_loss_mlp": 1.0253607,
      "epoch": 0.4364816930199002,
      "flos": 20302888135680.0,
      "grad_norm": 2.022455579760982,
      "language_loss": 0.73286033,
      "learning_rate": 2.503202248036612e-06,
      "loss": 0.75483644,
      "num_input_tokens_seen": 78213315,
      "step": 3630,
      "time_per_iteration": 4.556641578674316
    },
    {
      "auxiliary_loss_clip": 0.01190852,
      "auxiliary_loss_mlp": 0.01026487,
      "balance_loss_clip": 1.05545998,
      "balance_loss_mlp": 1.01835656,
      "epoch": 0.4366019359105393,
      "flos": 24061334699520.0,
      "grad_norm": 1.6475859406748012,
      "language_loss": 0.73651862,
      "learning_rate": 2.5024482969675927e-06,
      "loss": 0.75869203,
      "num_input_tokens_seen": 78233270,
      "step": 3631,
      "time_per_iteration": 2.6520493030548096
    },
    {
      "auxiliary_loss_clip": 0.01138276,
      "auxiliary_loss_mlp": 0.01026693,
      "balance_loss_clip": 1.04913259,
      "balance_loss_mlp": 1.01837778,
      "epoch": 0.43672217880117836,
      "flos": 21754109422080.0,
      "grad_norm": 2.3585639464057495,
      "language_loss": 0.84390378,
      "learning_rate": 2.501694269677566e-06,
      "loss": 0.8655535,
      "num_input_tokens_seen": 78251040,
      "step": 3632,
      "time_per_iteration": 2.8094675540924072
    },
    {
      "auxiliary_loss_clip": 0.01177098,
      "auxiliary_loss_mlp": 0.01028165,
      "balance_loss_clip": 1.04969203,
      "balance_loss_mlp": 1.01937342,
      "epoch": 0.4368424216918175,
      "flos": 18035200753920.0,
      "grad_norm": 3.202918260719624,
      "language_loss": 0.80487573,
      "learning_rate": 2.500940166280918e-06,
      "loss": 0.82692832,
      "num_input_tokens_seen": 78269470,
      "step": 3633,
      "time_per_iteration": 2.745352268218994
    },
    {
      "auxiliary_loss_clip": 0.01171609,
      "auxiliary_loss_mlp": 0.01034182,
      "balance_loss_clip": 1.05125546,
      "balance_loss_mlp": 1.02610552,
      "epoch": 0.4369626645824566,
      "flos": 25447127362560.0,
      "grad_norm": 2.0731942733404503,
      "language_loss": 0.79098916,
      "learning_rate": 2.500185986892045e-06,
      "loss": 0.81304705,
      "num_input_tokens_seen": 78288955,
      "step": 3634,
      "time_per_iteration": 2.682300090789795
    },
    {
      "auxiliary_loss_clip": 0.0116949,
      "auxiliary_loss_mlp": 0.010321,
      "balance_loss_clip": 1.04999232,
      "balance_loss_mlp": 1.02305841,
      "epoch": 0.43708290747309564,
      "flos": 25302694775040.0,
      "grad_norm": 2.126522284228904,
      "language_loss": 0.7747997,
      "learning_rate": 2.499431731625355e-06,
      "loss": 0.79681551,
      "num_input_tokens_seen": 78307980,
      "step": 3635,
      "time_per_iteration": 2.6690468788146973
    },
    {
      "auxiliary_loss_clip": 0.01190306,
      "auxiliary_loss_mlp": 0.01033377,
      "balance_loss_clip": 1.0533222,
      "balance_loss_mlp": 1.02421594,
      "epoch": 0.43720315036373475,
      "flos": 31575103344000.0,
      "grad_norm": 2.0671254077363006,
      "language_loss": 0.79612643,
      "learning_rate": 2.4986774005952686e-06,
      "loss": 0.81836331,
      "num_input_tokens_seen": 78330355,
      "step": 3636,
      "time_per_iteration": 2.749903917312622
    },
    {
      "auxiliary_loss_clip": 0.01171152,
      "auxiliary_loss_mlp": 0.01028063,
      "balance_loss_clip": 1.05230618,
      "balance_loss_mlp": 1.01956916,
      "epoch": 0.43732339325437386,
      "flos": 23112000195840.0,
      "grad_norm": 2.5440618583045054,
      "language_loss": 0.8481065,
      "learning_rate": 2.4979229939162166e-06,
      "loss": 0.87009865,
      "num_input_tokens_seen": 78349135,
      "step": 3637,
      "time_per_iteration": 2.6842153072357178
    },
    {
      "auxiliary_loss_clip": 0.01172421,
      "auxiliary_loss_mlp": 0.01027721,
      "balance_loss_clip": 1.05462885,
      "balance_loss_mlp": 1.01965666,
      "epoch": 0.4374436361450129,
      "flos": 27746272080000.0,
      "grad_norm": 1.8046246040435532,
      "language_loss": 0.80381012,
      "learning_rate": 2.4971685117026433e-06,
      "loss": 0.82581151,
      "num_input_tokens_seen": 78368900,
      "step": 3638,
      "time_per_iteration": 2.629162073135376
    },
    {
      "auxiliary_loss_clip": 0.01178834,
      "auxiliary_loss_mlp": 0.01041077,
      "balance_loss_clip": 1.05388379,
      "balance_loss_mlp": 1.03260136,
      "epoch": 0.437563879035652,
      "flos": 24172370616960.0,
      "grad_norm": 1.504283242781577,
      "language_loss": 0.76778221,
      "learning_rate": 2.4964139540690018e-06,
      "loss": 0.78998131,
      "num_input_tokens_seen": 78392235,
      "step": 3639,
      "time_per_iteration": 2.728782892227173
    },
    {
      "auxiliary_loss_clip": 0.01146862,
      "auxiliary_loss_mlp": 0.01038161,
      "balance_loss_clip": 1.04829431,
      "balance_loss_mlp": 1.02833855,
      "epoch": 0.4376841219262911,
      "flos": 23477211728640.0,
      "grad_norm": 2.013686258211536,
      "language_loss": 0.73139006,
      "learning_rate": 2.495659321129758e-06,
      "loss": 0.75324035,
      "num_input_tokens_seen": 78409980,
      "step": 3640,
      "time_per_iteration": 2.7193362712860107
    },
    {
      "auxiliary_loss_clip": 0.0117106,
      "auxiliary_loss_mlp": 0.01032054,
      "balance_loss_clip": 1.05028033,
      "balance_loss_mlp": 1.02328634,
      "epoch": 0.4378043648169302,
      "flos": 25447809720960.0,
      "grad_norm": 1.8934201955150876,
      "language_loss": 0.75248373,
      "learning_rate": 2.494904612999389e-06,
      "loss": 0.77451491,
      "num_input_tokens_seen": 78428690,
      "step": 3641,
      "time_per_iteration": 2.6830637454986572
    },
    {
      "auxiliary_loss_clip": 0.01071523,
      "auxiliary_loss_mlp": 0.01002095,
      "balance_loss_clip": 1.0180527,
      "balance_loss_mlp": 1.00046158,
      "epoch": 0.4379246077075693,
      "flos": 53914056986880.0,
      "grad_norm": 0.7676143853040126,
      "language_loss": 0.56518102,
      "learning_rate": 2.4941498297923843e-06,
      "loss": 0.58591723,
      "num_input_tokens_seen": 78489260,
      "step": 3642,
      "time_per_iteration": 3.246466636657715
    },
    {
      "auxiliary_loss_clip": 0.01171597,
      "auxiliary_loss_mlp": 0.01023963,
      "balance_loss_clip": 1.05152178,
      "balance_loss_mlp": 1.01531434,
      "epoch": 0.43804485059820836,
      "flos": 20588305605120.0,
      "grad_norm": 1.7455108822541578,
      "language_loss": 0.69837618,
      "learning_rate": 2.4933949716232424e-06,
      "loss": 0.72033179,
      "num_input_tokens_seen": 78506785,
      "step": 3643,
      "time_per_iteration": 2.6431987285614014
    },
    {
      "auxiliary_loss_clip": 0.0114624,
      "auxiliary_loss_mlp": 0.01028418,
      "balance_loss_clip": 1.05061781,
      "balance_loss_mlp": 1.02010942,
      "epoch": 0.43816509348884747,
      "flos": 23876214981120.0,
      "grad_norm": 1.9545658271162547,
      "language_loss": 0.7409637,
      "learning_rate": 2.492640038606476e-06,
      "loss": 0.76271027,
      "num_input_tokens_seen": 78525150,
      "step": 3644,
      "time_per_iteration": 2.7577614784240723
    },
    {
      "auxiliary_loss_clip": 0.01173514,
      "auxiliary_loss_mlp": 0.01037054,
      "balance_loss_clip": 1.05115438,
      "balance_loss_mlp": 1.02773166,
      "epoch": 0.4382853363794866,
      "flos": 14684448533760.0,
      "grad_norm": 2.1080819437874614,
      "language_loss": 0.78711116,
      "learning_rate": 2.491885030856608e-06,
      "loss": 0.80921686,
      "num_input_tokens_seen": 78543245,
      "step": 3645,
      "time_per_iteration": 2.6877214908599854
    },
    {
      "auxiliary_loss_clip": 0.01162733,
      "auxiliary_loss_mlp": 0.01031127,
      "balance_loss_clip": 1.05057263,
      "balance_loss_mlp": 1.02236485,
      "epoch": 0.43840557927012563,
      "flos": 17165301177600.0,
      "grad_norm": 2.1029898541907674,
      "language_loss": 0.82339597,
      "learning_rate": 2.4911299484881713e-06,
      "loss": 0.84533453,
      "num_input_tokens_seen": 78560775,
      "step": 3646,
      "time_per_iteration": 2.6882827281951904
    },
    {
      "auxiliary_loss_clip": 0.01153972,
      "auxiliary_loss_mlp": 0.01029091,
      "balance_loss_clip": 1.04748809,
      "balance_loss_mlp": 1.02111614,
      "epoch": 0.43852582216076474,
      "flos": 19390685316480.0,
      "grad_norm": 1.5973679921460024,
      "language_loss": 0.8084532,
      "learning_rate": 2.490374791615712e-06,
      "loss": 0.83028388,
      "num_input_tokens_seen": 78580800,
      "step": 3647,
      "time_per_iteration": 2.7253425121307373
    },
    {
      "auxiliary_loss_clip": 0.01196546,
      "auxiliary_loss_mlp": 0.00764299,
      "balance_loss_clip": 1.05665731,
      "balance_loss_mlp": 1.00035143,
      "epoch": 0.43864606505140386,
      "flos": 18075133699200.0,
      "grad_norm": 2.875957993142965,
      "language_loss": 0.77799833,
      "learning_rate": 2.4896195603537867e-06,
      "loss": 0.79760671,
      "num_input_tokens_seen": 78595410,
      "step": 3648,
      "time_per_iteration": 2.637045383453369
    },
    {
      "auxiliary_loss_clip": 0.01128237,
      "auxiliary_loss_mlp": 0.01033277,
      "balance_loss_clip": 1.05150843,
      "balance_loss_mlp": 1.02501607,
      "epoch": 0.4387663079420429,
      "flos": 19644896845440.0,
      "grad_norm": 2.0141729936908925,
      "language_loss": 0.73750466,
      "learning_rate": 2.488864254816964e-06,
      "loss": 0.75911975,
      "num_input_tokens_seen": 78614100,
      "step": 3649,
      "time_per_iteration": 2.7055630683898926
    },
    {
      "auxiliary_loss_clip": 0.01179039,
      "auxiliary_loss_mlp": 0.01033099,
      "balance_loss_clip": 1.05491269,
      "balance_loss_mlp": 1.02356803,
      "epoch": 0.438886550832682,
      "flos": 19719339782400.0,
      "grad_norm": 2.7609316107689157,
      "language_loss": 0.68331492,
      "learning_rate": 2.4881088751198218e-06,
      "loss": 0.70543635,
      "num_input_tokens_seen": 78632260,
      "step": 3650,
      "time_per_iteration": 3.6018640995025635
    },
    {
      "auxiliary_loss_clip": 0.01164122,
      "auxiliary_loss_mlp": 0.0103165,
      "balance_loss_clip": 1.05024052,
      "balance_loss_mlp": 1.0223037,
      "epoch": 0.43900679372332113,
      "flos": 14536675981440.0,
      "grad_norm": 2.846705354740708,
      "language_loss": 0.64311081,
      "learning_rate": 2.4873534213769517e-06,
      "loss": 0.66506851,
      "num_input_tokens_seen": 78647490,
      "step": 3651,
      "time_per_iteration": 2.677638053894043
    },
    {
      "auxiliary_loss_clip": 0.01142298,
      "auxiliary_loss_mlp": 0.0103067,
      "balance_loss_clip": 1.05200589,
      "balance_loss_mlp": 1.02216387,
      "epoch": 0.4391270366139602,
      "flos": 24056234968320.0,
      "grad_norm": 1.8309671877446523,
      "language_loss": 0.71860278,
      "learning_rate": 2.4865978937029547e-06,
      "loss": 0.74033248,
      "num_input_tokens_seen": 78666470,
      "step": 3652,
      "time_per_iteration": 3.7030718326568604
    },
    {
      "auxiliary_loss_clip": 0.01119676,
      "auxiliary_loss_mlp": 0.0103062,
      "balance_loss_clip": 1.04662538,
      "balance_loss_mlp": 1.02171516,
      "epoch": 0.4392472795045993,
      "flos": 31538510363520.0,
      "grad_norm": 1.8456448251091269,
      "language_loss": 0.66438723,
      "learning_rate": 2.485842292212445e-06,
      "loss": 0.6858902,
      "num_input_tokens_seen": 78687685,
      "step": 3653,
      "time_per_iteration": 2.832625389099121
    },
    {
      "auxiliary_loss_clip": 0.01193443,
      "auxiliary_loss_mlp": 0.01036519,
      "balance_loss_clip": 1.05676913,
      "balance_loss_mlp": 1.02716112,
      "epoch": 0.4393675223952384,
      "flos": 14866300114560.0,
      "grad_norm": 2.099772596910281,
      "language_loss": 0.80553383,
      "learning_rate": 2.485086617020045e-06,
      "loss": 0.82783341,
      "num_input_tokens_seen": 78706180,
      "step": 3654,
      "time_per_iteration": 2.5469486713409424
    },
    {
      "auxiliary_loss_clip": 0.01152483,
      "auxiliary_loss_mlp": 0.01031468,
      "balance_loss_clip": 1.04760838,
      "balance_loss_mlp": 1.0223366,
      "epoch": 0.43948776528587746,
      "flos": 14825900292480.0,
      "grad_norm": 2.036891662439873,
      "language_loss": 0.81894553,
      "learning_rate": 2.4843308682403903e-06,
      "loss": 0.84078509,
      "num_input_tokens_seen": 78723095,
      "step": 3655,
      "time_per_iteration": 2.6249020099639893
    },
    {
      "auxiliary_loss_clip": 0.01188766,
      "auxiliary_loss_mlp": 0.01030442,
      "balance_loss_clip": 1.05382478,
      "balance_loss_mlp": 1.02258003,
      "epoch": 0.4396080081765166,
      "flos": 13914523486080.0,
      "grad_norm": 1.75712339504953,
      "language_loss": 0.82804787,
      "learning_rate": 2.4835750459881294e-06,
      "loss": 0.85023993,
      "num_input_tokens_seen": 78739720,
      "step": 3656,
      "time_per_iteration": 4.3919665813446045
    },
    {
      "auxiliary_loss_clip": 0.01152246,
      "auxiliary_loss_mlp": 0.01037905,
      "balance_loss_clip": 1.04724336,
      "balance_loss_mlp": 1.02873826,
      "epoch": 0.43972825106715563,
      "flos": 18222978078720.0,
      "grad_norm": 1.71272184801934,
      "language_loss": 0.81906033,
      "learning_rate": 2.4828191503779177e-06,
      "loss": 0.84096181,
      "num_input_tokens_seen": 78757820,
      "step": 3657,
      "time_per_iteration": 2.6146507263183594
    },
    {
      "auxiliary_loss_clip": 0.01144836,
      "auxiliary_loss_mlp": 0.01030762,
      "balance_loss_clip": 1.04728198,
      "balance_loss_mlp": 1.02152896,
      "epoch": 0.43984849395779474,
      "flos": 16873239692160.0,
      "grad_norm": 1.9500005310713235,
      "language_loss": 0.8998602,
      "learning_rate": 2.482063181524425e-06,
      "loss": 0.9216162,
      "num_input_tokens_seen": 78773720,
      "step": 3658,
      "time_per_iteration": 2.7031266689300537
    },
    {
      "auxiliary_loss_clip": 0.01196731,
      "auxiliary_loss_mlp": 0.01036804,
      "balance_loss_clip": 1.05806053,
      "balance_loss_mlp": 1.0278933,
      "epoch": 0.43996873684843385,
      "flos": 18691504104960.0,
      "grad_norm": 2.5361305151582014,
      "language_loss": 0.81135154,
      "learning_rate": 2.4813071395423307e-06,
      "loss": 0.83368695,
      "num_input_tokens_seen": 78791285,
      "step": 3659,
      "time_per_iteration": 2.6063992977142334
    },
    {
      "auxiliary_loss_clip": 0.011761,
      "auxiliary_loss_mlp": 0.01030261,
      "balance_loss_clip": 1.05080235,
      "balance_loss_mlp": 1.0212431,
      "epoch": 0.4400889797390729,
      "flos": 23653460787840.0,
      "grad_norm": 2.864284903715874,
      "language_loss": 0.64751011,
      "learning_rate": 2.4805510245463263e-06,
      "loss": 0.66957366,
      "num_input_tokens_seen": 78811440,
      "step": 3660,
      "time_per_iteration": 2.7313833236694336
    },
    {
      "auxiliary_loss_clip": 0.01173397,
      "auxiliary_loss_mlp": 0.01038249,
      "balance_loss_clip": 1.04898262,
      "balance_loss_mlp": 1.02939141,
      "epoch": 0.440209222629712,
      "flos": 23149203707520.0,
      "grad_norm": 2.3341969009804413,
      "language_loss": 0.60399365,
      "learning_rate": 2.4797948366511137e-06,
      "loss": 0.62611008,
      "num_input_tokens_seen": 78831150,
      "step": 3661,
      "time_per_iteration": 2.6798031330108643
    },
    {
      "auxiliary_loss_clip": 0.01143137,
      "auxiliary_loss_mlp": 0.01029456,
      "balance_loss_clip": 1.04524183,
      "balance_loss_mlp": 1.02062893,
      "epoch": 0.4403294655203511,
      "flos": 24823394668800.0,
      "grad_norm": 6.603729715586676,
      "language_loss": 0.76626313,
      "learning_rate": 2.4790385759714055e-06,
      "loss": 0.78798908,
      "num_input_tokens_seen": 78850215,
      "step": 3662,
      "time_per_iteration": 2.7237746715545654
    },
    {
      "auxiliary_loss_clip": 0.01175686,
      "auxiliary_loss_mlp": 0.01027824,
      "balance_loss_clip": 1.05490923,
      "balance_loss_mlp": 1.01927686,
      "epoch": 0.4404497084109902,
      "flos": 22565080736640.0,
      "grad_norm": 1.781715182680308,
      "language_loss": 0.71464992,
      "learning_rate": 2.478282242621926e-06,
      "loss": 0.73668504,
      "num_input_tokens_seen": 78870675,
      "step": 3663,
      "time_per_iteration": 2.7357938289642334
    },
    {
      "auxiliary_loss_clip": 0.01050428,
      "auxiliary_loss_mlp": 0.01002372,
      "balance_loss_clip": 1.01416326,
      "balance_loss_mlp": 1.00063157,
      "epoch": 0.4405699513016293,
      "flos": 64967073448320.0,
      "grad_norm": 0.8377646876700964,
      "language_loss": 0.59570199,
      "learning_rate": 2.477525836717411e-06,
      "loss": 0.61623001,
      "num_input_tokens_seen": 78938440,
      "step": 3664,
      "time_per_iteration": 3.401033878326416
    },
    {
      "auxiliary_loss_clip": 0.01175752,
      "auxiliary_loss_mlp": 0.0103197,
      "balance_loss_clip": 1.05219877,
      "balance_loss_mlp": 1.02297592,
      "epoch": 0.4406901941922684,
      "flos": 35661952978560.0,
      "grad_norm": 2.963150216107812,
      "language_loss": 0.7974894,
      "learning_rate": 2.476769358372606e-06,
      "loss": 0.81956661,
      "num_input_tokens_seen": 78960090,
      "step": 3665,
      "time_per_iteration": 2.7882802486419678
    },
    {
      "auxiliary_loss_clip": 0.01147042,
      "auxiliary_loss_mlp": 0.01030053,
      "balance_loss_clip": 1.05353725,
      "balance_loss_mlp": 1.02164888,
      "epoch": 0.44081043708290746,
      "flos": 18040767361920.0,
      "grad_norm": 2.1932272231585777,
      "language_loss": 0.75011766,
      "learning_rate": 2.4760128077022683e-06,
      "loss": 0.77188861,
      "num_input_tokens_seen": 78978225,
      "step": 3666,
      "time_per_iteration": 2.646496295928955
    },
    {
      "auxiliary_loss_clip": 0.0112399,
      "auxiliary_loss_mlp": 0.01034193,
      "balance_loss_clip": 1.04677105,
      "balance_loss_mlp": 1.02546644,
      "epoch": 0.44093067997354657,
      "flos": 30153507799680.0,
      "grad_norm": 1.4289623974330963,
      "language_loss": 0.6834023,
      "learning_rate": 2.4752561848211672e-06,
      "loss": 0.70498413,
      "num_input_tokens_seen": 79000625,
      "step": 3667,
      "time_per_iteration": 2.8807568550109863
    },
    {
      "auxiliary_loss_clip": 0.01177002,
      "auxiliary_loss_mlp": 0.01028122,
      "balance_loss_clip": 1.05759525,
      "balance_loss_mlp": 1.01943803,
      "epoch": 0.4410509228641857,
      "flos": 23255068066560.0,
      "grad_norm": 1.9279639887968356,
      "language_loss": 0.7140187,
      "learning_rate": 2.4744994898440797e-06,
      "loss": 0.73606992,
      "num_input_tokens_seen": 79019415,
      "step": 3668,
      "time_per_iteration": 2.652148485183716
    },
    {
      "auxiliary_loss_clip": 0.0114929,
      "auxiliary_loss_mlp": 0.0102788,
      "balance_loss_clip": 1.04860258,
      "balance_loss_mlp": 1.01927269,
      "epoch": 0.44117116575482473,
      "flos": 19500571998720.0,
      "grad_norm": 2.1987294749154187,
      "language_loss": 0.83995783,
      "learning_rate": 2.473742722885797e-06,
      "loss": 0.8617295,
      "num_input_tokens_seen": 79038435,
      "step": 3669,
      "time_per_iteration": 2.6833221912384033
    },
    {
      "auxiliary_loss_clip": 0.01176356,
      "auxiliary_loss_mlp": 0.00764355,
      "balance_loss_clip": 1.05465055,
      "balance_loss_mlp": 1.00033581,
      "epoch": 0.44129140864546385,
      "flos": 27053124353280.0,
      "grad_norm": 2.2728954637744287,
      "language_loss": 0.65523899,
      "learning_rate": 2.4729858840611197e-06,
      "loss": 0.67464614,
      "num_input_tokens_seen": 79057345,
      "step": 3670,
      "time_per_iteration": 2.7127597332000732
    },
    {
      "auxiliary_loss_clip": 0.01189256,
      "auxiliary_loss_mlp": 0.01030352,
      "balance_loss_clip": 1.05443692,
      "balance_loss_mlp": 1.02215004,
      "epoch": 0.4414116515361029,
      "flos": 26102101910400.0,
      "grad_norm": 1.7786638519083433,
      "language_loss": 0.72827756,
      "learning_rate": 2.4722289734848605e-06,
      "loss": 0.75047362,
      "num_input_tokens_seen": 79077810,
      "step": 3671,
      "time_per_iteration": 2.663055419921875
    },
    {
      "auxiliary_loss_clip": 0.01145309,
      "auxiliary_loss_mlp": 0.01027758,
      "balance_loss_clip": 1.05255198,
      "balance_loss_mlp": 1.01946056,
      "epoch": 0.441531894426742,
      "flos": 21906083865600.0,
      "grad_norm": 1.9230970220592325,
      "language_loss": 0.77544498,
      "learning_rate": 2.471471991271841e-06,
      "loss": 0.79717571,
      "num_input_tokens_seen": 79094935,
      "step": 3672,
      "time_per_iteration": 2.7645130157470703
    },
    {
      "auxiliary_loss_clip": 0.01166021,
      "auxiliary_loss_mlp": 0.01036898,
      "balance_loss_clip": 1.04863834,
      "balance_loss_mlp": 1.02840459,
      "epoch": 0.4416521373173811,
      "flos": 23437099215360.0,
      "grad_norm": 3.171791582683192,
      "language_loss": 0.79453325,
      "learning_rate": 2.470714937536896e-06,
      "loss": 0.81656241,
      "num_input_tokens_seen": 79113660,
      "step": 3673,
      "time_per_iteration": 2.653517246246338
    },
    {
      "auxiliary_loss_clip": 0.0112897,
      "auxiliary_loss_mlp": 0.01027912,
      "balance_loss_clip": 1.04806066,
      "balance_loss_mlp": 1.01864934,
      "epoch": 0.4417723802080202,
      "flos": 20334345471360.0,
      "grad_norm": 1.7357159380252134,
      "language_loss": 0.703879,
      "learning_rate": 2.469957812394868e-06,
      "loss": 0.72544777,
      "num_input_tokens_seen": 79132470,
      "step": 3674,
      "time_per_iteration": 2.7627782821655273
    },
    {
      "auxiliary_loss_clip": 0.01191097,
      "auxiliary_loss_mlp": 0.01022861,
      "balance_loss_clip": 1.05695665,
      "balance_loss_mlp": 1.0142715,
      "epoch": 0.4418926230986593,
      "flos": 18880682060160.0,
      "grad_norm": 1.8906369535314556,
      "language_loss": 0.76415676,
      "learning_rate": 2.4692006159606148e-06,
      "loss": 0.78629637,
      "num_input_tokens_seen": 79150000,
      "step": 3675,
      "time_per_iteration": 2.6349289417266846
    },
    {
      "auxiliary_loss_clip": 0.01187429,
      "auxiliary_loss_mlp": 0.0103623,
      "balance_loss_clip": 1.0534029,
      "balance_loss_mlp": 1.02684855,
      "epoch": 0.4420128659892984,
      "flos": 19464409981440.0,
      "grad_norm": 1.764557602116056,
      "language_loss": 0.78231192,
      "learning_rate": 2.468443348349e-06,
      "loss": 0.8045485,
      "num_input_tokens_seen": 79167875,
      "step": 3676,
      "time_per_iteration": 2.654224157333374
    },
    {
      "auxiliary_loss_clip": 0.01130671,
      "auxiliary_loss_mlp": 0.01036314,
      "balance_loss_clip": 1.04563379,
      "balance_loss_mlp": 1.02708173,
      "epoch": 0.44213310887993745,
      "flos": 17894359526400.0,
      "grad_norm": 2.548508672115378,
      "language_loss": 0.82324702,
      "learning_rate": 2.467686009674902e-06,
      "loss": 0.84491688,
      "num_input_tokens_seen": 79182325,
      "step": 3677,
      "time_per_iteration": 3.705359935760498
    },
    {
      "auxiliary_loss_clip": 0.0116913,
      "auxiliary_loss_mlp": 0.01033282,
      "balance_loss_clip": 1.04970789,
      "balance_loss_mlp": 1.02457333,
      "epoch": 0.44225335177057656,
      "flos": 19204667758080.0,
      "grad_norm": 2.012377098272832,
      "language_loss": 0.85664546,
      "learning_rate": 2.466928600053209e-06,
      "loss": 0.8786695,
      "num_input_tokens_seen": 79197630,
      "step": 3678,
      "time_per_iteration": 3.6749086380004883
    },
    {
      "auxiliary_loss_clip": 0.01158071,
      "auxiliary_loss_mlp": 0.01028803,
      "balance_loss_clip": 1.04856491,
      "balance_loss_mlp": 1.02070832,
      "epoch": 0.4423735946612157,
      "flos": 23471321898240.0,
      "grad_norm": 1.916292391431073,
      "language_loss": 0.71573639,
      "learning_rate": 2.466171119598818e-06,
      "loss": 0.73760521,
      "num_input_tokens_seen": 79217600,
      "step": 3679,
      "time_per_iteration": 2.737898826599121
    },
    {
      "auxiliary_loss_clip": 0.01180921,
      "auxiliary_loss_mlp": 0.01028998,
      "balance_loss_clip": 1.05173624,
      "balance_loss_mlp": 1.01968813,
      "epoch": 0.44249383755185473,
      "flos": 26685398868480.0,
      "grad_norm": 1.885700049285115,
      "language_loss": 0.77584296,
      "learning_rate": 2.465413568426639e-06,
      "loss": 0.79794216,
      "num_input_tokens_seen": 79238550,
      "step": 3680,
      "time_per_iteration": 2.6661536693573
    },
    {
      "auxiliary_loss_clip": 0.01167261,
      "auxiliary_loss_mlp": 0.01030547,
      "balance_loss_clip": 1.05076897,
      "balance_loss_mlp": 1.02236927,
      "epoch": 0.44261408044249384,
      "flos": 23147659422720.0,
      "grad_norm": 1.572114119895612,
      "language_loss": 0.81549835,
      "learning_rate": 2.464655946651591e-06,
      "loss": 0.83747643,
      "num_input_tokens_seen": 79257555,
      "step": 3681,
      "time_per_iteration": 2.660778760910034
    },
    {
      "auxiliary_loss_clip": 0.01174918,
      "auxiliary_loss_mlp": 0.01031637,
      "balance_loss_clip": 1.05306196,
      "balance_loss_mlp": 1.02259481,
      "epoch": 0.44273432333313295,
      "flos": 24462564595200.0,
      "grad_norm": 1.8654908419979632,
      "language_loss": 0.80994827,
      "learning_rate": 2.4638982543886065e-06,
      "loss": 0.83201379,
      "num_input_tokens_seen": 79277595,
      "step": 3682,
      "time_per_iteration": 4.500851154327393
    },
    {
      "auxiliary_loss_clip": 0.01180309,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.05500889,
      "balance_loss_mlp": 1.02247238,
      "epoch": 0.442854566223772,
      "flos": 17528932512000.0,
      "grad_norm": 2.7293350687141738,
      "language_loss": 0.87106341,
      "learning_rate": 2.4631404917526254e-06,
      "loss": 0.8931815,
      "num_input_tokens_seen": 79294550,
      "step": 3683,
      "time_per_iteration": 2.5919923782348633
    },
    {
      "auxiliary_loss_clip": 0.0116482,
      "auxiliary_loss_mlp": 0.01035407,
      "balance_loss_clip": 1.04769373,
      "balance_loss_mlp": 1.02691913,
      "epoch": 0.4429748091144111,
      "flos": 24896293320960.0,
      "grad_norm": 1.50226329519889,
      "language_loss": 0.79706109,
      "learning_rate": 2.4623826588586e-06,
      "loss": 0.81906331,
      "num_input_tokens_seen": 79314820,
      "step": 3684,
      "time_per_iteration": 2.709543228149414
    },
    {
      "auxiliary_loss_clip": 0.01154547,
      "auxiliary_loss_mlp": 0.01029539,
      "balance_loss_clip": 1.0464673,
      "balance_loss_mlp": 1.02089,
      "epoch": 0.4430950520050502,
      "flos": 21614704738560.0,
      "grad_norm": 1.8300631747232492,
      "language_loss": 0.83053005,
      "learning_rate": 2.461624755821492e-06,
      "loss": 0.85237086,
      "num_input_tokens_seen": 79334300,
      "step": 3685,
      "time_per_iteration": 2.6639208793640137
    },
    {
      "auxiliary_loss_clip": 0.01150051,
      "auxiliary_loss_mlp": 0.01031635,
      "balance_loss_clip": 1.05169094,
      "balance_loss_mlp": 1.02252698,
      "epoch": 0.4432152948956893,
      "flos": 24572271709440.0,
      "grad_norm": 1.7863351606734668,
      "language_loss": 0.76662087,
      "learning_rate": 2.4608667827562763e-06,
      "loss": 0.78843772,
      "num_input_tokens_seen": 79353630,
      "step": 3686,
      "time_per_iteration": 2.727769613265991
    },
    {
      "auxiliary_loss_clip": 0.01181896,
      "auxiliary_loss_mlp": 0.01030952,
      "balance_loss_clip": 1.05472994,
      "balance_loss_mlp": 1.02230906,
      "epoch": 0.4433355377863284,
      "flos": 21762261809280.0,
      "grad_norm": 1.864729253474617,
      "language_loss": 0.90336102,
      "learning_rate": 2.460108739777936e-06,
      "loss": 0.92548954,
      "num_input_tokens_seen": 79372765,
      "step": 3687,
      "time_per_iteration": 2.6538448333740234
    },
    {
      "auxiliary_loss_clip": 0.01159539,
      "auxiliary_loss_mlp": 0.01035247,
      "balance_loss_clip": 1.05118704,
      "balance_loss_mlp": 1.02699792,
      "epoch": 0.44345578067696745,
      "flos": 20084479488000.0,
      "grad_norm": 1.9864796165648595,
      "language_loss": 0.76484847,
      "learning_rate": 2.4593506270014656e-06,
      "loss": 0.78679633,
      "num_input_tokens_seen": 79391735,
      "step": 3688,
      "time_per_iteration": 2.719886302947998
    },
    {
      "auxiliary_loss_clip": 0.0116022,
      "auxiliary_loss_mlp": 0.0103512,
      "balance_loss_clip": 1.04870105,
      "balance_loss_mlp": 1.02648926,
      "epoch": 0.44357602356760656,
      "flos": 24169497528960.0,
      "grad_norm": 1.4909973968486707,
      "language_loss": 0.81873405,
      "learning_rate": 2.45859244454187e-06,
      "loss": 0.84068745,
      "num_input_tokens_seen": 79411525,
      "step": 3689,
      "time_per_iteration": 2.784013509750366
    },
    {
      "auxiliary_loss_clip": 0.01170803,
      "auxiliary_loss_mlp": 0.01036745,
      "balance_loss_clip": 1.05049157,
      "balance_loss_mlp": 1.02842474,
      "epoch": 0.44369626645824567,
      "flos": 22707717644160.0,
      "grad_norm": 1.6195369754194644,
      "language_loss": 0.66389978,
      "learning_rate": 2.4578341925141655e-06,
      "loss": 0.68597531,
      "num_input_tokens_seen": 79430740,
      "step": 3690,
      "time_per_iteration": 2.6220197677612305
    },
    {
      "auxiliary_loss_clip": 0.0118093,
      "auxiliary_loss_mlp": 0.0103164,
      "balance_loss_clip": 1.05249512,
      "balance_loss_mlp": 1.02252054,
      "epoch": 0.4438165093488847,
      "flos": 38030225420160.0,
      "grad_norm": 1.9580520902409013,
      "language_loss": 0.72231179,
      "learning_rate": 2.457075871033378e-06,
      "loss": 0.74443746,
      "num_input_tokens_seen": 79452615,
      "step": 3691,
      "time_per_iteration": 2.910701274871826
    },
    {
      "auxiliary_loss_clip": 0.01143171,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.04935503,
      "balance_loss_mlp": 1.02041149,
      "epoch": 0.44393675223952384,
      "flos": 15523213996800.0,
      "grad_norm": 2.8604348363891234,
      "language_loss": 0.88606292,
      "learning_rate": 2.4563174802145445e-06,
      "loss": 0.90777743,
      "num_input_tokens_seen": 79469865,
      "step": 3692,
      "time_per_iteration": 2.7159037590026855
    },
    {
      "auxiliary_loss_clip": 0.01059512,
      "auxiliary_loss_mlp": 0.01003329,
      "balance_loss_clip": 1.01301098,
      "balance_loss_mlp": 1.00161254,
      "epoch": 0.44405699513016295,
      "flos": 64574893779840.0,
      "grad_norm": 0.6356160822279932,
      "language_loss": 0.48600069,
      "learning_rate": 2.455559020172712e-06,
      "loss": 0.50662911,
      "num_input_tokens_seen": 79537220,
      "step": 3693,
      "time_per_iteration": 3.408446788787842
    },
    {
      "auxiliary_loss_clip": 0.01136671,
      "auxiliary_loss_mlp": 0.0102945,
      "balance_loss_clip": 1.05126572,
      "balance_loss_mlp": 1.02115858,
      "epoch": 0.444177238020802,
      "flos": 23987394552960.0,
      "grad_norm": 1.987684780337597,
      "language_loss": 0.89591467,
      "learning_rate": 2.4548004910229385e-06,
      "loss": 0.91757596,
      "num_input_tokens_seen": 79554795,
      "step": 3694,
      "time_per_iteration": 2.761852741241455
    },
    {
      "auxiliary_loss_clip": 0.01174287,
      "auxiliary_loss_mlp": 0.00763444,
      "balance_loss_clip": 1.05029845,
      "balance_loss_mlp": 1.00036967,
      "epoch": 0.4442974809114411,
      "flos": 22563069575040.0,
      "grad_norm": 2.3475126399626487,
      "language_loss": 0.86876768,
      "learning_rate": 2.4540418928802913e-06,
      "loss": 0.88814497,
      "num_input_tokens_seen": 79573530,
      "step": 3695,
      "time_per_iteration": 2.7402238845825195
    },
    {
      "auxiliary_loss_clip": 0.01156689,
      "auxiliary_loss_mlp": 0.01027985,
      "balance_loss_clip": 1.04920638,
      "balance_loss_mlp": 1.01934862,
      "epoch": 0.4444177238020802,
      "flos": 17675699483520.0,
      "grad_norm": 2.0796532868567263,
      "language_loss": 0.66405356,
      "learning_rate": 2.4532832258598506e-06,
      "loss": 0.68590027,
      "num_input_tokens_seen": 79591360,
      "step": 3696,
      "time_per_iteration": 2.7102975845336914
    },
    {
      "auxiliary_loss_clip": 0.01185443,
      "auxiliary_loss_mlp": 0.01033549,
      "balance_loss_clip": 1.05217063,
      "balance_loss_mlp": 1.02516818,
      "epoch": 0.4445379666927193,
      "flos": 28621594609920.0,
      "grad_norm": 1.6977154941058177,
      "language_loss": 0.80404425,
      "learning_rate": 2.4525244900767047e-06,
      "loss": 0.8262341,
      "num_input_tokens_seen": 79612175,
      "step": 3697,
      "time_per_iteration": 2.714984178543091
    },
    {
      "auxiliary_loss_clip": 0.01065645,
      "auxiliary_loss_mlp": 0.01003383,
      "balance_loss_clip": 1.01315904,
      "balance_loss_mlp": 1.00159502,
      "epoch": 0.4446582095833584,
      "flos": 70487370115200.0,
      "grad_norm": 0.7775214546804061,
      "language_loss": 0.60492277,
      "learning_rate": 2.4517656856459536e-06,
      "loss": 0.62561309,
      "num_input_tokens_seen": 79678020,
      "step": 3698,
      "time_per_iteration": 3.319962739944458
    },
    {
      "auxiliary_loss_clip": 0.01169995,
      "auxiliary_loss_mlp": 0.01024554,
      "balance_loss_clip": 1.04904008,
      "balance_loss_mlp": 1.01596534,
      "epoch": 0.4447784524739975,
      "flos": 26505199313280.0,
      "grad_norm": 1.7230341897139447,
      "language_loss": 0.67932975,
      "learning_rate": 2.4510068126827073e-06,
      "loss": 0.70127523,
      "num_input_tokens_seen": 79699020,
      "step": 3699,
      "time_per_iteration": 2.699629783630371
    },
    {
      "auxiliary_loss_clip": 0.01158015,
      "auxiliary_loss_mlp": 0.01027966,
      "balance_loss_clip": 1.04895747,
      "balance_loss_mlp": 1.01928139,
      "epoch": 0.44489869536463655,
      "flos": 11656209553920.0,
      "grad_norm": 2.2501128830946806,
      "language_loss": 0.8231985,
      "learning_rate": 2.450247871302086e-06,
      "loss": 0.84505832,
      "num_input_tokens_seen": 79716795,
      "step": 3700,
      "time_per_iteration": 2.7203481197357178
    },
    {
      "auxiliary_loss_clip": 0.01178945,
      "auxiliary_loss_mlp": 0.01024791,
      "balance_loss_clip": 1.05246186,
      "balance_loss_mlp": 1.01605904,
      "epoch": 0.44501893825527566,
      "flos": 20448469958400.0,
      "grad_norm": 2.359187019720124,
      "language_loss": 0.8346411,
      "learning_rate": 2.44948886161922e-06,
      "loss": 0.85667849,
      "num_input_tokens_seen": 79735810,
      "step": 3701,
      "time_per_iteration": 2.612943410873413
    },
    {
      "auxiliary_loss_clip": 0.01175846,
      "auxiliary_loss_mlp": 0.01022259,
      "balance_loss_clip": 1.0510366,
      "balance_loss_mlp": 1.01436424,
      "epoch": 0.4451391811459148,
      "flos": 18261079430400.0,
      "grad_norm": 1.5677676291463951,
      "language_loss": 0.85075092,
      "learning_rate": 2.4487297837492524e-06,
      "loss": 0.87273204,
      "num_input_tokens_seen": 79754975,
      "step": 3702,
      "time_per_iteration": 2.7005562782287598
    },
    {
      "auxiliary_loss_clip": 0.01143918,
      "auxiliary_loss_mlp": 0.01028414,
      "balance_loss_clip": 1.04835701,
      "balance_loss_mlp": 1.01918757,
      "epoch": 0.44525942403655383,
      "flos": 16910155895040.0,
      "grad_norm": 1.8663613967310655,
      "language_loss": 0.62158978,
      "learning_rate": 2.4479706378073323e-06,
      "loss": 0.64331311,
      "num_input_tokens_seen": 79773515,
      "step": 3703,
      "time_per_iteration": 3.5774528980255127
    },
    {
      "auxiliary_loss_clip": 0.01133187,
      "auxiliary_loss_mlp": 0.01034003,
      "balance_loss_clip": 1.0434041,
      "balance_loss_mlp": 1.02599835,
      "epoch": 0.44537966692719294,
      "flos": 23258838994560.0,
      "grad_norm": 1.5586169113191524,
      "language_loss": 0.83846021,
      "learning_rate": 2.447211423908623e-06,
      "loss": 0.8601321,
      "num_input_tokens_seen": 79793560,
      "step": 3704,
      "time_per_iteration": 3.669285535812378
    },
    {
      "auxiliary_loss_clip": 0.01172487,
      "auxiliary_loss_mlp": 0.01025424,
      "balance_loss_clip": 1.0484066,
      "balance_loss_mlp": 1.01778841,
      "epoch": 0.445499909817832,
      "flos": 21724160457600.0,
      "grad_norm": 2.1163097233922805,
      "language_loss": 0.75000155,
      "learning_rate": 2.4464521421682966e-06,
      "loss": 0.77198064,
      "num_input_tokens_seen": 79811150,
      "step": 3705,
      "time_per_iteration": 2.6374497413635254
    },
    {
      "auxiliary_loss_clip": 0.01168456,
      "auxiliary_loss_mlp": 0.01029374,
      "balance_loss_clip": 1.05050969,
      "balance_loss_mlp": 1.02135754,
      "epoch": 0.4456201527084711,
      "flos": 23987969170560.0,
      "grad_norm": 1.4311894096633222,
      "language_loss": 0.87644976,
      "learning_rate": 2.4456927927015345e-06,
      "loss": 0.89842808,
      "num_input_tokens_seen": 79832190,
      "step": 3706,
      "time_per_iteration": 2.667948007583618
    },
    {
      "auxiliary_loss_clip": 0.01165953,
      "auxiliary_loss_mlp": 0.010351,
      "balance_loss_clip": 1.05316877,
      "balance_loss_mlp": 1.02567649,
      "epoch": 0.4457403955991102,
      "flos": 18807065136000.0,
      "grad_norm": 5.9274872244321894,
      "language_loss": 0.76259941,
      "learning_rate": 2.4449333756235307e-06,
      "loss": 0.78460991,
      "num_input_tokens_seen": 79848905,
      "step": 3707,
      "time_per_iteration": 3.643660545349121
    },
    {
      "auxiliary_loss_clip": 0.01177827,
      "auxiliary_loss_mlp": 0.01033823,
      "balance_loss_clip": 1.05269468,
      "balance_loss_mlp": 1.02577627,
      "epoch": 0.4458606384897493,
      "flos": 19207756327680.0,
      "grad_norm": 2.2977586470834828,
      "language_loss": 0.78487551,
      "learning_rate": 2.4441738910494876e-06,
      "loss": 0.80699199,
      "num_input_tokens_seen": 79863640,
      "step": 3708,
      "time_per_iteration": 3.550788640975952
    },
    {
      "auxiliary_loss_clip": 0.01165506,
      "auxiliary_loss_mlp": 0.01031935,
      "balance_loss_clip": 1.04889178,
      "balance_loss_mlp": 1.02341175,
      "epoch": 0.4459808813803884,
      "flos": 21361283308800.0,
      "grad_norm": 1.7981430834814152,
      "language_loss": 0.82355124,
      "learning_rate": 2.4434143390946176e-06,
      "loss": 0.84552562,
      "num_input_tokens_seen": 79882450,
      "step": 3709,
      "time_per_iteration": 2.7174766063690186
    },
    {
      "auxiliary_loss_clip": 0.01140289,
      "auxiliary_loss_mlp": 0.01031614,
      "balance_loss_clip": 1.0476203,
      "balance_loss_mlp": 1.02347481,
      "epoch": 0.4461011242710275,
      "flos": 23288967527040.0,
      "grad_norm": 1.891224166622154,
      "language_loss": 0.85978144,
      "learning_rate": 2.4426547198741457e-06,
      "loss": 0.88150048,
      "num_input_tokens_seen": 79900655,
      "step": 3710,
      "time_per_iteration": 2.746715545654297
    },
    {
      "auxiliary_loss_clip": 0.01129613,
      "auxiliary_loss_mlp": 0.01026547,
      "balance_loss_clip": 1.04963946,
      "balance_loss_mlp": 1.01747489,
      "epoch": 0.44622136716166655,
      "flos": 20193001453440.0,
      "grad_norm": 2.061965096083455,
      "language_loss": 0.7463305,
      "learning_rate": 2.441895033503305e-06,
      "loss": 0.76789212,
      "num_input_tokens_seen": 79918575,
      "step": 3711,
      "time_per_iteration": 2.7268617153167725
    },
    {
      "auxiliary_loss_clip": 0.01173085,
      "auxiliary_loss_mlp": 0.01031049,
      "balance_loss_clip": 1.05151701,
      "balance_loss_mlp": 1.02192354,
      "epoch": 0.44634161005230566,
      "flos": 21283033530240.0,
      "grad_norm": 1.7561245368745524,
      "language_loss": 0.82227176,
      "learning_rate": 2.4411352800973375e-06,
      "loss": 0.84431314,
      "num_input_tokens_seen": 79937010,
      "step": 3712,
      "time_per_iteration": 2.6689631938934326
    },
    {
      "auxiliary_loss_clip": 0.01138439,
      "auxiliary_loss_mlp": 0.01031024,
      "balance_loss_clip": 1.04519475,
      "balance_loss_mlp": 1.02263188,
      "epoch": 0.44646185294294477,
      "flos": 22929358515840.0,
      "grad_norm": 2.64742078489163,
      "language_loss": 0.7499373,
      "learning_rate": 2.4403754597715005e-06,
      "loss": 0.77163196,
      "num_input_tokens_seen": 79956455,
      "step": 3713,
      "time_per_iteration": 2.7410027980804443
    },
    {
      "auxiliary_loss_clip": 0.01160785,
      "auxiliary_loss_mlp": 0.01027201,
      "balance_loss_clip": 1.04634607,
      "balance_loss_mlp": 1.01799798,
      "epoch": 0.4465820958335838,
      "flos": 22637692080000.0,
      "grad_norm": 2.3638893793753826,
      "language_loss": 0.92977023,
      "learning_rate": 2.4396155726410553e-06,
      "loss": 0.95165014,
      "num_input_tokens_seen": 79975065,
      "step": 3714,
      "time_per_iteration": 2.687425374984741
    },
    {
      "auxiliary_loss_clip": 0.01177761,
      "auxiliary_loss_mlp": 0.01028566,
      "balance_loss_clip": 1.04959607,
      "balance_loss_mlp": 1.01993477,
      "epoch": 0.44670233872422294,
      "flos": 22672525294080.0,
      "grad_norm": 2.8461451588034317,
      "language_loss": 0.90624136,
      "learning_rate": 2.438855618821278e-06,
      "loss": 0.92830455,
      "num_input_tokens_seen": 79990865,
      "step": 3715,
      "time_per_iteration": 2.64646315574646
    },
    {
      "auxiliary_loss_clip": 0.0116278,
      "auxiliary_loss_mlp": 0.01027242,
      "balance_loss_clip": 1.04642415,
      "balance_loss_mlp": 1.01849198,
      "epoch": 0.44682258161486205,
      "flos": 23582178247680.0,
      "grad_norm": 1.6684953625130445,
      "language_loss": 0.66909754,
      "learning_rate": 2.4380955984274517e-06,
      "loss": 0.69099778,
      "num_input_tokens_seen": 80009520,
      "step": 3716,
      "time_per_iteration": 2.625842332839966
    },
    {
      "auxiliary_loss_clip": 0.01170924,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.04930568,
      "balance_loss_mlp": 1.019292,
      "epoch": 0.4469428245055011,
      "flos": 26501356558080.0,
      "grad_norm": 1.979159249626376,
      "language_loss": 0.76599419,
      "learning_rate": 2.4373355115748716e-06,
      "loss": 0.78798568,
      "num_input_tokens_seen": 80030350,
      "step": 3717,
      "time_per_iteration": 2.687736749649048
    },
    {
      "auxiliary_loss_clip": 0.01150266,
      "auxiliary_loss_mlp": 0.01029483,
      "balance_loss_clip": 1.04863071,
      "balance_loss_mlp": 1.0207808,
      "epoch": 0.4470630673961402,
      "flos": 21504925797120.0,
      "grad_norm": 1.5975646144150326,
      "language_loss": 0.72299576,
      "learning_rate": 2.436575358378842e-06,
      "loss": 0.7447933,
      "num_input_tokens_seen": 80049840,
      "step": 3718,
      "time_per_iteration": 2.679807424545288
    },
    {
      "auxiliary_loss_clip": 0.01163651,
      "auxiliary_loss_mlp": 0.01033644,
      "balance_loss_clip": 1.05053985,
      "balance_loss_mlp": 1.02523077,
      "epoch": 0.44718331028677927,
      "flos": 16173986653440.0,
      "grad_norm": 2.9096758162404157,
      "language_loss": 0.82978076,
      "learning_rate": 2.4358151389546782e-06,
      "loss": 0.85175371,
      "num_input_tokens_seen": 80066525,
      "step": 3719,
      "time_per_iteration": 2.8122947216033936
    },
    {
      "auxiliary_loss_clip": 0.01187144,
      "auxiliary_loss_mlp": 0.01032401,
      "balance_loss_clip": 1.05275917,
      "balance_loss_mlp": 1.02421701,
      "epoch": 0.4473035531774184,
      "flos": 19681238430720.0,
      "grad_norm": 4.630096588913423,
      "language_loss": 0.7538864,
      "learning_rate": 2.4350548534177035e-06,
      "loss": 0.77608186,
      "num_input_tokens_seen": 80083355,
      "step": 3720,
      "time_per_iteration": 2.6823770999908447
    },
    {
      "auxiliary_loss_clip": 0.01144634,
      "auxiliary_loss_mlp": 0.01032089,
      "balance_loss_clip": 1.04712725,
      "balance_loss_mlp": 1.02441764,
      "epoch": 0.4474237960680575,
      "flos": 41427590515200.0,
      "grad_norm": 1.6168103022164773,
      "language_loss": 0.66631538,
      "learning_rate": 2.434294501883254e-06,
      "loss": 0.68808258,
      "num_input_tokens_seen": 80106450,
      "step": 3721,
      "time_per_iteration": 2.9095029830932617
    },
    {
      "auxiliary_loss_clip": 0.01146952,
      "auxiliary_loss_mlp": 0.01028793,
      "balance_loss_clip": 1.04503858,
      "balance_loss_mlp": 1.020329,
      "epoch": 0.44754403895869654,
      "flos": 22891328991360.0,
      "grad_norm": 1.998662527345902,
      "language_loss": 0.65730309,
      "learning_rate": 2.433534084466674e-06,
      "loss": 0.67906052,
      "num_input_tokens_seen": 80125670,
      "step": 3722,
      "time_per_iteration": 2.7547476291656494
    },
    {
      "auxiliary_loss_clip": 0.01184137,
      "auxiliary_loss_mlp": 0.01028909,
      "balance_loss_clip": 1.05067444,
      "balance_loss_mlp": 1.02101767,
      "epoch": 0.44766428184933565,
      "flos": 25630271832960.0,
      "grad_norm": 1.780718511377368,
      "language_loss": 0.70666313,
      "learning_rate": 2.4327736012833178e-06,
      "loss": 0.72879356,
      "num_input_tokens_seen": 80147390,
      "step": 3723,
      "time_per_iteration": 2.653913974761963
    },
    {
      "auxiliary_loss_clip": 0.01171351,
      "auxiliary_loss_mlp": 0.01026896,
      "balance_loss_clip": 1.05061448,
      "balance_loss_mlp": 1.01832199,
      "epoch": 0.44778452473997477,
      "flos": 20448972748800.0,
      "grad_norm": 2.1773636712034814,
      "language_loss": 0.76705116,
      "learning_rate": 2.4320130524485506e-06,
      "loss": 0.78903359,
      "num_input_tokens_seen": 80166185,
      "step": 3724,
      "time_per_iteration": 2.678165912628174
    },
    {
      "auxiliary_loss_clip": 0.01150326,
      "auxiliary_loss_mlp": 0.01030189,
      "balance_loss_clip": 1.04945385,
      "balance_loss_mlp": 1.0222261,
      "epoch": 0.4479047676306138,
      "flos": 21975462984960.0,
      "grad_norm": 1.4793607652058478,
      "language_loss": 0.79697204,
      "learning_rate": 2.431252438077746e-06,
      "loss": 0.8187772,
      "num_input_tokens_seen": 80185685,
      "step": 3725,
      "time_per_iteration": 2.6977450847625732
    },
    {
      "auxiliary_loss_clip": 0.01176352,
      "auxiliary_loss_mlp": 0.00763426,
      "balance_loss_clip": 1.05042672,
      "balance_loss_mlp": 1.00031447,
      "epoch": 0.44802501052125293,
      "flos": 21467219495040.0,
      "grad_norm": 5.154584527475392,
      "language_loss": 0.77161348,
      "learning_rate": 2.4304917582862906e-06,
      "loss": 0.79101121,
      "num_input_tokens_seen": 80204865,
      "step": 3726,
      "time_per_iteration": 2.6747653484344482
    },
    {
      "auxiliary_loss_clip": 0.01187387,
      "auxiliary_loss_mlp": 0.01031923,
      "balance_loss_clip": 1.05290544,
      "balance_loss_mlp": 1.02413821,
      "epoch": 0.44814525341189204,
      "flos": 22126970551680.0,
      "grad_norm": 2.2023255207635164,
      "language_loss": 0.87729323,
      "learning_rate": 2.4297310131895774e-06,
      "loss": 0.8994863,
      "num_input_tokens_seen": 80223410,
      "step": 3727,
      "time_per_iteration": 2.581878185272217
    },
    {
      "auxiliary_loss_clip": 0.01169804,
      "auxiliary_loss_mlp": 0.01025962,
      "balance_loss_clip": 1.04938126,
      "balance_loss_mlp": 1.01784325,
      "epoch": 0.4482654963025311,
      "flos": 16653933204480.0,
      "grad_norm": 13.841496533601566,
      "language_loss": 0.74719018,
      "learning_rate": 2.4289702029030113e-06,
      "loss": 0.76914787,
      "num_input_tokens_seen": 80240880,
      "step": 3728,
      "time_per_iteration": 3.5480575561523438
    },
    {
      "auxiliary_loss_clip": 0.01175639,
      "auxiliary_loss_mlp": 0.01029412,
      "balance_loss_clip": 1.0536375,
      "balance_loss_mlp": 1.02147257,
      "epoch": 0.4483857391931702,
      "flos": 18841251905280.0,
      "grad_norm": 1.9264508614778388,
      "language_loss": 0.83130598,
      "learning_rate": 2.4282093275420057e-06,
      "loss": 0.85335648,
      "num_input_tokens_seen": 80259910,
      "step": 3729,
      "time_per_iteration": 2.671405792236328
    },
    {
      "auxiliary_loss_clip": 0.01177451,
      "auxiliary_loss_mlp": 0.01021788,
      "balance_loss_clip": 1.0516361,
      "balance_loss_mlp": 1.01355624,
      "epoch": 0.4485059820838093,
      "flos": 20372590477440.0,
      "grad_norm": 2.578670321438649,
      "language_loss": 0.70287108,
      "learning_rate": 2.4274483872219863e-06,
      "loss": 0.72486347,
      "num_input_tokens_seen": 80277270,
      "step": 3730,
      "time_per_iteration": 3.5934853553771973
    },
    {
      "auxiliary_loss_clip": 0.011696,
      "auxiliary_loss_mlp": 0.01027589,
      "balance_loss_clip": 1.04996026,
      "balance_loss_mlp": 1.01945853,
      "epoch": 0.4486262249744484,
      "flos": 20047742853120.0,
      "grad_norm": 2.061967806418435,
      "language_loss": 0.93652701,
      "learning_rate": 2.426687382058386e-06,
      "loss": 0.9584989,
      "num_input_tokens_seen": 80295550,
      "step": 3731,
      "time_per_iteration": 2.7246763706207275
    },
    {
      "auxiliary_loss_clip": 0.0106748,
      "auxiliary_loss_mlp": 0.01004634,
      "balance_loss_clip": 1.01476359,
      "balance_loss_mlp": 1.00291753,
      "epoch": 0.4487464678650875,
      "flos": 64595684776320.0,
      "grad_norm": 0.8673221143331336,
      "language_loss": 0.59838605,
      "learning_rate": 2.425926312166649e-06,
      "loss": 0.61910713,
      "num_input_tokens_seen": 80348425,
      "step": 3732,
      "time_per_iteration": 3.1514012813568115
    },
    {
      "auxiliary_loss_clip": 0.01163324,
      "auxiliary_loss_mlp": 0.01033687,
      "balance_loss_clip": 1.05027199,
      "balance_loss_mlp": 1.02499115,
      "epoch": 0.4488667107557266,
      "flos": 20769798049920.0,
      "grad_norm": 2.6126996734870844,
      "language_loss": 0.73397666,
      "learning_rate": 2.42516517766223e-06,
      "loss": 0.75594676,
      "num_input_tokens_seen": 80366505,
      "step": 3733,
      "time_per_iteration": 3.7910943031311035
    },
    {
      "auxiliary_loss_clip": 0.01189566,
      "auxiliary_loss_mlp": 0.01027154,
      "balance_loss_clip": 1.05509162,
      "balance_loss_mlp": 1.01897597,
      "epoch": 0.44898695364636565,
      "flos": 23951735326080.0,
      "grad_norm": 2.039417551397409,
      "language_loss": 0.67824233,
      "learning_rate": 2.4244039786605907e-06,
      "loss": 0.70040953,
      "num_input_tokens_seen": 80387510,
      "step": 3734,
      "time_per_iteration": 3.6253175735473633
    },
    {
      "auxiliary_loss_clip": 0.0112732,
      "auxiliary_loss_mlp": 0.0103317,
      "balance_loss_clip": 1.0429697,
      "balance_loss_mlp": 1.02465212,
      "epoch": 0.44910719653700476,
      "flos": 18624351628800.0,
      "grad_norm": 2.4740941299298798,
      "language_loss": 0.83188832,
      "learning_rate": 2.4236427152772055e-06,
      "loss": 0.85349321,
      "num_input_tokens_seen": 80405915,
      "step": 3735,
      "time_per_iteration": 2.7283754348754883
    },
    {
      "auxiliary_loss_clip": 0.01037301,
      "auxiliary_loss_mlp": 0.01002749,
      "balance_loss_clip": 1.01436007,
      "balance_loss_mlp": 1.00115168,
      "epoch": 0.4492274394276438,
      "flos": 57033435749760.0,
      "grad_norm": 0.838642635392877,
      "language_loss": 0.57317585,
      "learning_rate": 2.422881387627557e-06,
      "loss": 0.59357631,
      "num_input_tokens_seen": 80458365,
      "step": 3736,
      "time_per_iteration": 3.1052865982055664
    },
    {
      "auxiliary_loss_clip": 0.01158084,
      "auxiliary_loss_mlp": 0.01029693,
      "balance_loss_clip": 1.0480963,
      "balance_loss_mlp": 1.02102017,
      "epoch": 0.4493476823182829,
      "flos": 23254888498560.0,
      "grad_norm": 1.803218869283669,
      "language_loss": 0.77334529,
      "learning_rate": 2.422119995827139e-06,
      "loss": 0.79522306,
      "num_input_tokens_seen": 80478490,
      "step": 3737,
      "time_per_iteration": 2.6839122772216797
    },
    {
      "auxiliary_loss_clip": 0.01177546,
      "auxiliary_loss_mlp": 0.01030186,
      "balance_loss_clip": 1.05303955,
      "balance_loss_mlp": 1.02171016,
      "epoch": 0.44946792520892204,
      "flos": 15815131827840.0,
      "grad_norm": 3.386947649327146,
      "language_loss": 0.74842393,
      "learning_rate": 2.4213585399914528e-06,
      "loss": 0.77050126,
      "num_input_tokens_seen": 80495695,
      "step": 3738,
      "time_per_iteration": 2.662548065185547
    },
    {
      "auxiliary_loss_clip": 0.01172289,
      "auxiliary_loss_mlp": 0.01027591,
      "balance_loss_clip": 1.05266726,
      "balance_loss_mlp": 1.01979494,
      "epoch": 0.4495881680995611,
      "flos": 19610063631360.0,
      "grad_norm": 1.7531185098568636,
      "language_loss": 0.85532647,
      "learning_rate": 2.4205970202360113e-06,
      "loss": 0.87732524,
      "num_input_tokens_seen": 80515260,
      "step": 3739,
      "time_per_iteration": 2.6344974040985107
    },
    {
      "auxiliary_loss_clip": 0.01120667,
      "auxiliary_loss_mlp": 0.01025689,
      "balance_loss_clip": 1.04503083,
      "balance_loss_mlp": 1.01710558,
      "epoch": 0.4497084109902002,
      "flos": 26031465815040.0,
      "grad_norm": 1.9340890559043171,
      "language_loss": 0.78204107,
      "learning_rate": 2.4198354366763354e-06,
      "loss": 0.80350459,
      "num_input_tokens_seen": 80533900,
      "step": 3740,
      "time_per_iteration": 2.8254148960113525
    },
    {
      "auxiliary_loss_clip": 0.0116149,
      "auxiliary_loss_mlp": 0.01028992,
      "balance_loss_clip": 1.05013728,
      "balance_loss_mlp": 1.02148211,
      "epoch": 0.4498286538808393,
      "flos": 14793688771200.0,
      "grad_norm": 1.9243048671192629,
      "language_loss": 0.78278852,
      "learning_rate": 2.4190737894279587e-06,
      "loss": 0.80469334,
      "num_input_tokens_seen": 80551270,
      "step": 3741,
      "time_per_iteration": 2.7033510208129883
    },
    {
      "auxiliary_loss_clip": 0.01130105,
      "auxiliary_loss_mlp": 0.01032654,
      "balance_loss_clip": 1.04157627,
      "balance_loss_mlp": 1.02382028,
      "epoch": 0.44994889677147837,
      "flos": 15450171690240.0,
      "grad_norm": 2.012972342179157,
      "language_loss": 0.80854881,
      "learning_rate": 2.4183120786064203e-06,
      "loss": 0.83017641,
      "num_input_tokens_seen": 80568145,
      "step": 3742,
      "time_per_iteration": 2.6959125995635986
    },
    {
      "auxiliary_loss_clip": 0.01171189,
      "auxiliary_loss_mlp": 0.00763163,
      "balance_loss_clip": 1.05054927,
      "balance_loss_mlp": 1.00026524,
      "epoch": 0.4500691396621175,
      "flos": 21798316085760.0,
      "grad_norm": 2.749750298598797,
      "language_loss": 0.8568306,
      "learning_rate": 2.417550304327273e-06,
      "loss": 0.87617409,
      "num_input_tokens_seen": 80586185,
      "step": 3743,
      "time_per_iteration": 2.7928082942962646
    },
    {
      "auxiliary_loss_clip": 0.01189378,
      "auxiliary_loss_mlp": 0.01033277,
      "balance_loss_clip": 1.05354571,
      "balance_loss_mlp": 1.02394271,
      "epoch": 0.4501893825527566,
      "flos": 32382016421760.0,
      "grad_norm": 1.587820998816688,
      "language_loss": 0.75999749,
      "learning_rate": 2.4167884667060763e-06,
      "loss": 0.78222406,
      "num_input_tokens_seen": 80608895,
      "step": 3744,
      "time_per_iteration": 2.6924309730529785
    },
    {
      "auxiliary_loss_clip": 0.01158899,
      "auxiliary_loss_mlp": 0.01031549,
      "balance_loss_clip": 1.04904962,
      "balance_loss_mlp": 1.02303112,
      "epoch": 0.45030962544339564,
      "flos": 16544944362240.0,
      "grad_norm": 3.290643060498265,
      "language_loss": 0.87137222,
      "learning_rate": 2.4160265658584e-06,
      "loss": 0.89327663,
      "num_input_tokens_seen": 80623785,
      "step": 3745,
      "time_per_iteration": 2.659545660018921
    },
    {
      "auxiliary_loss_clip": 0.0117399,
      "auxiliary_loss_mlp": 0.01035937,
      "balance_loss_clip": 1.05082202,
      "balance_loss_mlp": 1.02771103,
      "epoch": 0.45042986833403476,
      "flos": 19573039687680.0,
      "grad_norm": 2.295046646712436,
      "language_loss": 0.68435204,
      "learning_rate": 2.4152646018998253e-06,
      "loss": 0.7064513,
      "num_input_tokens_seen": 80642735,
      "step": 3746,
      "time_per_iteration": 2.6366541385650635
    },
    {
      "auxiliary_loss_clip": 0.01168646,
      "auxiliary_loss_mlp": 0.01032356,
      "balance_loss_clip": 1.05047381,
      "balance_loss_mlp": 1.02414203,
      "epoch": 0.45055011122467387,
      "flos": 23112467072640.0,
      "grad_norm": 1.6962276933528193,
      "language_loss": 0.71977341,
      "learning_rate": 2.4145025749459403e-06,
      "loss": 0.74178344,
      "num_input_tokens_seen": 80663760,
      "step": 3747,
      "time_per_iteration": 2.7313196659088135
    },
    {
      "auxiliary_loss_clip": 0.01100632,
      "auxiliary_loss_mlp": 0.01027753,
      "balance_loss_clip": 1.04455078,
      "balance_loss_mlp": 1.01908016,
      "epoch": 0.4506703541153129,
      "flos": 19934623946880.0,
      "grad_norm": 2.31288153775069,
      "language_loss": 0.69944036,
      "learning_rate": 2.413740485112344e-06,
      "loss": 0.72072423,
      "num_input_tokens_seen": 80682100,
      "step": 3748,
      "time_per_iteration": 2.8301150798797607
    },
    {
      "auxiliary_loss_clip": 0.01151673,
      "auxiliary_loss_mlp": 0.01025586,
      "balance_loss_clip": 1.05000997,
      "balance_loss_mlp": 1.01782596,
      "epoch": 0.45079059700595203,
      "flos": 19499530504320.0,
      "grad_norm": 1.7311697912415969,
      "language_loss": 0.8192293,
      "learning_rate": 2.412978332514646e-06,
      "loss": 0.84100193,
      "num_input_tokens_seen": 80700880,
      "step": 3749,
      "time_per_iteration": 2.8018741607666016
    },
    {
      "auxiliary_loss_clip": 0.01159631,
      "auxiliary_loss_mlp": 0.01033265,
      "balance_loss_clip": 1.05004287,
      "balance_loss_mlp": 1.02415133,
      "epoch": 0.4509108398965911,
      "flos": 27636313570560.0,
      "grad_norm": 2.1296140080458223,
      "language_loss": 0.72336721,
      "learning_rate": 2.4122161172684623e-06,
      "loss": 0.74529612,
      "num_input_tokens_seen": 80721675,
      "step": 3750,
      "time_per_iteration": 2.841596841812134
    },
    {
      "auxiliary_loss_clip": 0.01158688,
      "auxiliary_loss_mlp": 0.01031077,
      "balance_loss_clip": 1.0487622,
      "balance_loss_mlp": 1.0226723,
      "epoch": 0.4510310827872302,
      "flos": 20995712640000.0,
      "grad_norm": 2.087078154673364,
      "language_loss": 0.84000063,
      "learning_rate": 2.4114538394894216e-06,
      "loss": 0.8618983,
      "num_input_tokens_seen": 80739315,
      "step": 3751,
      "time_per_iteration": 2.650799036026001
    },
    {
      "auxiliary_loss_clip": 0.01153704,
      "auxiliary_loss_mlp": 0.01026303,
      "balance_loss_clip": 1.0439868,
      "balance_loss_mlp": 1.01878715,
      "epoch": 0.4511513256778693,
      "flos": 16216684945920.0,
      "grad_norm": 1.8182877763830234,
      "language_loss": 0.83383983,
      "learning_rate": 2.410691499293161e-06,
      "loss": 0.85563993,
      "num_input_tokens_seen": 80757470,
      "step": 3752,
      "time_per_iteration": 2.846543550491333
    },
    {
      "auxiliary_loss_clip": 0.0116847,
      "auxiliary_loss_mlp": 0.01026367,
      "balance_loss_clip": 1.04698825,
      "balance_loss_mlp": 1.01836765,
      "epoch": 0.45127156856850836,
      "flos": 25186702780800.0,
      "grad_norm": 1.6314070629931532,
      "language_loss": 0.74215221,
      "learning_rate": 2.409929096795326e-06,
      "loss": 0.76410061,
      "num_input_tokens_seen": 80777840,
      "step": 3753,
      "time_per_iteration": 2.6440107822418213
    },
    {
      "auxiliary_loss_clip": 0.01169447,
      "auxiliary_loss_mlp": 0.01030674,
      "balance_loss_clip": 1.04726541,
      "balance_loss_mlp": 1.02244282,
      "epoch": 0.4513918114591475,
      "flos": 20412523422720.0,
      "grad_norm": 2.604885988786167,
      "language_loss": 0.79139578,
      "learning_rate": 2.409166632111573e-06,
      "loss": 0.81339699,
      "num_input_tokens_seen": 80795975,
      "step": 3754,
      "time_per_iteration": 2.7049014568328857
    },
    {
      "auxiliary_loss_clip": 0.0117901,
      "auxiliary_loss_mlp": 0.01031577,
      "balance_loss_clip": 1.05141616,
      "balance_loss_mlp": 1.02280927,
      "epoch": 0.4515120543497866,
      "flos": 26648482665600.0,
      "grad_norm": 1.9902238550495577,
      "language_loss": 0.80229127,
      "learning_rate": 2.4084041053575674e-06,
      "loss": 0.82439721,
      "num_input_tokens_seen": 80815395,
      "step": 3755,
      "time_per_iteration": 3.633913040161133
    },
    {
      "auxiliary_loss_clip": 0.01161433,
      "auxiliary_loss_mlp": 0.01025101,
      "balance_loss_clip": 1.04907346,
      "balance_loss_mlp": 1.01661885,
      "epoch": 0.45163229724042564,
      "flos": 20595093275520.0,
      "grad_norm": 1.987712151582793,
      "language_loss": 0.72094715,
      "learning_rate": 2.4076415166489834e-06,
      "loss": 0.74281251,
      "num_input_tokens_seen": 80834805,
      "step": 3756,
      "time_per_iteration": 3.5977869033813477
    },
    {
      "auxiliary_loss_clip": 0.01130867,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.04537892,
      "balance_loss_mlp": 1.02006078,
      "epoch": 0.45175254013106475,
      "flos": 21689004021120.0,
      "grad_norm": 1.5522171604790427,
      "language_loss": 0.79136974,
      "learning_rate": 2.406878866101506e-06,
      "loss": 0.81295568,
      "num_input_tokens_seen": 80853770,
      "step": 3757,
      "time_per_iteration": 2.797980785369873
    },
    {
      "auxiliary_loss_clip": 0.01185493,
      "auxiliary_loss_mlp": 0.01026674,
      "balance_loss_clip": 1.05370235,
      "balance_loss_mlp": 1.01829314,
      "epoch": 0.45187278302170386,
      "flos": 18878850466560.0,
      "grad_norm": 1.9422172811799012,
      "language_loss": 0.78264439,
      "learning_rate": 2.4061161538308273e-06,
      "loss": 0.80476606,
      "num_input_tokens_seen": 80870615,
      "step": 3758,
      "time_per_iteration": 2.583890676498413
    },
    {
      "auxiliary_loss_clip": 0.01169048,
      "auxiliary_loss_mlp": 0.0102518,
      "balance_loss_clip": 1.05010903,
      "balance_loss_mlp": 1.01711559,
      "epoch": 0.4519930259123429,
      "flos": 18582479349120.0,
      "grad_norm": 1.949067143845132,
      "language_loss": 0.88996321,
      "learning_rate": 2.4053533799526523e-06,
      "loss": 0.91190553,
      "num_input_tokens_seen": 80886335,
      "step": 3759,
      "time_per_iteration": 3.513416290283203
    },
    {
      "auxiliary_loss_clip": 0.01153704,
      "auxiliary_loss_mlp": 0.01024502,
      "balance_loss_clip": 1.05002058,
      "balance_loss_mlp": 1.01693845,
      "epoch": 0.452113268802982,
      "flos": 25192377129600.0,
      "grad_norm": 1.670694566229579,
      "language_loss": 0.86217773,
      "learning_rate": 2.404590544582691e-06,
      "loss": 0.88395977,
      "num_input_tokens_seen": 80904570,
      "step": 3760,
      "time_per_iteration": 3.612510919570923
    },
    {
      "auxiliary_loss_clip": 0.01131098,
      "auxiliary_loss_mlp": 0.01029318,
      "balance_loss_clip": 1.0424881,
      "balance_loss_mlp": 1.02067566,
      "epoch": 0.45223351169362114,
      "flos": 39378922312320.0,
      "grad_norm": 1.7394359063219174,
      "language_loss": 0.81384605,
      "learning_rate": 2.403827647836666e-06,
      "loss": 0.83545017,
      "num_input_tokens_seen": 80925125,
      "step": 3761,
      "time_per_iteration": 2.891256809234619
    },
    {
      "auxiliary_loss_clip": 0.01184293,
      "auxiliary_loss_mlp": 0.01029846,
      "balance_loss_clip": 1.05051017,
      "balance_loss_mlp": 1.02185869,
      "epoch": 0.4523537545842602,
      "flos": 21582169994880.0,
      "grad_norm": 2.056226691111121,
      "language_loss": 0.69536531,
      "learning_rate": 2.4030646898303075e-06,
      "loss": 0.71750671,
      "num_input_tokens_seen": 80946615,
      "step": 3762,
      "time_per_iteration": 2.6750059127807617
    },
    {
      "auxiliary_loss_clip": 0.01157653,
      "auxiliary_loss_mlp": 0.01031308,
      "balance_loss_clip": 1.04560137,
      "balance_loss_mlp": 1.02302849,
      "epoch": 0.4524739974748993,
      "flos": 28439527547520.0,
      "grad_norm": 1.9973584037012408,
      "language_loss": 0.81807506,
      "learning_rate": 2.4023016706793566e-06,
      "loss": 0.83996463,
      "num_input_tokens_seen": 80966410,
      "step": 3763,
      "time_per_iteration": 2.7446541786193848
    },
    {
      "auxiliary_loss_clip": 0.01056773,
      "auxiliary_loss_mlp": 0.0100427,
      "balance_loss_clip": 1.01553297,
      "balance_loss_mlp": 1.00254142,
      "epoch": 0.4525942403655384,
      "flos": 61556492148480.0,
      "grad_norm": 0.7690677125990084,
      "language_loss": 0.5687325,
      "learning_rate": 2.401538590499561e-06,
      "loss": 0.58934295,
      "num_input_tokens_seen": 81026865,
      "step": 3764,
      "time_per_iteration": 3.3574702739715576
    },
    {
      "auxiliary_loss_clip": 0.01170493,
      "auxiliary_loss_mlp": 0.00763614,
      "balance_loss_clip": 1.04831064,
      "balance_loss_mlp": 1.00025606,
      "epoch": 0.45271448325617747,
      "flos": 27529838680320.0,
      "grad_norm": 2.1747407383060153,
      "language_loss": 0.71750152,
      "learning_rate": 2.400775449406682e-06,
      "loss": 0.73684263,
      "num_input_tokens_seen": 81050060,
      "step": 3765,
      "time_per_iteration": 2.763550043106079
    },
    {
      "auxiliary_loss_clip": 0.01167059,
      "auxiliary_loss_mlp": 0.01025208,
      "balance_loss_clip": 1.04585242,
      "balance_loss_mlp": 1.01785231,
      "epoch": 0.4528347261468166,
      "flos": 22452608275200.0,
      "grad_norm": 1.9879067354228466,
      "language_loss": 0.7315942,
      "learning_rate": 2.400012247516485e-06,
      "loss": 0.75351691,
      "num_input_tokens_seen": 81070625,
      "step": 3766,
      "time_per_iteration": 2.654012680053711
    },
    {
      "auxiliary_loss_clip": 0.01145021,
      "auxiliary_loss_mlp": 0.01033792,
      "balance_loss_clip": 1.0464232,
      "balance_loss_mlp": 1.02535772,
      "epoch": 0.45295496903745563,
      "flos": 21103875469440.0,
      "grad_norm": 2.0004437472374974,
      "language_loss": 0.90366805,
      "learning_rate": 2.3992489849447484e-06,
      "loss": 0.92545617,
      "num_input_tokens_seen": 81089080,
      "step": 3767,
      "time_per_iteration": 2.763418197631836
    },
    {
      "auxiliary_loss_clip": 0.01146622,
      "auxiliary_loss_mlp": 0.01026655,
      "balance_loss_clip": 1.04809022,
      "balance_loss_mlp": 1.01768482,
      "epoch": 0.45307521192809475,
      "flos": 23221168606080.0,
      "grad_norm": 1.6992379321269822,
      "language_loss": 0.79000545,
      "learning_rate": 2.3984856618072584e-06,
      "loss": 0.81173825,
      "num_input_tokens_seen": 81109115,
      "step": 3768,
      "time_per_iteration": 2.7231523990631104
    },
    {
      "auxiliary_loss_clip": 0.0114943,
      "auxiliary_loss_mlp": 0.01030762,
      "balance_loss_clip": 1.04883504,
      "balance_loss_mlp": 1.02281642,
      "epoch": 0.45319545481873386,
      "flos": 15560094286080.0,
      "grad_norm": 1.8501080292220489,
      "language_loss": 0.74324691,
      "learning_rate": 2.3977222782198098e-06,
      "loss": 0.76504886,
      "num_input_tokens_seen": 81127750,
      "step": 3769,
      "time_per_iteration": 2.7318825721740723
    },
    {
      "auxiliary_loss_clip": 0.01134542,
      "auxiliary_loss_mlp": 0.010254,
      "balance_loss_clip": 1.04494858,
      "balance_loss_mlp": 1.01701999,
      "epoch": 0.4533156977093729,
      "flos": 21944759834880.0,
      "grad_norm": 2.0209836197747557,
      "language_loss": 0.75196445,
      "learning_rate": 2.3969588342982077e-06,
      "loss": 0.77356386,
      "num_input_tokens_seen": 81147125,
      "step": 3770,
      "time_per_iteration": 2.7186527252197266
    },
    {
      "auxiliary_loss_clip": 0.01167133,
      "auxiliary_loss_mlp": 0.01028307,
      "balance_loss_clip": 1.04968131,
      "balance_loss_mlp": 1.02040362,
      "epoch": 0.453435940600012,
      "flos": 24242180699520.0,
      "grad_norm": 1.653628939923085,
      "language_loss": 0.72739148,
      "learning_rate": 2.396195330158267e-06,
      "loss": 0.74934596,
      "num_input_tokens_seen": 81167015,
      "step": 3771,
      "time_per_iteration": 2.752680778503418
    },
    {
      "auxiliary_loss_clip": 0.01184276,
      "auxiliary_loss_mlp": 0.01032995,
      "balance_loss_clip": 1.05028069,
      "balance_loss_mlp": 1.025038,
      "epoch": 0.45355618349065113,
      "flos": 23440367352960.0,
      "grad_norm": 1.7424126397020594,
      "language_loss": 0.79460514,
      "learning_rate": 2.3954317659158094e-06,
      "loss": 0.81677788,
      "num_input_tokens_seen": 81187350,
      "step": 3772,
      "time_per_iteration": 2.6147401332855225
    },
    {
      "auxiliary_loss_clip": 0.01084079,
      "auxiliary_loss_mlp": 0.0100222,
      "balance_loss_clip": 1.01596618,
      "balance_loss_mlp": 1.00043201,
      "epoch": 0.4536764263812902,
      "flos": 66903161448960.0,
      "grad_norm": 0.992214202620696,
      "language_loss": 0.56943893,
      "learning_rate": 2.394668141686667e-06,
      "loss": 0.59030193,
      "num_input_tokens_seen": 81249315,
      "step": 3773,
      "time_per_iteration": 3.224578380584717
    },
    {
      "auxiliary_loss_clip": 0.01166225,
      "auxiliary_loss_mlp": 0.01026493,
      "balance_loss_clip": 1.04785931,
      "balance_loss_mlp": 1.01885128,
      "epoch": 0.4537966692719293,
      "flos": 42739766254080.0,
      "grad_norm": 2.1126769203219555,
      "language_loss": 0.69490826,
      "learning_rate": 2.3939044575866813e-06,
      "loss": 0.71683544,
      "num_input_tokens_seen": 81272065,
      "step": 3774,
      "time_per_iteration": 2.8738088607788086
    },
    {
      "auxiliary_loss_clip": 0.01150539,
      "auxiliary_loss_mlp": 0.00763557,
      "balance_loss_clip": 1.0472641,
      "balance_loss_mlp": 1.00020862,
      "epoch": 0.4539169121625684,
      "flos": 35549480517120.0,
      "grad_norm": 2.163951829345864,
      "language_loss": 0.75248575,
      "learning_rate": 2.3931407137317024e-06,
      "loss": 0.77162671,
      "num_input_tokens_seen": 81292220,
      "step": 3775,
      "time_per_iteration": 2.761521100997925
    },
    {
      "auxiliary_loss_clip": 0.01138765,
      "auxiliary_loss_mlp": 0.01030612,
      "balance_loss_clip": 1.04337525,
      "balance_loss_mlp": 1.02215993,
      "epoch": 0.45403715505320746,
      "flos": 18514716341760.0,
      "grad_norm": 1.7245377508497888,
      "language_loss": 0.84790313,
      "learning_rate": 2.3923769102375907e-06,
      "loss": 0.86959696,
      "num_input_tokens_seen": 81311085,
      "step": 3776,
      "time_per_iteration": 2.742717742919922
    },
    {
      "auxiliary_loss_clip": 0.01141108,
      "auxiliary_loss_mlp": 0.01033443,
      "balance_loss_clip": 1.04467249,
      "balance_loss_mlp": 1.02455044,
      "epoch": 0.4541573979438466,
      "flos": 25045825639680.0,
      "grad_norm": 2.2899578580626097,
      "language_loss": 0.78621304,
      "learning_rate": 2.391613047220213e-06,
      "loss": 0.80795848,
      "num_input_tokens_seen": 81330985,
      "step": 3777,
      "time_per_iteration": 2.668560028076172
    },
    {
      "auxiliary_loss_clip": 0.01136545,
      "auxiliary_loss_mlp": 0.01031711,
      "balance_loss_clip": 1.04527807,
      "balance_loss_mlp": 1.02334237,
      "epoch": 0.4542776408344857,
      "flos": 18332397884160.0,
      "grad_norm": 1.7960918455177004,
      "language_loss": 0.78776139,
      "learning_rate": 2.390849124795447e-06,
      "loss": 0.80944395,
      "num_input_tokens_seen": 81346985,
      "step": 3778,
      "time_per_iteration": 2.6834897994995117
    },
    {
      "auxiliary_loss_clip": 0.0118587,
      "auxiliary_loss_mlp": 0.01029058,
      "balance_loss_clip": 1.05212522,
      "balance_loss_mlp": 1.02073753,
      "epoch": 0.45439788372512474,
      "flos": 20701173116160.0,
      "grad_norm": 2.011187735931306,
      "language_loss": 0.84064734,
      "learning_rate": 2.3900851430791804e-06,
      "loss": 0.8627966,
      "num_input_tokens_seen": 81365005,
      "step": 3779,
      "time_per_iteration": 2.636666774749756
    },
    {
      "auxiliary_loss_clip": 0.01187287,
      "auxiliary_loss_mlp": 0.01029054,
      "balance_loss_clip": 1.05045247,
      "balance_loss_mlp": 1.02007127,
      "epoch": 0.45451812661576385,
      "flos": 22309432663680.0,
      "grad_norm": 2.2579276623578455,
      "language_loss": 0.84659874,
      "learning_rate": 2.389321102187307e-06,
      "loss": 0.86876214,
      "num_input_tokens_seen": 81383785,
      "step": 3780,
      "time_per_iteration": 2.577648639678955
    },
    {
      "auxiliary_loss_clip": 0.01153412,
      "auxiliary_loss_mlp": 0.00763631,
      "balance_loss_clip": 1.04808378,
      "balance_loss_mlp": 1.00027728,
      "epoch": 0.4546383695064029,
      "flos": 21763303303680.0,
      "grad_norm": 2.035230610942664,
      "language_loss": 0.81760383,
      "learning_rate": 2.3885570022357326e-06,
      "loss": 0.83677423,
      "num_input_tokens_seen": 81402915,
      "step": 3781,
      "time_per_iteration": 3.4947707653045654
    },
    {
      "auxiliary_loss_clip": 0.01056896,
      "auxiliary_loss_mlp": 0.01002267,
      "balance_loss_clip": 1.01551509,
      "balance_loss_mlp": 1.00039566,
      "epoch": 0.454758612397042,
      "flos": 64242755694720.0,
      "grad_norm": 0.7968785258889399,
      "language_loss": 0.60880476,
      "learning_rate": 2.38779284334037e-06,
      "loss": 0.62939644,
      "num_input_tokens_seen": 81467890,
      "step": 3782,
      "time_per_iteration": 4.178065061569214
    },
    {
      "auxiliary_loss_clip": 0.01110697,
      "auxiliary_loss_mlp": 0.01024176,
      "balance_loss_clip": 1.04062593,
      "balance_loss_mlp": 1.01581323,
      "epoch": 0.4548788552876811,
      "flos": 27304175485440.0,
      "grad_norm": 2.1072543104814443,
      "language_loss": 0.7842083,
      "learning_rate": 2.387028625617141e-06,
      "loss": 0.80555701,
      "num_input_tokens_seen": 81487105,
      "step": 3783,
      "time_per_iteration": 2.7396762371063232
    },
    {
      "auxiliary_loss_clip": 0.01143669,
      "auxiliary_loss_mlp": 0.01030043,
      "balance_loss_clip": 1.0443058,
      "balance_loss_mlp": 1.02196038,
      "epoch": 0.4549990981783202,
      "flos": 22857142222080.0,
      "grad_norm": 1.9047748447225141,
      "language_loss": 0.84932172,
      "learning_rate": 2.3862643491819766e-06,
      "loss": 0.87105882,
      "num_input_tokens_seen": 81505670,
      "step": 3784,
      "time_per_iteration": 2.640009641647339
    },
    {
      "auxiliary_loss_clip": 0.01164894,
      "auxiliary_loss_mlp": 0.0102707,
      "balance_loss_clip": 1.04534793,
      "balance_loss_mlp": 1.0191431,
      "epoch": 0.4551193410689593,
      "flos": 23258587599360.0,
      "grad_norm": 1.6857143769581937,
      "language_loss": 0.84770703,
      "learning_rate": 2.3855000141508186e-06,
      "loss": 0.8696267,
      "num_input_tokens_seen": 81525825,
      "step": 3785,
      "time_per_iteration": 3.5934386253356934
    },
    {
      "auxiliary_loss_clip": 0.01159447,
      "auxiliary_loss_mlp": 0.01030636,
      "balance_loss_clip": 1.04959106,
      "balance_loss_mlp": 1.0224998,
      "epoch": 0.4552395839595984,
      "flos": 20777519473920.0,
      "grad_norm": 2.509303787151921,
      "language_loss": 0.84749997,
      "learning_rate": 2.3847356206396143e-06,
      "loss": 0.86940074,
      "num_input_tokens_seen": 81543135,
      "step": 3786,
      "time_per_iteration": 3.662792205810547
    },
    {
      "auxiliary_loss_clip": 0.01186679,
      "auxiliary_loss_mlp": 0.01028849,
      "balance_loss_clip": 1.05290186,
      "balance_loss_mlp": 1.02079618,
      "epoch": 0.45535982685023746,
      "flos": 23257510191360.0,
      "grad_norm": 1.6815497593397553,
      "language_loss": 0.78570557,
      "learning_rate": 2.3839711687643227e-06,
      "loss": 0.80786085,
      "num_input_tokens_seen": 81564360,
      "step": 3787,
      "time_per_iteration": 2.6299681663513184
    },
    {
      "auxiliary_loss_clip": 0.01171462,
      "auxiliary_loss_mlp": 0.01032496,
      "balance_loss_clip": 1.05115628,
      "balance_loss_mlp": 1.0236032,
      "epoch": 0.45548006974087657,
      "flos": 19646117907840.0,
      "grad_norm": 2.302172489667698,
      "language_loss": 0.74058402,
      "learning_rate": 2.38320665864091e-06,
      "loss": 0.76262367,
      "num_input_tokens_seen": 81583710,
      "step": 3788,
      "time_per_iteration": 2.6754024028778076
    },
    {
      "auxiliary_loss_clip": 0.01112396,
      "auxiliary_loss_mlp": 0.01032987,
      "balance_loss_clip": 1.03972554,
      "balance_loss_mlp": 1.02348578,
      "epoch": 0.4556003126315157,
      "flos": 20047778766720.0,
      "grad_norm": 2.1263410381049077,
      "language_loss": 0.81948066,
      "learning_rate": 2.3824420903853516e-06,
      "loss": 0.84093451,
      "num_input_tokens_seen": 81602175,
      "step": 3789,
      "time_per_iteration": 2.8183977603912354
    },
    {
      "auxiliary_loss_clip": 0.01172468,
      "auxiliary_loss_mlp": 0.01025056,
      "balance_loss_clip": 1.05202723,
      "balance_loss_mlp": 1.01692557,
      "epoch": 0.45572055552215474,
      "flos": 22959738443520.0,
      "grad_norm": 2.3375679170963157,
      "language_loss": 0.81669652,
      "learning_rate": 2.3816774641136324e-06,
      "loss": 0.83867168,
      "num_input_tokens_seen": 81619430,
      "step": 3790,
      "time_per_iteration": 2.65752911567688
    },
    {
      "auxiliary_loss_clip": 0.01167366,
      "auxiliary_loss_mlp": 0.00763283,
      "balance_loss_clip": 1.04864526,
      "balance_loss_mlp": 1.00016356,
      "epoch": 0.45584079841279385,
      "flos": 33109925535360.0,
      "grad_norm": 1.7943307482343498,
      "language_loss": 0.71138734,
      "learning_rate": 2.380912779941745e-06,
      "loss": 0.73069382,
      "num_input_tokens_seen": 81642550,
      "step": 3791,
      "time_per_iteration": 2.8226561546325684
    },
    {
      "auxiliary_loss_clip": 0.01168272,
      "auxiliary_loss_mlp": 0.01031665,
      "balance_loss_clip": 1.04428411,
      "balance_loss_mlp": 1.02267087,
      "epoch": 0.45596104130343296,
      "flos": 27272179445760.0,
      "grad_norm": 4.1790098498457455,
      "language_loss": 0.8354494,
      "learning_rate": 2.3801480379856918e-06,
      "loss": 0.85744876,
      "num_input_tokens_seen": 81664260,
      "step": 3792,
      "time_per_iteration": 2.7123584747314453
    },
    {
      "auxiliary_loss_clip": 0.01156652,
      "auxiliary_loss_mlp": 0.01030704,
      "balance_loss_clip": 1.05001116,
      "balance_loss_mlp": 1.02251387,
      "epoch": 0.456081284194072,
      "flos": 21579799697280.0,
      "grad_norm": 2.0471353402476136,
      "language_loss": 0.838588,
      "learning_rate": 2.379383238361484e-06,
      "loss": 0.86046153,
      "num_input_tokens_seen": 81683620,
      "step": 3793,
      "time_per_iteration": 2.7634336948394775
    },
    {
      "auxiliary_loss_clip": 0.01167659,
      "auxiliary_loss_mlp": 0.01024496,
      "balance_loss_clip": 1.04732227,
      "balance_loss_mlp": 1.01703644,
      "epoch": 0.4562015270847111,
      "flos": 35918822113920.0,
      "grad_norm": 2.198276031758361,
      "language_loss": 0.79751611,
      "learning_rate": 2.3786183811851407e-06,
      "loss": 0.81943762,
      "num_input_tokens_seen": 81704325,
      "step": 3794,
      "time_per_iteration": 2.8194656372070312
    },
    {
      "auxiliary_loss_clip": 0.01186235,
      "auxiliary_loss_mlp": 0.01035554,
      "balance_loss_clip": 1.05191207,
      "balance_loss_mlp": 1.02711368,
      "epoch": 0.45632176997535023,
      "flos": 13589783602560.0,
      "grad_norm": 1.8426128131300044,
      "language_loss": 0.80074048,
      "learning_rate": 2.3778534665726892e-06,
      "loss": 0.82295835,
      "num_input_tokens_seen": 81721155,
      "step": 3795,
      "time_per_iteration": 2.684628963470459
    },
    {
      "auxiliary_loss_clip": 0.01162906,
      "auxiliary_loss_mlp": 0.01038914,
      "balance_loss_clip": 1.05065536,
      "balance_loss_mlp": 1.03092396,
      "epoch": 0.4564420128659893,
      "flos": 32635401937920.0,
      "grad_norm": 1.857027331481611,
      "language_loss": 0.72885811,
      "learning_rate": 2.377088494640168e-06,
      "loss": 0.75087637,
      "num_input_tokens_seen": 81742905,
      "step": 3796,
      "time_per_iteration": 2.6828017234802246
    },
    {
      "auxiliary_loss_clip": 0.01166717,
      "auxiliary_loss_mlp": 0.01031179,
      "balance_loss_clip": 1.05035233,
      "balance_loss_mlp": 1.02307236,
      "epoch": 0.4565622557566284,
      "flos": 20377690208640.0,
      "grad_norm": 1.7908306434636296,
      "language_loss": 0.78147316,
      "learning_rate": 2.3763234655036216e-06,
      "loss": 0.80345213,
      "num_input_tokens_seen": 81762105,
      "step": 3797,
      "time_per_iteration": 2.69478440284729
    },
    {
      "auxiliary_loss_clip": 0.01139073,
      "auxiliary_loss_mlp": 0.01030467,
      "balance_loss_clip": 1.04252326,
      "balance_loss_mlp": 1.0226469,
      "epoch": 0.45668249864726745,
      "flos": 25374372364800.0,
      "grad_norm": 2.289305841498056,
      "language_loss": 0.86790657,
      "learning_rate": 2.3755583792791046e-06,
      "loss": 0.88960201,
      "num_input_tokens_seen": 81781975,
      "step": 3798,
      "time_per_iteration": 2.7829136848449707
    },
    {
      "auxiliary_loss_clip": 0.01173688,
      "auxiliary_loss_mlp": 0.01028805,
      "balance_loss_clip": 1.05053592,
      "balance_loss_mlp": 1.02033532,
      "epoch": 0.45680274153790656,
      "flos": 15559806977280.0,
      "grad_norm": 2.092698032288108,
      "language_loss": 0.74330807,
      "learning_rate": 2.3747932360826803e-06,
      "loss": 0.765333,
      "num_input_tokens_seen": 81798905,
      "step": 3799,
      "time_per_iteration": 2.655580520629883
    },
    {
      "auxiliary_loss_clip": 0.01170217,
      "auxiliary_loss_mlp": 0.0102724,
      "balance_loss_clip": 1.05076432,
      "balance_loss_mlp": 1.01944971,
      "epoch": 0.4569229844285457,
      "flos": 19792884879360.0,
      "grad_norm": 1.922964460493744,
      "language_loss": 0.82423472,
      "learning_rate": 2.3740280360304205e-06,
      "loss": 0.84620929,
      "num_input_tokens_seen": 81816630,
      "step": 3800,
      "time_per_iteration": 2.6316416263580322
    },
    {
      "auxiliary_loss_clip": 0.01140988,
      "auxiliary_loss_mlp": 0.01026472,
      "balance_loss_clip": 1.04856706,
      "balance_loss_mlp": 1.01876771,
      "epoch": 0.45704322731918473,
      "flos": 24093941270400.0,
      "grad_norm": 1.734900450550298,
      "language_loss": 0.68390501,
      "learning_rate": 2.3732627792384038e-06,
      "loss": 0.70557964,
      "num_input_tokens_seen": 81837700,
      "step": 3801,
      "time_per_iteration": 2.7590556144714355
    },
    {
      "auxiliary_loss_clip": 0.01185689,
      "auxiliary_loss_mlp": 0.0102643,
      "balance_loss_clip": 1.05093622,
      "balance_loss_mlp": 1.01813889,
      "epoch": 0.45716347020982384,
      "flos": 31317803245440.0,
      "grad_norm": 1.986594110911609,
      "language_loss": 0.75203609,
      "learning_rate": 2.3724974658227207e-06,
      "loss": 0.77415729,
      "num_input_tokens_seen": 81858490,
      "step": 3802,
      "time_per_iteration": 2.6969940662384033
    },
    {
      "auxiliary_loss_clip": 0.01154483,
      "auxiliary_loss_mlp": 0.00762988,
      "balance_loss_clip": 1.04831457,
      "balance_loss_mlp": 1.00019026,
      "epoch": 0.45728371310046295,
      "flos": 26501392471680.0,
      "grad_norm": 2.1154809202072733,
      "language_loss": 0.71372551,
      "learning_rate": 2.3717320958994687e-06,
      "loss": 0.73290026,
      "num_input_tokens_seen": 81876050,
      "step": 3803,
      "time_per_iteration": 2.707887887954712
    },
    {
      "auxiliary_loss_clip": 0.01141777,
      "auxiliary_loss_mlp": 0.01029328,
      "balance_loss_clip": 1.04199564,
      "balance_loss_mlp": 1.02083397,
      "epoch": 0.457403955991102,
      "flos": 17929408222080.0,
      "grad_norm": 1.948001932407079,
      "language_loss": 0.70452142,
      "learning_rate": 2.3709666695847534e-06,
      "loss": 0.72623247,
      "num_input_tokens_seen": 81894230,
      "step": 3804,
      "time_per_iteration": 2.7754030227661133
    },
    {
      "auxiliary_loss_clip": 0.01118759,
      "auxiliary_loss_mlp": 0.01030172,
      "balance_loss_clip": 1.04261088,
      "balance_loss_mlp": 1.0226438,
      "epoch": 0.4575241988817411,
      "flos": 42230660837760.0,
      "grad_norm": 1.6378909739955605,
      "language_loss": 0.70500433,
      "learning_rate": 2.370201186994689e-06,
      "loss": 0.7264936,
      "num_input_tokens_seen": 81917915,
      "step": 3805,
      "time_per_iteration": 2.866208076477051
    },
    {
      "auxiliary_loss_clip": 0.01148134,
      "auxiliary_loss_mlp": 0.01028549,
      "balance_loss_clip": 1.04751372,
      "balance_loss_mlp": 1.02023447,
      "epoch": 0.45764444177238023,
      "flos": 30117309868800.0,
      "grad_norm": 1.8554025236356495,
      "language_loss": 0.69894546,
      "learning_rate": 2.369435648245399e-06,
      "loss": 0.7207123,
      "num_input_tokens_seen": 81938130,
      "step": 3806,
      "time_per_iteration": 2.690476417541504
    },
    {
      "auxiliary_loss_clip": 0.0115639,
      "auxiliary_loss_mlp": 0.01038349,
      "balance_loss_clip": 1.04814303,
      "balance_loss_mlp": 1.03045142,
      "epoch": 0.4577646846630193,
      "flos": 24060293205120.0,
      "grad_norm": 1.729903981343227,
      "language_loss": 0.85032493,
      "learning_rate": 2.368670053453015e-06,
      "loss": 0.87227225,
      "num_input_tokens_seen": 81959820,
      "step": 3807,
      "time_per_iteration": 3.644468307495117
    },
    {
      "auxiliary_loss_clip": 0.01178751,
      "auxiliary_loss_mlp": 0.01023622,
      "balance_loss_clip": 1.05355835,
      "balance_loss_mlp": 1.01568866,
      "epoch": 0.4578849275536584,
      "flos": 17418578952960.0,
      "grad_norm": 2.2202937536825145,
      "language_loss": 0.74259067,
      "learning_rate": 2.3679044027336757e-06,
      "loss": 0.7646144,
      "num_input_tokens_seen": 81975710,
      "step": 3808,
      "time_per_iteration": 3.5734386444091797
    },
    {
      "auxiliary_loss_clip": 0.01186022,
      "auxiliary_loss_mlp": 0.01031171,
      "balance_loss_clip": 1.05068493,
      "balance_loss_mlp": 1.02243948,
      "epoch": 0.4580051704442975,
      "flos": 13510169107200.0,
      "grad_norm": 3.275086906572407,
      "language_loss": 0.69558609,
      "learning_rate": 2.3671386962035326e-06,
      "loss": 0.717758,
      "num_input_tokens_seen": 81993180,
      "step": 3809,
      "time_per_iteration": 2.689065456390381
    },
    {
      "auxiliary_loss_clip": 0.01172707,
      "auxiliary_loss_mlp": 0.01030941,
      "balance_loss_clip": 1.05056989,
      "balance_loss_mlp": 1.02258432,
      "epoch": 0.45812541333493656,
      "flos": 18037606965120.0,
      "grad_norm": 2.3782803002006583,
      "language_loss": 0.68889087,
      "learning_rate": 2.3663729339787405e-06,
      "loss": 0.71092737,
      "num_input_tokens_seen": 82010115,
      "step": 3810,
      "time_per_iteration": 2.638394594192505
    },
    {
      "auxiliary_loss_clip": 0.01186051,
      "auxiliary_loss_mlp": 0.01034181,
      "balance_loss_clip": 1.05175996,
      "balance_loss_mlp": 1.02534795,
      "epoch": 0.45824565622557567,
      "flos": 20222196232320.0,
      "grad_norm": 2.722942157169025,
      "language_loss": 0.73177552,
      "learning_rate": 2.365607116175466e-06,
      "loss": 0.75397778,
      "num_input_tokens_seen": 82025540,
      "step": 3811,
      "time_per_iteration": 3.522378444671631
    },
    {
      "auxiliary_loss_clip": 0.01184483,
      "auxiliary_loss_mlp": 0.01024535,
      "balance_loss_clip": 1.05075645,
      "balance_loss_mlp": 1.01651788,
      "epoch": 0.4583658991162148,
      "flos": 19864885691520.0,
      "grad_norm": 3.3254266495646956,
      "language_loss": 0.67179221,
      "learning_rate": 2.3648412429098825e-06,
      "loss": 0.69388241,
      "num_input_tokens_seen": 82043890,
      "step": 3812,
      "time_per_iteration": 3.56215238571167
    },
    {
      "auxiliary_loss_clip": 0.01137894,
      "auxiliary_loss_mlp": 0.0103124,
      "balance_loss_clip": 1.04700923,
      "balance_loss_mlp": 1.0216496,
      "epoch": 0.45848614200685384,
      "flos": 21029935322880.0,
      "grad_norm": 2.5449367748011387,
      "language_loss": 0.82080263,
      "learning_rate": 2.364075314298172e-06,
      "loss": 0.84249401,
      "num_input_tokens_seen": 82061345,
      "step": 3813,
      "time_per_iteration": 2.653594493865967
    },
    {
      "auxiliary_loss_clip": 0.01173935,
      "auxiliary_loss_mlp": 0.00763785,
      "balance_loss_clip": 1.04968166,
      "balance_loss_mlp": 1.00021076,
      "epoch": 0.45860638489749295,
      "flos": 21069293650560.0,
      "grad_norm": 1.9660841603347303,
      "language_loss": 0.69973695,
      "learning_rate": 2.3633093304565267e-06,
      "loss": 0.71911418,
      "num_input_tokens_seen": 82080400,
      "step": 3814,
      "time_per_iteration": 2.733041763305664
    },
    {
      "auxiliary_loss_clip": 0.01192118,
      "auxiliary_loss_mlp": 0.01028955,
      "balance_loss_clip": 1.05523968,
      "balance_loss_mlp": 1.02044892,
      "epoch": 0.458726627788132,
      "flos": 26833889692800.0,
      "grad_norm": 1.9613259922265862,
      "language_loss": 0.63070661,
      "learning_rate": 2.3625432915011443e-06,
      "loss": 0.65291733,
      "num_input_tokens_seen": 82102310,
      "step": 3815,
      "time_per_iteration": 2.636105537414551
    },
    {
      "auxiliary_loss_clip": 0.01148989,
      "auxiliary_loss_mlp": 0.01033379,
      "balance_loss_clip": 1.04598403,
      "balance_loss_mlp": 1.02450347,
      "epoch": 0.4588468706787711,
      "flos": 24097927680000.0,
      "grad_norm": 1.7390973396102092,
      "language_loss": 0.65573537,
      "learning_rate": 2.3617771975482334e-06,
      "loss": 0.67755902,
      "num_input_tokens_seen": 82121140,
      "step": 3816,
      "time_per_iteration": 2.7017769813537598
    },
    {
      "auxiliary_loss_clip": 0.01124399,
      "auxiliary_loss_mlp": 0.01034733,
      "balance_loss_clip": 1.04464364,
      "balance_loss_mlp": 1.0265379,
      "epoch": 0.4589671135694102,
      "flos": 17889331622400.0,
      "grad_norm": 1.6910284386742453,
      "language_loss": 0.74327016,
      "learning_rate": 2.3610110487140083e-06,
      "loss": 0.76486152,
      "num_input_tokens_seen": 82139575,
      "step": 3817,
      "time_per_iteration": 2.7090156078338623
    },
    {
      "auxiliary_loss_clip": 0.01157608,
      "auxiliary_loss_mlp": 0.01027126,
      "balance_loss_clip": 1.04968405,
      "balance_loss_mlp": 1.01910329,
      "epoch": 0.4590873564600493,
      "flos": 25626967781760.0,
      "grad_norm": 1.7776623124767856,
      "language_loss": 0.80670756,
      "learning_rate": 2.360244845114695e-06,
      "loss": 0.82855487,
      "num_input_tokens_seen": 82159195,
      "step": 3818,
      "time_per_iteration": 2.7197089195251465
    },
    {
      "auxiliary_loss_clip": 0.0115057,
      "auxiliary_loss_mlp": 0.01028055,
      "balance_loss_clip": 1.0491209,
      "balance_loss_mlp": 1.02010369,
      "epoch": 0.4592075993506884,
      "flos": 18514788168960.0,
      "grad_norm": 2.3070480130475914,
      "language_loss": 0.68525219,
      "learning_rate": 2.3594785868665245e-06,
      "loss": 0.7070384,
      "num_input_tokens_seen": 82175500,
      "step": 3819,
      "time_per_iteration": 2.6559667587280273
    },
    {
      "auxiliary_loss_clip": 0.01145237,
      "auxiliary_loss_mlp": 0.00763428,
      "balance_loss_clip": 1.04859281,
      "balance_loss_mlp": 1.00016046,
      "epoch": 0.4593278422413275,
      "flos": 20631111638400.0,
      "grad_norm": 1.9523955743573738,
      "language_loss": 0.80428356,
      "learning_rate": 2.3587122740857386e-06,
      "loss": 0.82337022,
      "num_input_tokens_seen": 82192600,
      "step": 3820,
      "time_per_iteration": 2.7656543254852295
    },
    {
      "auxiliary_loss_clip": 0.01168806,
      "auxiliary_loss_mlp": 0.01023422,
      "balance_loss_clip": 1.04884219,
      "balance_loss_mlp": 1.01651955,
      "epoch": 0.45944808513196655,
      "flos": 21358517961600.0,
      "grad_norm": 1.7667110949097864,
      "language_loss": 0.7804119,
      "learning_rate": 2.357945906888586e-06,
      "loss": 0.80233419,
      "num_input_tokens_seen": 82212040,
      "step": 3821,
      "time_per_iteration": 2.7187771797180176
    },
    {
      "auxiliary_loss_clip": 0.0117185,
      "auxiliary_loss_mlp": 0.01033061,
      "balance_loss_clip": 1.05050504,
      "balance_loss_mlp": 1.0242393,
      "epoch": 0.45956832802260567,
      "flos": 21427789340160.0,
      "grad_norm": 2.680941359972959,
      "language_loss": 0.80118525,
      "learning_rate": 2.357179485391324e-06,
      "loss": 0.82323438,
      "num_input_tokens_seen": 82229895,
      "step": 3822,
      "time_per_iteration": 2.7255406379699707
    },
    {
      "auxiliary_loss_clip": 0.01184003,
      "auxiliary_loss_mlp": 0.01030312,
      "balance_loss_clip": 1.05197001,
      "balance_loss_mlp": 1.02208066,
      "epoch": 0.4596885709132448,
      "flos": 22382654538240.0,
      "grad_norm": 2.508324661546939,
      "language_loss": 0.86311257,
      "learning_rate": 2.3564130097102173e-06,
      "loss": 0.88525569,
      "num_input_tokens_seen": 82249550,
      "step": 3823,
      "time_per_iteration": 2.675727128982544
    },
    {
      "auxiliary_loss_clip": 0.01148263,
      "auxiliary_loss_mlp": 0.01032315,
      "balance_loss_clip": 1.04824567,
      "balance_loss_mlp": 1.02403545,
      "epoch": 0.45980881380388383,
      "flos": 28981957806720.0,
      "grad_norm": 1.7674743409379776,
      "language_loss": 0.75412929,
      "learning_rate": 2.355646479961541e-06,
      "loss": 0.77593505,
      "num_input_tokens_seen": 82268860,
      "step": 3824,
      "time_per_iteration": 2.7737581729888916
    },
    {
      "auxiliary_loss_clip": 0.01185078,
      "auxiliary_loss_mlp": 0.01024507,
      "balance_loss_clip": 1.05121303,
      "balance_loss_mlp": 1.01652026,
      "epoch": 0.45992905669452294,
      "flos": 33396599980800.0,
      "grad_norm": 1.8598177290271187,
      "language_loss": 0.71620399,
      "learning_rate": 2.354879896261576e-06,
      "loss": 0.73829985,
      "num_input_tokens_seen": 82289070,
      "step": 3825,
      "time_per_iteration": 2.731172800064087
    },
    {
      "auxiliary_loss_clip": 0.01137687,
      "auxiliary_loss_mlp": 0.01030767,
      "balance_loss_clip": 1.0481149,
      "balance_loss_mlp": 1.0222497,
      "epoch": 0.46004929958516205,
      "flos": 36318184502400.0,
      "grad_norm": 2.0361364384270644,
      "language_loss": 0.5669719,
      "learning_rate": 2.3541132587266133e-06,
      "loss": 0.58865643,
      "num_input_tokens_seen": 82311790,
      "step": 3826,
      "time_per_iteration": 2.8080198764801025
    },
    {
      "auxiliary_loss_clip": 0.01148223,
      "auxiliary_loss_mlp": 0.01030635,
      "balance_loss_clip": 1.04961205,
      "balance_loss_mlp": 1.0219568,
      "epoch": 0.4601695424758011,
      "flos": 17238451224960.0,
      "grad_norm": 4.180153879432144,
      "language_loss": 0.69292665,
      "learning_rate": 2.3533465674729515e-06,
      "loss": 0.71471524,
      "num_input_tokens_seen": 82329020,
      "step": 3827,
      "time_per_iteration": 2.694967746734619
    },
    {
      "auxiliary_loss_clip": 0.01187506,
      "auxiliary_loss_mlp": 0.01036195,
      "balance_loss_clip": 1.05341446,
      "balance_loss_mlp": 1.02759361,
      "epoch": 0.4602897853664402,
      "flos": 15888425529600.0,
      "grad_norm": 2.1342636477737753,
      "language_loss": 0.72766173,
      "learning_rate": 2.352579822616895e-06,
      "loss": 0.74989867,
      "num_input_tokens_seen": 82346455,
      "step": 3828,
      "time_per_iteration": 2.654542922973633
    },
    {
      "auxiliary_loss_clip": 0.0115907,
      "auxiliary_loss_mlp": 0.01021498,
      "balance_loss_clip": 1.05054283,
      "balance_loss_mlp": 1.01356745,
      "epoch": 0.4604100282570793,
      "flos": 25412617370880.0,
      "grad_norm": 1.8964156694111172,
      "language_loss": 0.77667391,
      "learning_rate": 2.351813024274761e-06,
      "loss": 0.79847962,
      "num_input_tokens_seen": 82367810,
      "step": 3829,
      "time_per_iteration": 2.6943414211273193
    },
    {
      "auxiliary_loss_clip": 0.01146484,
      "auxiliary_loss_mlp": 0.01025476,
      "balance_loss_clip": 1.04884255,
      "balance_loss_mlp": 1.01761127,
      "epoch": 0.4605302711477184,
      "flos": 27630711048960.0,
      "grad_norm": 1.8083484584285114,
      "language_loss": 0.73786402,
      "learning_rate": 2.3510461725628693e-06,
      "loss": 0.75958365,
      "num_input_tokens_seen": 82388275,
      "step": 3830,
      "time_per_iteration": 2.693460702896118
    },
    {
      "auxiliary_loss_clip": 0.0114455,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 1.04831338,
      "balance_loss_mlp": 1.0232892,
      "epoch": 0.4606505140383575,
      "flos": 23839657914240.0,
      "grad_norm": 2.15721949759543,
      "language_loss": 0.7107932,
      "learning_rate": 2.350279267597554e-06,
      "loss": 0.73255098,
      "num_input_tokens_seen": 82408915,
      "step": 3831,
      "time_per_iteration": 2.7996551990509033
    },
    {
      "auxiliary_loss_clip": 0.01171953,
      "auxiliary_loss_mlp": 0.01027899,
      "balance_loss_clip": 1.05049241,
      "balance_loss_mlp": 1.01926827,
      "epoch": 0.46077075692899655,
      "flos": 16107013745280.0,
      "grad_norm": 2.305715771050497,
      "language_loss": 0.83032143,
      "learning_rate": 2.3495123094951515e-06,
      "loss": 0.85231996,
      "num_input_tokens_seen": 82427260,
      "step": 3832,
      "time_per_iteration": 2.6804981231689453
    },
    {
      "auxiliary_loss_clip": 0.01148995,
      "auxiliary_loss_mlp": 0.01028715,
      "balance_loss_clip": 1.04835498,
      "balance_loss_mlp": 1.02032292,
      "epoch": 0.46089099981963566,
      "flos": 48798147634560.0,
      "grad_norm": 2.6011501881987216,
      "language_loss": 0.75793958,
      "learning_rate": 2.34874529837201e-06,
      "loss": 0.77971667,
      "num_input_tokens_seen": 82450805,
      "step": 3833,
      "time_per_iteration": 4.834528923034668
    },
    {
      "auxiliary_loss_clip": 0.01104998,
      "auxiliary_loss_mlp": 0.01029453,
      "balance_loss_clip": 1.04086947,
      "balance_loss_mlp": 1.02134073,
      "epoch": 0.46101124271027477,
      "flos": 19099234362240.0,
      "grad_norm": 2.123662846856711,
      "language_loss": 0.79130423,
      "learning_rate": 2.347978234344483e-06,
      "loss": 0.81264877,
      "num_input_tokens_seen": 82467010,
      "step": 3834,
      "time_per_iteration": 2.7935197353363037
    },
    {
      "auxiliary_loss_clip": 0.01177461,
      "auxiliary_loss_mlp": 0.01035178,
      "balance_loss_clip": 1.05286276,
      "balance_loss_mlp": 1.02589488,
      "epoch": 0.4611314856009138,
      "flos": 39347931853440.0,
      "grad_norm": 1.830981154374917,
      "language_loss": 0.69118893,
      "learning_rate": 2.347211117528935e-06,
      "loss": 0.71331537,
      "num_input_tokens_seen": 82489310,
      "step": 3835,
      "time_per_iteration": 2.8139333724975586
    },
    {
      "auxiliary_loss_clip": 0.01151984,
      "auxiliary_loss_mlp": 0.01034845,
      "balance_loss_clip": 1.0519768,
      "balance_loss_mlp": 1.02659583,
      "epoch": 0.46125172849155294,
      "flos": 20810772489600.0,
      "grad_norm": 1.5587743291800389,
      "language_loss": 0.7187106,
      "learning_rate": 2.3464439480417374e-06,
      "loss": 0.74057889,
      "num_input_tokens_seen": 82508830,
      "step": 3836,
      "time_per_iteration": 3.6322555541992188
    },
    {
      "auxiliary_loss_clip": 0.0117897,
      "auxiliary_loss_mlp": 0.01028747,
      "balance_loss_clip": 1.05469489,
      "balance_loss_mlp": 1.01920986,
      "epoch": 0.46137197138219205,
      "flos": 17930808852480.0,
      "grad_norm": 2.6436250763463542,
      "language_loss": 0.77405608,
      "learning_rate": 2.3456767259992676e-06,
      "loss": 0.79613328,
      "num_input_tokens_seen": 82526475,
      "step": 3837,
      "time_per_iteration": 2.6096034049987793
    },
    {
      "auxiliary_loss_clip": 0.0118727,
      "auxiliary_loss_mlp": 0.00763198,
      "balance_loss_clip": 1.05133927,
      "balance_loss_mlp": 1.00015545,
      "epoch": 0.4614922142728311,
      "flos": 16836610798080.0,
      "grad_norm": 2.547858527077774,
      "language_loss": 0.88318008,
      "learning_rate": 2.3449094515179135e-06,
      "loss": 0.90268475,
      "num_input_tokens_seen": 82543935,
      "step": 3838,
      "time_per_iteration": 3.5027754306793213
    },
    {
      "auxiliary_loss_clip": 0.01160092,
      "auxiliary_loss_mlp": 0.01026096,
      "balance_loss_clip": 1.04692447,
      "balance_loss_mlp": 1.01809108,
      "epoch": 0.4616124571634702,
      "flos": 26614906427520.0,
      "grad_norm": 1.568236967828726,
      "language_loss": 0.81784105,
      "learning_rate": 2.34414212471407e-06,
      "loss": 0.8397029,
      "num_input_tokens_seen": 82563730,
      "step": 3839,
      "time_per_iteration": 2.6916425228118896
    },
    {
      "auxiliary_loss_clip": 0.0117905,
      "auxiliary_loss_mlp": 0.01029896,
      "balance_loss_clip": 1.05133379,
      "balance_loss_mlp": 1.02062726,
      "epoch": 0.4617327000541093,
      "flos": 20340127560960.0,
      "grad_norm": 1.9965868580968595,
      "language_loss": 0.7295295,
      "learning_rate": 2.3433747457041394e-06,
      "loss": 0.75161886,
      "num_input_tokens_seen": 82582435,
      "step": 3840,
      "time_per_iteration": 2.6610267162323
    },
    {
      "auxiliary_loss_clip": 0.01143233,
      "auxiliary_loss_mlp": 0.01028695,
      "balance_loss_clip": 1.04903841,
      "balance_loss_mlp": 1.01955748,
      "epoch": 0.4618529429447484,
      "flos": 29570749545600.0,
      "grad_norm": 1.9347505496803592,
      "language_loss": 0.84917074,
      "learning_rate": 2.342607314604533e-06,
      "loss": 0.87089008,
      "num_input_tokens_seen": 82602185,
      "step": 3841,
      "time_per_iteration": 2.7545342445373535
    },
    {
      "auxiliary_loss_clip": 0.01171788,
      "auxiliary_loss_mlp": 0.01029459,
      "balance_loss_clip": 1.05355382,
      "balance_loss_mlp": 1.01995838,
      "epoch": 0.4619731858353875,
      "flos": 19787030962560.0,
      "grad_norm": 6.666626297566391,
      "language_loss": 0.84304082,
      "learning_rate": 2.3418398315316694e-06,
      "loss": 0.8650533,
      "num_input_tokens_seen": 82620005,
      "step": 3842,
      "time_per_iteration": 2.7001709938049316
    },
    {
      "auxiliary_loss_clip": 0.01187307,
      "auxiliary_loss_mlp": 0.01023824,
      "balance_loss_clip": 1.05411696,
      "balance_loss_mlp": 1.01577187,
      "epoch": 0.4620934287260266,
      "flos": 18951138587520.0,
      "grad_norm": 2.2766224116393405,
      "language_loss": 0.78259599,
      "learning_rate": 2.3410722966019755e-06,
      "loss": 0.80470729,
      "num_input_tokens_seen": 82635120,
      "step": 3843,
      "time_per_iteration": 2.647512674331665
    },
    {
      "auxiliary_loss_clip": 0.0116939,
      "auxiliary_loss_mlp": 0.01031142,
      "balance_loss_clip": 1.05019999,
      "balance_loss_mlp": 1.02339935,
      "epoch": 0.46221367161666566,
      "flos": 37341674634240.0,
      "grad_norm": 2.5564385581536007,
      "language_loss": 0.65524805,
      "learning_rate": 2.3403047099318848e-06,
      "loss": 0.67725337,
      "num_input_tokens_seen": 82659190,
      "step": 3844,
      "time_per_iteration": 2.783294677734375
    },
    {
      "auxiliary_loss_clip": 0.01120497,
      "auxiliary_loss_mlp": 0.01025188,
      "balance_loss_clip": 1.0418998,
      "balance_loss_mlp": 1.01696825,
      "epoch": 0.46233391450730477,
      "flos": 14428549065600.0,
      "grad_norm": 2.5797584752317313,
      "language_loss": 0.74743116,
      "learning_rate": 2.3395370716378405e-06,
      "loss": 0.768888,
      "num_input_tokens_seen": 82676635,
      "step": 3845,
      "time_per_iteration": 2.674377679824829
    },
    {
      "auxiliary_loss_clip": 0.01175237,
      "auxiliary_loss_mlp": 0.01025083,
      "balance_loss_clip": 1.05083787,
      "balance_loss_mlp": 1.01642871,
      "epoch": 0.4624541573979438,
      "flos": 22493044010880.0,
      "grad_norm": 2.2679206490784822,
      "language_loss": 0.72343349,
      "learning_rate": 2.338769381836292e-06,
      "loss": 0.74543667,
      "num_input_tokens_seen": 82696245,
      "step": 3846,
      "time_per_iteration": 2.857036828994751
    },
    {
      "auxiliary_loss_clip": 0.01143087,
      "auxiliary_loss_mlp": 0.01030077,
      "balance_loss_clip": 1.05101824,
      "balance_loss_mlp": 1.02217388,
      "epoch": 0.46257440028858293,
      "flos": 14465070218880.0,
      "grad_norm": 1.9066503631982015,
      "language_loss": 0.73297554,
      "learning_rate": 2.3380016406436984e-06,
      "loss": 0.75470716,
      "num_input_tokens_seen": 82713725,
      "step": 3847,
      "time_per_iteration": 2.679553747177124
    },
    {
      "auxiliary_loss_clip": 0.01127516,
      "auxiliary_loss_mlp": 0.0102585,
      "balance_loss_clip": 1.04882884,
      "balance_loss_mlp": 1.01668906,
      "epoch": 0.46269464317922204,
      "flos": 23332204523520.0,
      "grad_norm": 2.041135317237289,
      "language_loss": 0.81624365,
      "learning_rate": 2.337233848176524e-06,
      "loss": 0.83777732,
      "num_input_tokens_seen": 82731495,
      "step": 3848,
      "time_per_iteration": 2.729567766189575
    },
    {
      "auxiliary_loss_clip": 0.01120437,
      "auxiliary_loss_mlp": 0.01030813,
      "balance_loss_clip": 1.0438211,
      "balance_loss_mlp": 1.02203357,
      "epoch": 0.4628148860698611,
      "flos": 18552027594240.0,
      "grad_norm": 1.9321330338105165,
      "language_loss": 0.83513039,
      "learning_rate": 2.3364660045512435e-06,
      "loss": 0.85664296,
      "num_input_tokens_seen": 82750255,
      "step": 3849,
      "time_per_iteration": 2.7877683639526367
    },
    {
      "auxiliary_loss_clip": 0.01060181,
      "auxiliary_loss_mlp": 0.01009802,
      "balance_loss_clip": 1.01761532,
      "balance_loss_mlp": 1.00809693,
      "epoch": 0.4629351289605002,
      "flos": 70667569670400.0,
      "grad_norm": 0.8352963142381657,
      "language_loss": 0.58232033,
      "learning_rate": 2.335698109884337e-06,
      "loss": 0.60302019,
      "num_input_tokens_seen": 82815460,
      "step": 3850,
      "time_per_iteration": 3.4033608436584473
    },
    {
      "auxiliary_loss_clip": 0.01052099,
      "auxiliary_loss_mlp": 0.01004457,
      "balance_loss_clip": 1.02902687,
      "balance_loss_mlp": 1.00278759,
      "epoch": 0.4630553718511393,
      "flos": 59687200465920.0,
      "grad_norm": 0.7852880779529042,
      "language_loss": 0.59865332,
      "learning_rate": 2.334930164292294e-06,
      "loss": 0.61921889,
      "num_input_tokens_seen": 82878010,
      "step": 3851,
      "time_per_iteration": 3.459230899810791
    },
    {
      "auxiliary_loss_clip": 0.0112122,
      "auxiliary_loss_mlp": 0.01032099,
      "balance_loss_clip": 1.0427264,
      "balance_loss_mlp": 1.02396929,
      "epoch": 0.4631756147417784,
      "flos": 15960605909760.0,
      "grad_norm": 1.954813854944133,
      "language_loss": 0.79889292,
      "learning_rate": 2.334162167891612e-06,
      "loss": 0.82042611,
      "num_input_tokens_seen": 82895275,
      "step": 3852,
      "time_per_iteration": 2.933835029602051
    },
    {
      "auxiliary_loss_clip": 0.01162005,
      "auxiliary_loss_mlp": 0.01028227,
      "balance_loss_clip": 1.04913259,
      "balance_loss_mlp": 1.01957214,
      "epoch": 0.4632958576324175,
      "flos": 16472907636480.0,
      "grad_norm": 2.076889048683533,
      "language_loss": 0.75121981,
      "learning_rate": 2.333394120798795e-06,
      "loss": 0.77312213,
      "num_input_tokens_seen": 82914010,
      "step": 3853,
      "time_per_iteration": 2.6549072265625
    },
    {
      "auxiliary_loss_clip": 0.01156637,
      "auxiliary_loss_mlp": 0.01024321,
      "balance_loss_clip": 1.04721332,
      "balance_loss_mlp": 1.0164237,
      "epoch": 0.4634161005230566,
      "flos": 22346492520960.0,
      "grad_norm": 2.529322269152144,
      "language_loss": 0.72005647,
      "learning_rate": 2.3326260231303545e-06,
      "loss": 0.74186611,
      "num_input_tokens_seen": 82932610,
      "step": 3854,
      "time_per_iteration": 2.722343921661377
    },
    {
      "auxiliary_loss_clip": 0.01183604,
      "auxiliary_loss_mlp": 0.01026289,
      "balance_loss_clip": 1.0521425,
      "balance_loss_mlp": 1.01871669,
      "epoch": 0.46353634341369565,
      "flos": 15742233175680.0,
      "grad_norm": 1.6618001694861786,
      "language_loss": 0.87005103,
      "learning_rate": 2.331857875002811e-06,
      "loss": 0.89214998,
      "num_input_tokens_seen": 82951210,
      "step": 3855,
      "time_per_iteration": 2.6663739681243896
    },
    {
      "auxiliary_loss_clip": 0.01159065,
      "auxiliary_loss_mlp": 0.01024856,
      "balance_loss_clip": 1.05302477,
      "balance_loss_mlp": 1.01711345,
      "epoch": 0.46365658630433476,
      "flos": 28329820433280.0,
      "grad_norm": 1.6125485612152428,
      "language_loss": 0.76332641,
      "learning_rate": 2.3310896765326916e-06,
      "loss": 0.78516561,
      "num_input_tokens_seen": 82972210,
      "step": 3856,
      "time_per_iteration": 2.8275814056396484
    },
    {
      "auxiliary_loss_clip": 0.01137613,
      "auxiliary_loss_mlp": 0.01028114,
      "balance_loss_clip": 1.04733109,
      "balance_loss_mlp": 1.01976299,
      "epoch": 0.46377682919497387,
      "flos": 24608074590720.0,
      "grad_norm": 1.6634255282702313,
      "language_loss": 0.84311587,
      "learning_rate": 2.330321427836531e-06,
      "loss": 0.86477315,
      "num_input_tokens_seen": 82994080,
      "step": 3857,
      "time_per_iteration": 2.7586252689361572
    },
    {
      "auxiliary_loss_clip": 0.01169658,
      "auxiliary_loss_mlp": 0.01026824,
      "balance_loss_clip": 1.05126429,
      "balance_loss_mlp": 1.01908112,
      "epoch": 0.4638970720856129,
      "flos": 19060953442560.0,
      "grad_norm": 4.993732182188965,
      "language_loss": 0.8280341,
      "learning_rate": 2.3295531290308733e-06,
      "loss": 0.84999889,
      "num_input_tokens_seen": 83012230,
      "step": 3858,
      "time_per_iteration": 2.6841845512390137
    },
    {
      "auxiliary_loss_clip": 0.01189824,
      "auxiliary_loss_mlp": 0.00764034,
      "balance_loss_clip": 1.05464041,
      "balance_loss_mlp": 1.00021005,
      "epoch": 0.46401731497625204,
      "flos": 18471012468480.0,
      "grad_norm": 3.4836264901156975,
      "language_loss": 0.76062083,
      "learning_rate": 2.3287847802322678e-06,
      "loss": 0.78015941,
      "num_input_tokens_seen": 83027800,
      "step": 3859,
      "time_per_iteration": 3.5625555515289307
    },
    {
      "auxiliary_loss_clip": 0.0116461,
      "auxiliary_loss_mlp": 0.01029743,
      "balance_loss_clip": 1.05216444,
      "balance_loss_mlp": 1.02034307,
      "epoch": 0.4641375578668911,
      "flos": 26067053214720.0,
      "grad_norm": 1.925636636515387,
      "language_loss": 0.84229344,
      "learning_rate": 2.3280163815572723e-06,
      "loss": 0.86423695,
      "num_input_tokens_seen": 83048395,
      "step": 3860,
      "time_per_iteration": 2.7129099369049072
    },
    {
      "auxiliary_loss_clip": 0.01149464,
      "auxiliary_loss_mlp": 0.01029719,
      "balance_loss_clip": 1.04840922,
      "balance_loss_mlp": 1.02125514,
      "epoch": 0.4642578007575302,
      "flos": 19570382081280.0,
      "grad_norm": 2.3899367267100375,
      "language_loss": 0.77333945,
      "learning_rate": 2.3272479331224522e-06,
      "loss": 0.79513133,
      "num_input_tokens_seen": 83065825,
      "step": 3861,
      "time_per_iteration": 2.734492063522339
    },
    {
      "auxiliary_loss_clip": 0.01185802,
      "auxiliary_loss_mlp": 0.01030294,
      "balance_loss_clip": 1.0506587,
      "balance_loss_mlp": 1.0217886,
      "epoch": 0.4643780436481693,
      "flos": 28186249772160.0,
      "grad_norm": 1.7788738651644669,
      "language_loss": 0.78168547,
      "learning_rate": 2.3264794350443817e-06,
      "loss": 0.80384636,
      "num_input_tokens_seen": 83087920,
      "step": 3862,
      "time_per_iteration": 3.577354669570923
    },
    {
      "auxiliary_loss_clip": 0.01173997,
      "auxiliary_loss_mlp": 0.01029522,
      "balance_loss_clip": 1.04983401,
      "balance_loss_mlp": 1.02080178,
      "epoch": 0.46449828653880837,
      "flos": 25375270204800.0,
      "grad_norm": 1.8306698245970212,
      "language_loss": 0.78700507,
      "learning_rate": 2.3257108874396396e-06,
      "loss": 0.80904025,
      "num_input_tokens_seen": 83109015,
      "step": 3863,
      "time_per_iteration": 2.769343376159668
    },
    {
      "auxiliary_loss_clip": 0.01155549,
      "auxiliary_loss_mlp": 0.01035987,
      "balance_loss_clip": 1.0468142,
      "balance_loss_mlp": 1.02779722,
      "epoch": 0.4646185294294475,
      "flos": 16034330574720.0,
      "grad_norm": 1.9144599639290787,
      "language_loss": 0.74081886,
      "learning_rate": 2.3249422904248152e-06,
      "loss": 0.76273423,
      "num_input_tokens_seen": 83127450,
      "step": 3864,
      "time_per_iteration": 3.7400617599487305
    },
    {
      "auxiliary_loss_clip": 0.01175828,
      "auxiliary_loss_mlp": 0.01032971,
      "balance_loss_clip": 1.05170524,
      "balance_loss_mlp": 1.02504611,
      "epoch": 0.4647387723200866,
      "flos": 26363101109760.0,
      "grad_norm": 1.9812387411174233,
      "language_loss": 0.8695817,
      "learning_rate": 2.324173644116504e-06,
      "loss": 0.89166963,
      "num_input_tokens_seen": 83150300,
      "step": 3865,
      "time_per_iteration": 2.7115299701690674
    },
    {
      "auxiliary_loss_clip": 0.0116937,
      "auxiliary_loss_mlp": 0.01030796,
      "balance_loss_clip": 1.05199504,
      "balance_loss_mlp": 1.02215374,
      "epoch": 0.46485901521072565,
      "flos": 27160209774720.0,
      "grad_norm": 1.6341360138761267,
      "language_loss": 0.81403613,
      "learning_rate": 2.3234049486313087e-06,
      "loss": 0.83603781,
      "num_input_tokens_seen": 83171750,
      "step": 3866,
      "time_per_iteration": 2.852799415588379
    },
    {
      "auxiliary_loss_clip": 0.01171471,
      "auxiliary_loss_mlp": 0.01026814,
      "balance_loss_clip": 1.05132329,
      "balance_loss_mlp": 1.01922011,
      "epoch": 0.46497925810136476,
      "flos": 24279851088000.0,
      "grad_norm": 1.8405904263757935,
      "language_loss": 0.75671649,
      "learning_rate": 2.322636204085839e-06,
      "loss": 0.77869934,
      "num_input_tokens_seen": 83191820,
      "step": 3867,
      "time_per_iteration": 2.7998647689819336
    },
    {
      "auxiliary_loss_clip": 0.01146601,
      "auxiliary_loss_mlp": 0.01030021,
      "balance_loss_clip": 1.04451084,
      "balance_loss_mlp": 1.02163458,
      "epoch": 0.46509950099200387,
      "flos": 16253134272000.0,
      "grad_norm": 2.2200537038963937,
      "language_loss": 0.78956378,
      "learning_rate": 2.3218674105967143e-06,
      "loss": 0.81133002,
      "num_input_tokens_seen": 83210085,
      "step": 3868,
      "time_per_iteration": 2.7260873317718506
    },
    {
      "auxiliary_loss_clip": 0.01149234,
      "auxiliary_loss_mlp": 0.01035078,
      "balance_loss_clip": 1.04635096,
      "balance_loss_mlp": 1.02675152,
      "epoch": 0.4652197438826429,
      "flos": 23442270773760.0,
      "grad_norm": 1.522847998352799,
      "language_loss": 0.8342036,
      "learning_rate": 2.3210985682805593e-06,
      "loss": 0.85604674,
      "num_input_tokens_seen": 83231865,
      "step": 3869,
      "time_per_iteration": 2.742035150527954
    },
    {
      "auxiliary_loss_clip": 0.01188117,
      "auxiliary_loss_mlp": 0.01030858,
      "balance_loss_clip": 1.05451369,
      "balance_loss_mlp": 1.02225685,
      "epoch": 0.46533998677328203,
      "flos": 16216397637120.0,
      "grad_norm": 2.538426337038713,
      "language_loss": 0.68076175,
      "learning_rate": 2.320329677254007e-06,
      "loss": 0.70295149,
      "num_input_tokens_seen": 83249195,
      "step": 3870,
      "time_per_iteration": 2.589747190475464
    },
    {
      "auxiliary_loss_clip": 0.01189783,
      "auxiliary_loss_mlp": 0.01033064,
      "balance_loss_clip": 1.0565486,
      "balance_loss_mlp": 1.02467191,
      "epoch": 0.46546022966392114,
      "flos": 21141869080320.0,
      "grad_norm": 2.303044567066429,
      "language_loss": 0.72253466,
      "learning_rate": 2.319560737633697e-06,
      "loss": 0.74476314,
      "num_input_tokens_seen": 83267915,
      "step": 3871,
      "time_per_iteration": 2.602261781692505
    },
    {
      "auxiliary_loss_clip": 0.01148298,
      "auxiliary_loss_mlp": 0.0103108,
      "balance_loss_clip": 1.04580522,
      "balance_loss_mlp": 1.02233052,
      "epoch": 0.4655804725545602,
      "flos": 41171942442240.0,
      "grad_norm": 1.7225242059536656,
      "language_loss": 0.6830616,
      "learning_rate": 2.3187917495362775e-06,
      "loss": 0.70485544,
      "num_input_tokens_seen": 83292325,
      "step": 3872,
      "time_per_iteration": 2.8763175010681152
    },
    {
      "auxiliary_loss_clip": 0.01126956,
      "auxiliary_loss_mlp": 0.01028161,
      "balance_loss_clip": 1.04787397,
      "balance_loss_mlp": 1.02033174,
      "epoch": 0.4657007154451993,
      "flos": 19570956698880.0,
      "grad_norm": 3.226177377721573,
      "language_loss": 0.76722568,
      "learning_rate": 2.318022713078403e-06,
      "loss": 0.78877681,
      "num_input_tokens_seen": 83306905,
      "step": 3873,
      "time_per_iteration": 2.8014166355133057
    },
    {
      "auxiliary_loss_clip": 0.01159028,
      "auxiliary_loss_mlp": 0.01025266,
      "balance_loss_clip": 1.04997563,
      "balance_loss_mlp": 1.01739192,
      "epoch": 0.4658209583358384,
      "flos": 15517826956800.0,
      "grad_norm": 2.2893660919300047,
      "language_loss": 0.85500157,
      "learning_rate": 2.3172536283767354e-06,
      "loss": 0.87684453,
      "num_input_tokens_seen": 83320665,
      "step": 3874,
      "time_per_iteration": 2.661252021789551
    },
    {
      "auxiliary_loss_clip": 0.0113822,
      "auxiliary_loss_mlp": 0.01029447,
      "balance_loss_clip": 1.04788351,
      "balance_loss_mlp": 1.02074504,
      "epoch": 0.4659412012264775,
      "flos": 14903180403840.0,
      "grad_norm": 1.9936366509920724,
      "language_loss": 0.80923271,
      "learning_rate": 2.3164844955479447e-06,
      "loss": 0.83090937,
      "num_input_tokens_seen": 83336475,
      "step": 3875,
      "time_per_iteration": 2.809168577194214
    },
    {
      "auxiliary_loss_clip": 0.01134962,
      "auxiliary_loss_mlp": 0.01028867,
      "balance_loss_clip": 1.04584444,
      "balance_loss_mlp": 1.02044535,
      "epoch": 0.4660614441171166,
      "flos": 24425612478720.0,
      "grad_norm": 2.110800098459517,
      "language_loss": 0.70792925,
      "learning_rate": 2.3157153147087082e-06,
      "loss": 0.72956753,
      "num_input_tokens_seen": 83358365,
      "step": 3876,
      "time_per_iteration": 2.7619433403015137
    },
    {
      "auxiliary_loss_clip": 0.01138648,
      "auxiliary_loss_mlp": 0.0102929,
      "balance_loss_clip": 1.05070949,
      "balance_loss_mlp": 1.02080202,
      "epoch": 0.46618168700775564,
      "flos": 22091095843200.0,
      "grad_norm": 1.7118852573851393,
      "language_loss": 0.82956785,
      "learning_rate": 2.314946085975709e-06,
      "loss": 0.85124725,
      "num_input_tokens_seen": 83377345,
      "step": 3877,
      "time_per_iteration": 2.8471899032592773
    },
    {
      "auxiliary_loss_clip": 0.01132497,
      "auxiliary_loss_mlp": 0.01030182,
      "balance_loss_clip": 1.04869437,
      "balance_loss_mlp": 1.02229643,
      "epoch": 0.46630192989839475,
      "flos": 26176975810560.0,
      "grad_norm": 1.7311050698130053,
      "language_loss": 0.82439876,
      "learning_rate": 2.3141768094656393e-06,
      "loss": 0.84602559,
      "num_input_tokens_seen": 83395920,
      "step": 3878,
      "time_per_iteration": 2.7453835010528564
    },
    {
      "auxiliary_loss_clip": 0.01106367,
      "auxiliary_loss_mlp": 0.01027463,
      "balance_loss_clip": 1.0422852,
      "balance_loss_mlp": 1.01952052,
      "epoch": 0.46642217278903386,
      "flos": 11509622150400.0,
      "grad_norm": 2.34003994231913,
      "language_loss": 0.82827914,
      "learning_rate": 2.3134074852951966e-06,
      "loss": 0.84961748,
      "num_input_tokens_seen": 83412510,
      "step": 3879,
      "time_per_iteration": 2.842590808868408
    },
    {
      "auxiliary_loss_clip": 0.01121201,
      "auxiliary_loss_mlp": 0.01028459,
      "balance_loss_clip": 1.04191113,
      "balance_loss_mlp": 1.02057934,
      "epoch": 0.4665424156796729,
      "flos": 32306819299200.0,
      "grad_norm": 1.679315316127553,
      "language_loss": 0.77714789,
      "learning_rate": 2.312638113581088e-06,
      "loss": 0.79864454,
      "num_input_tokens_seen": 83432995,
      "step": 3880,
      "time_per_iteration": 2.899304151535034
    },
    {
      "auxiliary_loss_clip": 0.01168961,
      "auxiliary_loss_mlp": 0.01030593,
      "balance_loss_clip": 1.04736447,
      "balance_loss_mlp": 1.0219326,
      "epoch": 0.46666265857031203,
      "flos": 18436179254400.0,
      "grad_norm": 2.6972281801373064,
      "language_loss": 0.77824664,
      "learning_rate": 2.311868694440027e-06,
      "loss": 0.80024219,
      "num_input_tokens_seen": 83447415,
      "step": 3881,
      "time_per_iteration": 2.6214747428894043
    },
    {
      "auxiliary_loss_clip": 0.01082556,
      "auxiliary_loss_mlp": 0.0100399,
      "balance_loss_clip": 1.01549935,
      "balance_loss_mlp": 1.00236917,
      "epoch": 0.46678290146095114,
      "flos": 68438989221120.0,
      "grad_norm": 0.7409824968993908,
      "language_loss": 0.6245923,
      "learning_rate": 2.3110992279887323e-06,
      "loss": 0.6454578,
      "num_input_tokens_seen": 83519340,
      "step": 3882,
      "time_per_iteration": 3.3362624645233154
    },
    {
      "auxiliary_loss_clip": 0.0115097,
      "auxiliary_loss_mlp": 0.01031784,
      "balance_loss_clip": 1.05143702,
      "balance_loss_mlp": 1.02367163,
      "epoch": 0.4669031443515902,
      "flos": 17712507945600.0,
      "grad_norm": 2.449782917073229,
      "language_loss": 0.84478247,
      "learning_rate": 2.310329714343932e-06,
      "loss": 0.86661005,
      "num_input_tokens_seen": 83535490,
      "step": 3883,
      "time_per_iteration": 2.714383363723755
    },
    {
      "auxiliary_loss_clip": 0.01152324,
      "auxiliary_loss_mlp": 0.01027303,
      "balance_loss_clip": 1.04915118,
      "balance_loss_mlp": 1.01926255,
      "epoch": 0.4670233872422293,
      "flos": 23947748916480.0,
      "grad_norm": 2.8413399790418628,
      "language_loss": 0.81512868,
      "learning_rate": 2.309560153622361e-06,
      "loss": 0.83692497,
      "num_input_tokens_seen": 83552400,
      "step": 3884,
      "time_per_iteration": 2.730412483215332
    },
    {
      "auxiliary_loss_clip": 0.01144299,
      "auxiliary_loss_mlp": 0.01029861,
      "balance_loss_clip": 1.0492332,
      "balance_loss_mlp": 1.02127755,
      "epoch": 0.4671436301328684,
      "flos": 28111268131200.0,
      "grad_norm": 1.9750414302784336,
      "language_loss": 0.74525845,
      "learning_rate": 2.3087905459407602e-06,
      "loss": 0.76700008,
      "num_input_tokens_seen": 83571340,
      "step": 3885,
      "time_per_iteration": 3.6685636043548584
    },
    {
      "auxiliary_loss_clip": 0.01074422,
      "auxiliary_loss_mlp": 0.01002539,
      "balance_loss_clip": 1.01599216,
      "balance_loss_mlp": 1.00091779,
      "epoch": 0.46726387302350747,
      "flos": 69369684566400.0,
      "grad_norm": 0.8066923735779634,
      "language_loss": 0.62863255,
      "learning_rate": 2.3080208914158795e-06,
      "loss": 0.64940214,
      "num_input_tokens_seen": 83634340,
      "step": 3886,
      "time_per_iteration": 4.176991701126099
    },
    {
      "auxiliary_loss_clip": 0.01156356,
      "auxiliary_loss_mlp": 0.01027265,
      "balance_loss_clip": 1.05195284,
      "balance_loss_mlp": 1.01909351,
      "epoch": 0.4673841159141466,
      "flos": 25519666878720.0,
      "grad_norm": 2.944177526182004,
      "language_loss": 0.71992576,
      "learning_rate": 2.3072511901644753e-06,
      "loss": 0.74176192,
      "num_input_tokens_seen": 83653410,
      "step": 3887,
      "time_per_iteration": 2.726440668106079
    },
    {
      "auxiliary_loss_clip": 0.01182874,
      "auxiliary_loss_mlp": 0.01023682,
      "balance_loss_clip": 1.05224633,
      "balance_loss_mlp": 1.01617181,
      "epoch": 0.4675043588047857,
      "flos": 24499265316480.0,
      "grad_norm": 2.0444820183398753,
      "language_loss": 0.80899876,
      "learning_rate": 2.306481442303309e-06,
      "loss": 0.83106434,
      "num_input_tokens_seen": 83672985,
      "step": 3888,
      "time_per_iteration": 2.628535747528076
    },
    {
      "auxiliary_loss_clip": 0.01174107,
      "auxiliary_loss_mlp": 0.01030115,
      "balance_loss_clip": 1.0523684,
      "balance_loss_mlp": 1.02143621,
      "epoch": 0.46762460169542475,
      "flos": 20960771685120.0,
      "grad_norm": 1.8920606141268406,
      "language_loss": 0.73078066,
      "learning_rate": 2.3057116479491515e-06,
      "loss": 0.75282288,
      "num_input_tokens_seen": 83692395,
      "step": 3889,
      "time_per_iteration": 3.575392007827759
    },
    {
      "auxiliary_loss_clip": 0.01166545,
      "auxiliary_loss_mlp": 0.01026194,
      "balance_loss_clip": 1.04799247,
      "balance_loss_mlp": 1.01840973,
      "epoch": 0.46774484458606386,
      "flos": 19171666137600.0,
      "grad_norm": 2.166637837977406,
      "language_loss": 0.76016784,
      "learning_rate": 2.30494180721878e-06,
      "loss": 0.78209519,
      "num_input_tokens_seen": 83709735,
      "step": 3890,
      "time_per_iteration": 3.6049256324768066
    },
    {
      "auxiliary_loss_clip": 0.01168143,
      "auxiliary_loss_mlp": 0.01025975,
      "balance_loss_clip": 1.04869592,
      "balance_loss_mlp": 1.01849437,
      "epoch": 0.4678650874767029,
      "flos": 17967689141760.0,
      "grad_norm": 1.8541916985298201,
      "language_loss": 0.89638579,
      "learning_rate": 2.3041719202289794e-06,
      "loss": 0.91832691,
      "num_input_tokens_seen": 83725910,
      "step": 3891,
      "time_per_iteration": 2.5900015830993652
    },
    {
      "auxiliary_loss_clip": 0.0117206,
      "auxiliary_loss_mlp": 0.01029569,
      "balance_loss_clip": 1.05198574,
      "balance_loss_mlp": 1.0222789,
      "epoch": 0.467985330367342,
      "flos": 21360816432000.0,
      "grad_norm": 1.7649414595063042,
      "language_loss": 0.80657291,
      "learning_rate": 2.30340198709654e-06,
      "loss": 0.8285892,
      "num_input_tokens_seen": 83745745,
      "step": 3892,
      "time_per_iteration": 2.6073925495147705
    },
    {
      "auxiliary_loss_clip": 0.01157018,
      "auxiliary_loss_mlp": 0.01029083,
      "balance_loss_clip": 1.04576385,
      "balance_loss_mlp": 1.0211978,
      "epoch": 0.46810557325798113,
      "flos": 20521835487360.0,
      "grad_norm": 2.108018237107952,
      "language_loss": 0.74169803,
      "learning_rate": 2.3026320079382605e-06,
      "loss": 0.76355904,
      "num_input_tokens_seen": 83762680,
      "step": 3893,
      "time_per_iteration": 2.6797237396240234
    },
    {
      "auxiliary_loss_clip": 0.0118206,
      "auxiliary_loss_mlp": 0.01022861,
      "balance_loss_clip": 1.05103076,
      "balance_loss_mlp": 1.01473653,
      "epoch": 0.4682258161486202,
      "flos": 30117848572800.0,
      "grad_norm": 1.9168360487852627,
      "language_loss": 0.76515847,
      "learning_rate": 2.3018619828709454e-06,
      "loss": 0.78720766,
      "num_input_tokens_seen": 83784220,
      "step": 3894,
      "time_per_iteration": 2.694000244140625
    },
    {
      "auxiliary_loss_clip": 0.01169687,
      "auxiliary_loss_mlp": 0.00763253,
      "balance_loss_clip": 1.05257738,
      "balance_loss_mlp": 1.00016057,
      "epoch": 0.4683460590392593,
      "flos": 25293357239040.0,
      "grad_norm": 2.2012067370136905,
      "language_loss": 0.82024956,
      "learning_rate": 2.3010919120114084e-06,
      "loss": 0.83957899,
      "num_input_tokens_seen": 83800750,
      "step": 3895,
      "time_per_iteration": 2.7824513912200928
    },
    {
      "auxiliary_loss_clip": 0.01163691,
      "auxiliary_loss_mlp": 0.01029344,
      "balance_loss_clip": 1.04440665,
      "balance_loss_mlp": 1.0208745,
      "epoch": 0.4684663019298984,
      "flos": 15368330551680.0,
      "grad_norm": 2.1787387031238503,
      "language_loss": 0.6587981,
      "learning_rate": 2.3003217954764672e-06,
      "loss": 0.68072844,
      "num_input_tokens_seen": 83815455,
      "step": 3896,
      "time_per_iteration": 2.6696741580963135
    },
    {
      "auxiliary_loss_clip": 0.0117025,
      "auxiliary_loss_mlp": 0.01027902,
      "balance_loss_clip": 1.04634356,
      "balance_loss_mlp": 1.01986194,
      "epoch": 0.46858654482053747,
      "flos": 27778842737280.0,
      "grad_norm": 1.7695336474781151,
      "language_loss": 0.79493558,
      "learning_rate": 2.299551633382949e-06,
      "loss": 0.81691706,
      "num_input_tokens_seen": 83835765,
      "step": 3897,
      "time_per_iteration": 2.668097972869873
    },
    {
      "auxiliary_loss_clip": 0.01149287,
      "auxiliary_loss_mlp": 0.01027988,
      "balance_loss_clip": 1.04658496,
      "balance_loss_mlp": 1.01961315,
      "epoch": 0.4687067877111766,
      "flos": 18040623707520.0,
      "grad_norm": 1.778320475914632,
      "language_loss": 0.85807306,
      "learning_rate": 2.2987814258476854e-06,
      "loss": 0.8798458,
      "num_input_tokens_seen": 83853565,
      "step": 3898,
      "time_per_iteration": 2.605916976928711
    },
    {
      "auxiliary_loss_clip": 0.01130135,
      "auxiliary_loss_mlp": 0.01034382,
      "balance_loss_clip": 1.04291177,
      "balance_loss_mlp": 1.02641273,
      "epoch": 0.4688270306018157,
      "flos": 16977380198400.0,
      "grad_norm": 2.4184816217833056,
      "language_loss": 0.68383425,
      "learning_rate": 2.2980111729875177e-06,
      "loss": 0.7054795,
      "num_input_tokens_seen": 83869815,
      "step": 3899,
      "time_per_iteration": 2.7167932987213135
    },
    {
      "auxiliary_loss_clip": 0.01150649,
      "auxiliary_loss_mlp": 0.010267,
      "balance_loss_clip": 1.04990554,
      "balance_loss_mlp": 1.01883805,
      "epoch": 0.46894727349245474,
      "flos": 17821640442240.0,
      "grad_norm": 1.617673316916541,
      "language_loss": 0.82571316,
      "learning_rate": 2.2972408749192917e-06,
      "loss": 0.84748667,
      "num_input_tokens_seen": 83887545,
      "step": 3900,
      "time_per_iteration": 2.6457011699676514
    },
    {
      "auxiliary_loss_clip": 0.01165559,
      "auxiliary_loss_mlp": 0.00763178,
      "balance_loss_clip": 1.04956901,
      "balance_loss_mlp": 1.00016665,
      "epoch": 0.46906751638309385,
      "flos": 21471349559040.0,
      "grad_norm": 2.13158539933793,
      "language_loss": 0.67168552,
      "learning_rate": 2.296470531759861e-06,
      "loss": 0.69097292,
      "num_input_tokens_seen": 83905645,
      "step": 3901,
      "time_per_iteration": 2.69840407371521
    },
    {
      "auxiliary_loss_clip": 0.01134125,
      "auxiliary_loss_mlp": 0.01024341,
      "balance_loss_clip": 1.04497468,
      "balance_loss_mlp": 1.01633036,
      "epoch": 0.46918775927373296,
      "flos": 20337829090560.0,
      "grad_norm": 1.7854642589681655,
      "language_loss": 0.79367745,
      "learning_rate": 2.2957001436260866e-06,
      "loss": 0.81526214,
      "num_input_tokens_seen": 83922705,
      "step": 3902,
      "time_per_iteration": 2.6706223487854004
    },
    {
      "auxiliary_loss_clip": 0.01153413,
      "auxiliary_loss_mlp": 0.01028044,
      "balance_loss_clip": 1.04906332,
      "balance_loss_mlp": 1.01999712,
      "epoch": 0.469308002164372,
      "flos": 18403249461120.0,
      "grad_norm": 1.5498450378757413,
      "language_loss": 0.73447442,
      "learning_rate": 2.294929710634836e-06,
      "loss": 0.75628901,
      "num_input_tokens_seen": 83940795,
      "step": 3903,
      "time_per_iteration": 2.692601203918457
    },
    {
      "auxiliary_loss_clip": 0.01165487,
      "auxiliary_loss_mlp": 0.01025806,
      "balance_loss_clip": 1.04518104,
      "balance_loss_mlp": 1.01788425,
      "epoch": 0.46942824505501113,
      "flos": 37962067363200.0,
      "grad_norm": 1.786793108020436,
      "language_loss": 0.61168468,
      "learning_rate": 2.2941592329029823e-06,
      "loss": 0.63359761,
      "num_input_tokens_seen": 83961900,
      "step": 3904,
      "time_per_iteration": 2.735229015350342
    },
    {
      "auxiliary_loss_clip": 0.01166923,
      "auxiliary_loss_mlp": 0.01032009,
      "balance_loss_clip": 1.05068111,
      "balance_loss_mlp": 1.02405143,
      "epoch": 0.46954848794565024,
      "flos": 21872507627520.0,
      "grad_norm": 1.8922636607638492,
      "language_loss": 0.79251981,
      "learning_rate": 2.2933887105474067e-06,
      "loss": 0.81450915,
      "num_input_tokens_seen": 83980075,
      "step": 3905,
      "time_per_iteration": 2.6590332984924316
    },
    {
      "auxiliary_loss_clip": 0.01164129,
      "auxiliary_loss_mlp": 0.01032254,
      "balance_loss_clip": 1.04933083,
      "balance_loss_mlp": 1.02444553,
      "epoch": 0.4696687308362893,
      "flos": 22016545165440.0,
      "grad_norm": 1.634207241718594,
      "language_loss": 0.81484556,
      "learning_rate": 2.2926181436849974e-06,
      "loss": 0.8368094,
      "num_input_tokens_seen": 83999430,
      "step": 3906,
      "time_per_iteration": 2.6255295276641846
    },
    {
      "auxiliary_loss_clip": 0.01166856,
      "auxiliary_loss_mlp": 0.01027127,
      "balance_loss_clip": 1.04973722,
      "balance_loss_mlp": 1.01880598,
      "epoch": 0.4697889737269284,
      "flos": 21613663244160.0,
      "grad_norm": 1.6836581439967466,
      "language_loss": 0.72499716,
      "learning_rate": 2.2918475324326478e-06,
      "loss": 0.74693704,
      "num_input_tokens_seen": 84019150,
      "step": 3907,
      "time_per_iteration": 2.682119369506836
    },
    {
      "auxiliary_loss_clip": 0.01175315,
      "auxiliary_loss_mlp": 0.00763528,
      "balance_loss_clip": 1.05156171,
      "balance_loss_mlp": 1.00028419,
      "epoch": 0.46990921661756746,
      "flos": 25228323665280.0,
      "grad_norm": 2.0634771656570883,
      "language_loss": 0.91505826,
      "learning_rate": 2.2910768769072603e-06,
      "loss": 0.93444669,
      "num_input_tokens_seen": 84037930,
      "step": 3908,
      "time_per_iteration": 2.720750570297241
    },
    {
      "auxiliary_loss_clip": 0.01161812,
      "auxiliary_loss_mlp": 0.01029466,
      "balance_loss_clip": 1.04704618,
      "balance_loss_mlp": 1.02150273,
      "epoch": 0.47002945950820657,
      "flos": 13844031045120.0,
      "grad_norm": 1.9259247386896547,
      "language_loss": 0.7612645,
      "learning_rate": 2.2903061772257417e-06,
      "loss": 0.78317726,
      "num_input_tokens_seen": 84055915,
      "step": 3909,
      "time_per_iteration": 2.6470632553100586
    },
    {
      "auxiliary_loss_clip": 0.01168575,
      "auxiliary_loss_mlp": 0.01026477,
      "balance_loss_clip": 1.0502249,
      "balance_loss_mlp": 1.01810861,
      "epoch": 0.4701497023988457,
      "flos": 26247001374720.0,
      "grad_norm": 1.6581451009108852,
      "language_loss": 0.7870419,
      "learning_rate": 2.289535433505007e-06,
      "loss": 0.80899239,
      "num_input_tokens_seen": 84077270,
      "step": 3910,
      "time_per_iteration": 2.6944665908813477
    },
    {
      "auxiliary_loss_clip": 0.01156262,
      "auxiliary_loss_mlp": 0.01031856,
      "balance_loss_clip": 1.04566669,
      "balance_loss_mlp": 1.02371955,
      "epoch": 0.47026994528948474,
      "flos": 25629517647360.0,
      "grad_norm": 1.8233746763570382,
      "language_loss": 0.6375317,
      "learning_rate": 2.2887646458619767e-06,
      "loss": 0.65941286,
      "num_input_tokens_seen": 84098635,
      "step": 3911,
      "time_per_iteration": 3.6398494243621826
    },
    {
      "auxiliary_loss_clip": 0.01145502,
      "auxiliary_loss_mlp": 0.01036601,
      "balance_loss_clip": 1.04710388,
      "balance_loss_mlp": 1.02777982,
      "epoch": 0.47039018818012385,
      "flos": 20554406144640.0,
      "grad_norm": 3.035679892845977,
      "language_loss": 0.76374543,
      "learning_rate": 2.2879938144135797e-06,
      "loss": 0.78556645,
      "num_input_tokens_seen": 84114740,
      "step": 3912,
      "time_per_iteration": 2.7206196784973145
    },
    {
      "auxiliary_loss_clip": 0.01137661,
      "auxiliary_loss_mlp": 0.00762606,
      "balance_loss_clip": 1.04405713,
      "balance_loss_mlp": 1.00015223,
      "epoch": 0.47051043107076296,
      "flos": 21577249831680.0,
      "grad_norm": 1.5977673217300887,
      "language_loss": 0.7528193,
      "learning_rate": 2.2872229392767496e-06,
      "loss": 0.77182204,
      "num_input_tokens_seen": 84134845,
      "step": 3913,
      "time_per_iteration": 2.659719944000244
    },
    {
      "auxiliary_loss_clip": 0.01172406,
      "auxiliary_loss_mlp": 0.01029654,
      "balance_loss_clip": 1.04964089,
      "balance_loss_mlp": 1.02153611,
      "epoch": 0.470630673961402,
      "flos": 18953185662720.0,
      "grad_norm": 1.6367482835182665,
      "language_loss": 0.74737841,
      "learning_rate": 2.286452020568428e-06,
      "loss": 0.76939899,
      "num_input_tokens_seen": 84152920,
      "step": 3914,
      "time_per_iteration": 2.616288661956787
    },
    {
      "auxiliary_loss_clip": 0.01189779,
      "auxiliary_loss_mlp": 0.01031175,
      "balance_loss_clip": 1.05239248,
      "balance_loss_mlp": 1.02278233,
      "epoch": 0.4707509168520411,
      "flos": 19938969492480.0,
      "grad_norm": 1.7819579020675602,
      "language_loss": 0.73141491,
      "learning_rate": 2.2856810584055637e-06,
      "loss": 0.75362444,
      "num_input_tokens_seen": 84170455,
      "step": 3915,
      "time_per_iteration": 3.4468448162078857
    },
    {
      "auxiliary_loss_clip": 0.01170785,
      "auxiliary_loss_mlp": 0.01031332,
      "balance_loss_clip": 1.04952776,
      "balance_loss_mlp": 1.02316654,
      "epoch": 0.47087115974268023,
      "flos": 40118754741120.0,
      "grad_norm": 1.536206092465581,
      "language_loss": 0.67632687,
      "learning_rate": 2.2849100529051085e-06,
      "loss": 0.69834805,
      "num_input_tokens_seen": 84197390,
      "step": 3916,
      "time_per_iteration": 3.705092430114746
    },
    {
      "auxiliary_loss_clip": 0.01180099,
      "auxiliary_loss_mlp": 0.01025438,
      "balance_loss_clip": 1.04926872,
      "balance_loss_mlp": 1.01720631,
      "epoch": 0.4709914026333193,
      "flos": 13552723745280.0,
      "grad_norm": 2.342345401020367,
      "language_loss": 0.80124813,
      "learning_rate": 2.284139004184026e-06,
      "loss": 0.82330346,
      "num_input_tokens_seen": 84214620,
      "step": 3917,
      "time_per_iteration": 2.5804765224456787
    },
    {
      "auxiliary_loss_clip": 0.01183274,
      "auxiliary_loss_mlp": 0.01028961,
      "balance_loss_clip": 1.04957032,
      "balance_loss_mlp": 1.02077103,
      "epoch": 0.4711116455239584,
      "flos": 19974628719360.0,
      "grad_norm": 1.969366981393763,
      "language_loss": 0.74122071,
      "learning_rate": 2.2833679123592814e-06,
      "loss": 0.7633431,
      "num_input_tokens_seen": 84231880,
      "step": 3918,
      "time_per_iteration": 2.5867271423339844
    },
    {
      "auxiliary_loss_clip": 0.01153748,
      "auxiliary_loss_mlp": 0.01031479,
      "balance_loss_clip": 1.04872799,
      "balance_loss_mlp": 1.0234921,
      "epoch": 0.4712318884145975,
      "flos": 32124824064000.0,
      "grad_norm": 1.7794949584426258,
      "language_loss": 0.63665456,
      "learning_rate": 2.2825967775478508e-06,
      "loss": 0.65850687,
      "num_input_tokens_seen": 84252980,
      "step": 3919,
      "time_per_iteration": 2.921816349029541
    },
    {
      "auxiliary_loss_clip": 0.01180289,
      "auxiliary_loss_mlp": 0.01027683,
      "balance_loss_clip": 1.04776347,
      "balance_loss_mlp": 1.01981497,
      "epoch": 0.47135213130523657,
      "flos": 20047850593920.0,
      "grad_norm": 2.022245965816566,
      "language_loss": 0.83098698,
      "learning_rate": 2.2818255998667135e-06,
      "loss": 0.85306668,
      "num_input_tokens_seen": 84271490,
      "step": 3920,
      "time_per_iteration": 2.5784261226654053
    },
    {
      "auxiliary_loss_clip": 0.01168472,
      "auxiliary_loss_mlp": 0.01030307,
      "balance_loss_clip": 1.05042183,
      "balance_loss_mlp": 1.02193856,
      "epoch": 0.4714723741958757,
      "flos": 19426990988160.0,
      "grad_norm": 1.630558804238722,
      "language_loss": 0.79044384,
      "learning_rate": 2.2810543794328566e-06,
      "loss": 0.81243157,
      "num_input_tokens_seen": 84290525,
      "step": 3921,
      "time_per_iteration": 2.6516711711883545
    },
    {
      "auxiliary_loss_clip": 0.01173993,
      "auxiliary_loss_mlp": 0.0103224,
      "balance_loss_clip": 1.05035472,
      "balance_loss_mlp": 1.02370417,
      "epoch": 0.4715926170865148,
      "flos": 20373883367040.0,
      "grad_norm": 1.8886906200965397,
      "language_loss": 0.82782489,
      "learning_rate": 2.2802831163632735e-06,
      "loss": 0.84988725,
      "num_input_tokens_seen": 84309245,
      "step": 3922,
      "time_per_iteration": 2.7843987941741943
    },
    {
      "auxiliary_loss_clip": 0.01114266,
      "auxiliary_loss_mlp": 0.01031733,
      "balance_loss_clip": 1.04333627,
      "balance_loss_mlp": 1.02314067,
      "epoch": 0.47171285997715384,
      "flos": 22672884430080.0,
      "grad_norm": 2.0459884281006318,
      "language_loss": 0.74210852,
      "learning_rate": 2.279511810774965e-06,
      "loss": 0.76356852,
      "num_input_tokens_seen": 84330775,
      "step": 3923,
      "time_per_iteration": 2.799251079559326
    },
    {
      "auxiliary_loss_clip": 0.01183682,
      "auxiliary_loss_mlp": 0.01024621,
      "balance_loss_clip": 1.05063725,
      "balance_loss_mlp": 1.01675868,
      "epoch": 0.47183310286779295,
      "flos": 21105419754240.0,
      "grad_norm": 2.1014775191296775,
      "language_loss": 0.71568191,
      "learning_rate": 2.2787404627849364e-06,
      "loss": 0.73776489,
      "num_input_tokens_seen": 84349985,
      "step": 3924,
      "time_per_iteration": 2.628480911254883
    },
    {
      "auxiliary_loss_clip": 0.0115192,
      "auxiliary_loss_mlp": 0.01025295,
      "balance_loss_clip": 1.04588711,
      "balance_loss_mlp": 1.01751077,
      "epoch": 0.471953345758432,
      "flos": 21726566668800.0,
      "grad_norm": 2.170345459286011,
      "language_loss": 0.79207259,
      "learning_rate": 2.277969072510202e-06,
      "loss": 0.81384474,
      "num_input_tokens_seen": 84368965,
      "step": 3925,
      "time_per_iteration": 2.6631126403808594
    },
    {
      "auxiliary_loss_clip": 0.01152591,
      "auxiliary_loss_mlp": 0.01028011,
      "balance_loss_clip": 1.04680908,
      "balance_loss_mlp": 1.02054262,
      "epoch": 0.4720735886490711,
      "flos": 19861078849920.0,
      "grad_norm": 1.4996389393293528,
      "language_loss": 0.81531072,
      "learning_rate": 2.2771976400677803e-06,
      "loss": 0.83711672,
      "num_input_tokens_seen": 84387795,
      "step": 3926,
      "time_per_iteration": 2.7199976444244385
    },
    {
      "auxiliary_loss_clip": 0.01113857,
      "auxiliary_loss_mlp": 0.01029342,
      "balance_loss_clip": 1.04211104,
      "balance_loss_mlp": 1.0217185,
      "epoch": 0.47219383153971023,
      "flos": 19171809792000.0,
      "grad_norm": 1.7168396859768613,
      "language_loss": 0.79157591,
      "learning_rate": 2.2764261655746965e-06,
      "loss": 0.81300789,
      "num_input_tokens_seen": 84405290,
      "step": 3927,
      "time_per_iteration": 2.681540012359619
    },
    {
      "auxiliary_loss_clip": 0.01135496,
      "auxiliary_loss_mlp": 0.01033326,
      "balance_loss_clip": 1.04426098,
      "balance_loss_mlp": 1.02519011,
      "epoch": 0.4723140744303493,
      "flos": 23224005780480.0,
      "grad_norm": 1.627887260947625,
      "language_loss": 0.76053417,
      "learning_rate": 2.2756546491479832e-06,
      "loss": 0.78222239,
      "num_input_tokens_seen": 84426205,
      "step": 3928,
      "time_per_iteration": 2.713150978088379
    },
    {
      "auxiliary_loss_clip": 0.01183703,
      "auxiliary_loss_mlp": 0.00762944,
      "balance_loss_clip": 1.04982424,
      "balance_loss_mlp": 1.00019979,
      "epoch": 0.4724343173209884,
      "flos": 18223265387520.0,
      "grad_norm": 3.646009860075139,
      "language_loss": 0.80714303,
      "learning_rate": 2.274883090904679e-06,
      "loss": 0.82660949,
      "num_input_tokens_seen": 84443970,
      "step": 3929,
      "time_per_iteration": 2.5717389583587646
    },
    {
      "auxiliary_loss_clip": 0.01189094,
      "auxiliary_loss_mlp": 0.01028704,
      "balance_loss_clip": 1.05457163,
      "balance_loss_mlp": 1.02057958,
      "epoch": 0.4725545602116275,
      "flos": 21251037490560.0,
      "grad_norm": 2.3750285108622564,
      "language_loss": 0.67579138,
      "learning_rate": 2.2741114909618283e-06,
      "loss": 0.69796938,
      "num_input_tokens_seen": 84459865,
      "step": 3930,
      "time_per_iteration": 2.627019166946411
    },
    {
      "auxiliary_loss_clip": 0.01141065,
      "auxiliary_loss_mlp": 0.01027053,
      "balance_loss_clip": 1.04693627,
      "balance_loss_mlp": 1.01952529,
      "epoch": 0.47267480310226656,
      "flos": 21434002392960.0,
      "grad_norm": 1.8343322832021847,
      "language_loss": 0.72283751,
      "learning_rate": 2.2733398494364828e-06,
      "loss": 0.7445187,
      "num_input_tokens_seen": 84479110,
      "step": 3931,
      "time_per_iteration": 2.758167266845703
    },
    {
      "auxiliary_loss_clip": 0.01150549,
      "auxiliary_loss_mlp": 0.0102598,
      "balance_loss_clip": 1.05047798,
      "balance_loss_mlp": 1.01801121,
      "epoch": 0.47279504599290567,
      "flos": 18770508069120.0,
      "grad_norm": 1.8079465608232885,
      "language_loss": 0.84436619,
      "learning_rate": 2.272568166445699e-06,
      "loss": 0.86613148,
      "num_input_tokens_seen": 84497675,
      "step": 3932,
      "time_per_iteration": 2.7788946628570557
    },
    {
      "auxiliary_loss_clip": 0.01167735,
      "auxiliary_loss_mlp": 0.01023497,
      "balance_loss_clip": 1.04856563,
      "balance_loss_mlp": 1.01559949,
      "epoch": 0.4729152888835448,
      "flos": 21105742976640.0,
      "grad_norm": 3.557033031846401,
      "language_loss": 0.64503849,
      "learning_rate": 2.271796442106541e-06,
      "loss": 0.66695082,
      "num_input_tokens_seen": 84517030,
      "step": 3933,
      "time_per_iteration": 2.6185598373413086
    },
    {
      "auxiliary_loss_clip": 0.01052402,
      "auxiliary_loss_mlp": 0.01005275,
      "balance_loss_clip": 1.02016342,
      "balance_loss_mlp": 1.00347543,
      "epoch": 0.47303553177418384,
      "flos": 70201877840640.0,
      "grad_norm": 0.7957961616325003,
      "language_loss": 0.56526256,
      "learning_rate": 2.271024676536079e-06,
      "loss": 0.58583939,
      "num_input_tokens_seen": 84577290,
      "step": 3934,
      "time_per_iteration": 3.2556071281433105
    },
    {
      "auxiliary_loss_clip": 0.01163496,
      "auxiliary_loss_mlp": 0.01032635,
      "balance_loss_clip": 1.05332696,
      "balance_loss_mlp": 1.02358079,
      "epoch": 0.47315577466482295,
      "flos": 22455122227200.0,
      "grad_norm": 2.289915400589802,
      "language_loss": 0.73200274,
      "learning_rate": 2.2702528698513894e-06,
      "loss": 0.75396407,
      "num_input_tokens_seen": 84598415,
      "step": 3935,
      "time_per_iteration": 2.697964668273926
    },
    {
      "auxiliary_loss_clip": 0.01155503,
      "auxiliary_loss_mlp": 0.01025157,
      "balance_loss_clip": 1.04581928,
      "balance_loss_mlp": 1.01672268,
      "epoch": 0.47327601755546206,
      "flos": 24352857480960.0,
      "grad_norm": 4.047232993148734,
      "language_loss": 0.79255569,
      "learning_rate": 2.269481022169554e-06,
      "loss": 0.81436229,
      "num_input_tokens_seen": 84617010,
      "step": 3936,
      "time_per_iteration": 2.699193000793457
    },
    {
      "auxiliary_loss_clip": 0.01161139,
      "auxiliary_loss_mlp": 0.0102883,
      "balance_loss_clip": 1.04756975,
      "balance_loss_mlp": 1.0205034,
      "epoch": 0.4733962604461011,
      "flos": 22926772736640.0,
      "grad_norm": 1.914430432976261,
      "language_loss": 0.8071357,
      "learning_rate": 2.2687091336076614e-06,
      "loss": 0.8290354,
      "num_input_tokens_seen": 84636350,
      "step": 3937,
      "time_per_iteration": 3.6518943309783936
    },
    {
      "auxiliary_loss_clip": 0.01168252,
      "auxiliary_loss_mlp": 0.01029308,
      "balance_loss_clip": 1.0506041,
      "balance_loss_mlp": 1.02085054,
      "epoch": 0.4735165033367402,
      "flos": 18327369980160.0,
      "grad_norm": 2.0877700182174737,
      "language_loss": 0.80217516,
      "learning_rate": 2.267937204282807e-06,
      "loss": 0.8241508,
      "num_input_tokens_seen": 84653490,
      "step": 3938,
      "time_per_iteration": 2.6783976554870605
    },
    {
      "auxiliary_loss_clip": 0.01174044,
      "auxiliary_loss_mlp": 0.01030641,
      "balance_loss_clip": 1.05128074,
      "balance_loss_mlp": 1.02197409,
      "epoch": 0.4736367462273793,
      "flos": 23037018554880.0,
      "grad_norm": 2.1470910048598753,
      "language_loss": 0.79270804,
      "learning_rate": 2.2671652343120926e-06,
      "loss": 0.81475484,
      "num_input_tokens_seen": 84673965,
      "step": 3939,
      "time_per_iteration": 2.690129280090332
    },
    {
      "auxiliary_loss_clip": 0.0118392,
      "auxiliary_loss_mlp": 0.01026335,
      "balance_loss_clip": 1.05277848,
      "balance_loss_mlp": 1.01843762,
      "epoch": 0.4737569891180184,
      "flos": 25374336451200.0,
      "grad_norm": 1.6273739982711202,
      "language_loss": 0.80745655,
      "learning_rate": 2.2663932238126236e-06,
      "loss": 0.82955909,
      "num_input_tokens_seen": 84692525,
      "step": 3940,
      "time_per_iteration": 3.4611480236053467
    },
    {
      "auxiliary_loss_clip": 0.01166345,
      "auxiliary_loss_mlp": 0.01026468,
      "balance_loss_clip": 1.046597,
      "balance_loss_mlp": 1.01802766,
      "epoch": 0.4738772320086575,
      "flos": 25849326925440.0,
      "grad_norm": 1.5376727522629519,
      "language_loss": 0.80144477,
      "learning_rate": 2.265621172901515e-06,
      "loss": 0.82337296,
      "num_input_tokens_seen": 84715640,
      "step": 3941,
      "time_per_iteration": 2.720466136932373
    },
    {
      "auxiliary_loss_clip": 0.01189161,
      "auxiliary_loss_mlp": 0.01030734,
      "balance_loss_clip": 1.05593491,
      "balance_loss_mlp": 1.02232373,
      "epoch": 0.47399747489929656,
      "flos": 27564420499200.0,
      "grad_norm": 2.6342842132308184,
      "language_loss": 0.71447343,
      "learning_rate": 2.2648490816958854e-06,
      "loss": 0.7366724,
      "num_input_tokens_seen": 84736635,
      "step": 3942,
      "time_per_iteration": 3.5601305961608887
    },
    {
      "auxiliary_loss_clip": 0.01170294,
      "auxiliary_loss_mlp": 0.01031472,
      "balance_loss_clip": 1.05022335,
      "balance_loss_mlp": 1.02270448,
      "epoch": 0.47411771778993567,
      "flos": 24863650836480.0,
      "grad_norm": 2.3813178247295452,
      "language_loss": 0.73565108,
      "learning_rate": 2.264076950312861e-06,
      "loss": 0.75766873,
      "num_input_tokens_seen": 84755445,
      "step": 3943,
      "time_per_iteration": 2.700568199157715
    },
    {
      "auxiliary_loss_clip": 0.01161089,
      "auxiliary_loss_mlp": 0.01029582,
      "balance_loss_clip": 1.04977965,
      "balance_loss_mlp": 1.02170253,
      "epoch": 0.4742379606805748,
      "flos": 22748009725440.0,
      "grad_norm": 2.022951617531913,
      "language_loss": 0.82670105,
      "learning_rate": 2.2633047788695727e-06,
      "loss": 0.84860772,
      "num_input_tokens_seen": 84775750,
      "step": 3944,
      "time_per_iteration": 2.667673349380493
    },
    {
      "auxiliary_loss_clip": 0.01153035,
      "auxiliary_loss_mlp": 0.01023714,
      "balance_loss_clip": 1.04810643,
      "balance_loss_mlp": 1.01634097,
      "epoch": 0.47435820357121383,
      "flos": 19681130689920.0,
      "grad_norm": 1.8325208015627505,
      "language_loss": 0.64043617,
      "learning_rate": 2.262532567483159e-06,
      "loss": 0.66220361,
      "num_input_tokens_seen": 84794310,
      "step": 3945,
      "time_per_iteration": 2.670921564102173
    },
    {
      "auxiliary_loss_clip": 0.01186119,
      "auxiliary_loss_mlp": 0.00763127,
      "balance_loss_clip": 1.0538888,
      "balance_loss_mlp": 1.00016093,
      "epoch": 0.47447844646185294,
      "flos": 25228718714880.0,
      "grad_norm": 1.7756056218117124,
      "language_loss": 0.80099446,
      "learning_rate": 2.2617603162707635e-06,
      "loss": 0.8204869,
      "num_input_tokens_seen": 84814720,
      "step": 3946,
      "time_per_iteration": 2.6224801540374756
    },
    {
      "auxiliary_loss_clip": 0.01183934,
      "auxiliary_loss_mlp": 0.01026882,
      "balance_loss_clip": 1.05191612,
      "balance_loss_mlp": 1.01876378,
      "epoch": 0.47459868935249205,
      "flos": 24570619683840.0,
      "grad_norm": 1.782044247930579,
      "language_loss": 0.82608724,
      "learning_rate": 2.2609880253495363e-06,
      "loss": 0.84819543,
      "num_input_tokens_seen": 84834355,
      "step": 3947,
      "time_per_iteration": 2.621115207672119
    },
    {
      "auxiliary_loss_clip": 0.01150972,
      "auxiliary_loss_mlp": 0.01034206,
      "balance_loss_clip": 1.0473671,
      "balance_loss_mlp": 1.02577138,
      "epoch": 0.4747189322431311,
      "flos": 20558500295040.0,
      "grad_norm": 1.9013329601573432,
      "language_loss": 0.86016905,
      "learning_rate": 2.260215694836633e-06,
      "loss": 0.88202083,
      "num_input_tokens_seen": 84853530,
      "step": 3948,
      "time_per_iteration": 2.7053840160369873
    },
    {
      "auxiliary_loss_clip": 0.01130913,
      "auxiliary_loss_mlp": 0.00763306,
      "balance_loss_clip": 1.04472136,
      "balance_loss_mlp": 1.00015521,
      "epoch": 0.4748391751337702,
      "flos": 25995231970560.0,
      "grad_norm": 2.7956499669574817,
      "language_loss": 0.64882207,
      "learning_rate": 2.2594433248492157e-06,
      "loss": 0.66776431,
      "num_input_tokens_seen": 84872505,
      "step": 3949,
      "time_per_iteration": 2.7704782485961914
    },
    {
      "auxiliary_loss_clip": 0.01173397,
      "auxiliary_loss_mlp": 0.0102814,
      "balance_loss_clip": 1.04883182,
      "balance_loss_mlp": 1.01966977,
      "epoch": 0.47495941802440933,
      "flos": 22821052032000.0,
      "grad_norm": 2.6005216967421365,
      "language_loss": 0.80117553,
      "learning_rate": 2.2586709155044527e-06,
      "loss": 0.82319087,
      "num_input_tokens_seen": 84893105,
      "step": 3950,
      "time_per_iteration": 2.5835225582122803
    },
    {
      "auxiliary_loss_clip": 0.01185809,
      "auxiliary_loss_mlp": 0.01031311,
      "balance_loss_clip": 1.05322707,
      "balance_loss_mlp": 1.02321088,
      "epoch": 0.4750796609150484,
      "flos": 27891782075520.0,
      "grad_norm": 1.4885386052513598,
      "language_loss": 0.75740016,
      "learning_rate": 2.2578984669195167e-06,
      "loss": 0.77957141,
      "num_input_tokens_seen": 84914070,
      "step": 3951,
      "time_per_iteration": 2.6251516342163086
    },
    {
      "auxiliary_loss_clip": 0.01168615,
      "auxiliary_loss_mlp": 0.01029363,
      "balance_loss_clip": 1.04789221,
      "balance_loss_mlp": 1.02173972,
      "epoch": 0.4751999038056875,
      "flos": 35660085471360.0,
      "grad_norm": 1.9148108559994252,
      "language_loss": 0.67907393,
      "learning_rate": 2.2571259792115887e-06,
      "loss": 0.70105368,
      "num_input_tokens_seen": 84935290,
      "step": 3952,
      "time_per_iteration": 2.7447779178619385
    },
    {
      "auxiliary_loss_clip": 0.01164977,
      "auxiliary_loss_mlp": 0.01030755,
      "balance_loss_clip": 1.04939497,
      "balance_loss_mlp": 1.02336371,
      "epoch": 0.4753201466963266,
      "flos": 22090880361600.0,
      "grad_norm": 1.644245364624808,
      "language_loss": 0.79042131,
      "learning_rate": 2.2563534524978544e-06,
      "loss": 0.81237864,
      "num_input_tokens_seen": 84952760,
      "step": 3953,
      "time_per_iteration": 2.6966745853424072
    },
    {
      "auxiliary_loss_clip": 0.01138311,
      "auxiliary_loss_mlp": 0.01023986,
      "balance_loss_clip": 1.04926205,
      "balance_loss_mlp": 1.01620722,
      "epoch": 0.47544038958696566,
      "flos": 30190854965760.0,
      "grad_norm": 1.6878755648688595,
      "language_loss": 0.70576966,
      "learning_rate": 2.2555808868955052e-06,
      "loss": 0.72739261,
      "num_input_tokens_seen": 84974890,
      "step": 3954,
      "time_per_iteration": 2.7929909229278564
    },
    {
      "auxiliary_loss_clip": 0.0112712,
      "auxiliary_loss_mlp": 0.01029703,
      "balance_loss_clip": 1.04582274,
      "balance_loss_mlp": 1.02081048,
      "epoch": 0.47556063247760477,
      "flos": 23472219738240.0,
      "grad_norm": 2.202777393319252,
      "language_loss": 0.74172592,
      "learning_rate": 2.254808282521738e-06,
      "loss": 0.76329416,
      "num_input_tokens_seen": 84993640,
      "step": 3955,
      "time_per_iteration": 2.7400150299072266
    },
    {
      "auxiliary_loss_clip": 0.01145006,
      "auxiliary_loss_mlp": 0.00762155,
      "balance_loss_clip": 1.04661787,
      "balance_loss_mlp": 1.00016975,
      "epoch": 0.4756808753682438,
      "flos": 25155209531520.0,
      "grad_norm": 1.7148496679955048,
      "language_loss": 0.811948,
      "learning_rate": 2.2540356394937573e-06,
      "loss": 0.83101964,
      "num_input_tokens_seen": 85012340,
      "step": 3956,
      "time_per_iteration": 2.7114598751068115
    },
    {
      "auxiliary_loss_clip": 0.01145967,
      "auxiliary_loss_mlp": 0.01031017,
      "balance_loss_clip": 1.04621732,
      "balance_loss_mlp": 1.02252352,
      "epoch": 0.47580111825888294,
      "flos": 15669729573120.0,
      "grad_norm": 2.1763157251676803,
      "language_loss": 0.83755732,
      "learning_rate": 2.253262957928772e-06,
      "loss": 0.85932714,
      "num_input_tokens_seen": 85029225,
      "step": 3957,
      "time_per_iteration": 2.6541965007781982
    },
    {
      "auxiliary_loss_clip": 0.01150928,
      "auxiliary_loss_mlp": 0.0103156,
      "balance_loss_clip": 1.04681528,
      "balance_loss_mlp": 1.02266061,
      "epoch": 0.47592136114952205,
      "flos": 17636556637440.0,
      "grad_norm": 1.5872102878416015,
      "language_loss": 0.72143751,
      "learning_rate": 2.2524902379439976e-06,
      "loss": 0.74326241,
      "num_input_tokens_seen": 85047895,
      "step": 3958,
      "time_per_iteration": 2.69250226020813
    },
    {
      "auxiliary_loss_clip": 0.01037053,
      "auxiliary_loss_mlp": 0.01003883,
      "balance_loss_clip": 1.02920079,
      "balance_loss_mlp": 1.00219011,
      "epoch": 0.4760416040401611,
      "flos": 61417159292160.0,
      "grad_norm": 0.7428303450181923,
      "language_loss": 0.63686359,
      "learning_rate": 2.251717479656655e-06,
      "loss": 0.65727299,
      "num_input_tokens_seen": 85112690,
      "step": 3959,
      "time_per_iteration": 3.4690439701080322
    },
    {
      "auxiliary_loss_clip": 0.01186332,
      "auxiliary_loss_mlp": 0.01031048,
      "balance_loss_clip": 1.05289102,
      "balance_loss_mlp": 1.02258992,
      "epoch": 0.4761618469308002,
      "flos": 18405871153920.0,
      "grad_norm": 5.154605544053129,
      "language_loss": 0.76413631,
      "learning_rate": 2.2509446831839704e-06,
      "loss": 0.78631008,
      "num_input_tokens_seen": 85132130,
      "step": 3960,
      "time_per_iteration": 3.261329174041748
    },
    {
      "auxiliary_loss_clip": 0.01160005,
      "auxiliary_loss_mlp": 0.0102779,
      "balance_loss_clip": 1.04909039,
      "balance_loss_mlp": 1.01982713,
      "epoch": 0.4762820898214393,
      "flos": 18040911016320.0,
      "grad_norm": 2.2185557167431513,
      "language_loss": 0.82768124,
      "learning_rate": 2.250171848643177e-06,
      "loss": 0.84955919,
      "num_input_tokens_seen": 85149420,
      "step": 3961,
      "time_per_iteration": 2.6906702518463135
    },
    {
      "auxiliary_loss_clip": 0.01152141,
      "auxiliary_loss_mlp": 0.01026868,
      "balance_loss_clip": 1.05023694,
      "balance_loss_mlp": 1.01923835,
      "epoch": 0.4764023327120784,
      "flos": 19318253541120.0,
      "grad_norm": 1.6723456047537,
      "language_loss": 0.86036706,
      "learning_rate": 2.249398976151513e-06,
      "loss": 0.88215721,
      "num_input_tokens_seen": 85166970,
      "step": 3962,
      "time_per_iteration": 2.654501438140869
    },
    {
      "auxiliary_loss_clip": 0.01181643,
      "auxiliary_loss_mlp": 0.01029824,
      "balance_loss_clip": 1.05188119,
      "balance_loss_mlp": 1.02138948,
      "epoch": 0.4765225756027175,
      "flos": 22747255539840.0,
      "grad_norm": 2.0285725811347417,
      "language_loss": 0.78468478,
      "learning_rate": 2.248626065826223e-06,
      "loss": 0.80679941,
      "num_input_tokens_seen": 85185175,
      "step": 3963,
      "time_per_iteration": 4.569107294082642
    },
    {
      "auxiliary_loss_clip": 0.01088953,
      "auxiliary_loss_mlp": 0.01001974,
      "balance_loss_clip": 1.02148819,
      "balance_loss_mlp": 1.00017357,
      "epoch": 0.4766428184933566,
      "flos": 65933392106880.0,
      "grad_norm": 0.763417656487407,
      "language_loss": 0.62573355,
      "learning_rate": 2.2478531177845564e-06,
      "loss": 0.6466428,
      "num_input_tokens_seen": 85246170,
      "step": 3964,
      "time_per_iteration": 3.1997880935668945
    },
    {
      "auxiliary_loss_clip": 0.01159765,
      "auxiliary_loss_mlp": 0.01028209,
      "balance_loss_clip": 1.05250192,
      "balance_loss_mlp": 1.01988173,
      "epoch": 0.47676306138399566,
      "flos": 24136495908480.0,
      "grad_norm": 1.6581597361631368,
      "language_loss": 0.84920394,
      "learning_rate": 2.247080132143769e-06,
      "loss": 0.87108368,
      "num_input_tokens_seen": 85268525,
      "step": 3965,
      "time_per_iteration": 2.7295210361480713
    },
    {
      "auxiliary_loss_clip": 0.01140742,
      "auxiliary_loss_mlp": 0.01028754,
      "balance_loss_clip": 1.04327941,
      "balance_loss_mlp": 1.02027273,
      "epoch": 0.47688330427463477,
      "flos": 12604322995200.0,
      "grad_norm": 2.4499439906313856,
      "language_loss": 0.69355035,
      "learning_rate": 2.246307109021121e-06,
      "loss": 0.71524537,
      "num_input_tokens_seen": 85285930,
      "step": 3966,
      "time_per_iteration": 2.7127819061279297
    },
    {
      "auxiliary_loss_clip": 0.01150482,
      "auxiliary_loss_mlp": 0.01035425,
      "balance_loss_clip": 1.04583359,
      "balance_loss_mlp": 1.027843,
      "epoch": 0.4770035471652739,
      "flos": 21390585828480.0,
      "grad_norm": 1.811836212587798,
      "language_loss": 0.82751036,
      "learning_rate": 2.2455340485338817e-06,
      "loss": 0.84936947,
      "num_input_tokens_seen": 85303565,
      "step": 3967,
      "time_per_iteration": 4.940474033355713
    },
    {
      "auxiliary_loss_clip": 0.01171105,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.04934001,
      "balance_loss_mlp": 1.02257288,
      "epoch": 0.47712379005591293,
      "flos": 25156251025920.0,
      "grad_norm": 2.0412288515019514,
      "language_loss": 0.67681575,
      "learning_rate": 2.244760950799322e-06,
      "loss": 0.69883144,
      "num_input_tokens_seen": 85321835,
      "step": 3968,
      "time_per_iteration": 2.7316582202911377
    },
    {
      "auxiliary_loss_clip": 0.01128029,
      "auxiliary_loss_mlp": 0.01027356,
      "balance_loss_clip": 1.0465343,
      "balance_loss_mlp": 1.01929212,
      "epoch": 0.47724403294655204,
      "flos": 22054323294720.0,
      "grad_norm": 2.020654857760967,
      "language_loss": 0.72624123,
      "learning_rate": 2.2439878159347203e-06,
      "loss": 0.7477951,
      "num_input_tokens_seen": 85341260,
      "step": 3969,
      "time_per_iteration": 2.753474473953247
    },
    {
      "auxiliary_loss_clip": 0.01088554,
      "auxiliary_loss_mlp": 0.01001308,
      "balance_loss_clip": 1.02105451,
      "balance_loss_mlp": 0.99955606,
      "epoch": 0.4773642758371911,
      "flos": 70229387658240.0,
      "grad_norm": 0.7307679284108406,
      "language_loss": 0.55236185,
      "learning_rate": 2.2432146440573616e-06,
      "loss": 0.57326049,
      "num_input_tokens_seen": 85407220,
      "step": 3970,
      "time_per_iteration": 3.3317105770111084
    },
    {
      "auxiliary_loss_clip": 0.01154744,
      "auxiliary_loss_mlp": 0.0103099,
      "balance_loss_clip": 1.048967,
      "balance_loss_mlp": 1.02278197,
      "epoch": 0.4774845187278302,
      "flos": 23548602009600.0,
      "grad_norm": 1.835564883086054,
      "language_loss": 0.6651659,
      "learning_rate": 2.242441435284534e-06,
      "loss": 0.68702322,
      "num_input_tokens_seen": 85426095,
      "step": 3971,
      "time_per_iteration": 2.7350192070007324
    },
    {
      "auxiliary_loss_clip": 0.01172048,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.05150771,
      "balance_loss_mlp": 1.01732373,
      "epoch": 0.4776047616184693,
      "flos": 23075371301760.0,
      "grad_norm": 2.172577641383062,
      "language_loss": 0.85641766,
      "learning_rate": 2.2416681897335337e-06,
      "loss": 0.87838995,
      "num_input_tokens_seen": 85444245,
      "step": 3972,
      "time_per_iteration": 2.631610155105591
    },
    {
      "auxiliary_loss_clip": 0.01126753,
      "auxiliary_loss_mlp": 0.01029949,
      "balance_loss_clip": 1.04761493,
      "balance_loss_mlp": 1.02182519,
      "epoch": 0.4777250045091084,
      "flos": 31898119374720.0,
      "grad_norm": 2.0953834862242404,
      "language_loss": 0.67016757,
      "learning_rate": 2.240894907521661e-06,
      "loss": 0.69173467,
      "num_input_tokens_seen": 85463325,
      "step": 3973,
      "time_per_iteration": 2.8140227794647217
    },
    {
      "auxiliary_loss_clip": 0.01157635,
      "auxiliary_loss_mlp": 0.01025695,
      "balance_loss_clip": 1.04873371,
      "balance_loss_mlp": 1.01751089,
      "epoch": 0.4778452473997475,
      "flos": 24278163148800.0,
      "grad_norm": 1.5874780186685356,
      "language_loss": 0.64036131,
      "learning_rate": 2.240121588766223e-06,
      "loss": 0.66219461,
      "num_input_tokens_seen": 85483375,
      "step": 3974,
      "time_per_iteration": 2.704516887664795
    },
    {
      "auxiliary_loss_clip": 0.01147859,
      "auxiliary_loss_mlp": 0.01031583,
      "balance_loss_clip": 1.04514384,
      "balance_loss_mlp": 1.02382255,
      "epoch": 0.4779654902903866,
      "flos": 31575031516800.0,
      "grad_norm": 1.6885560856438555,
      "language_loss": 0.71547079,
      "learning_rate": 2.239348233584531e-06,
      "loss": 0.73726517,
      "num_input_tokens_seen": 85504230,
      "step": 3975,
      "time_per_iteration": 2.7292416095733643
    },
    {
      "auxiliary_loss_clip": 0.01170641,
      "auxiliary_loss_mlp": 0.01032767,
      "balance_loss_clip": 1.04973602,
      "balance_loss_mlp": 1.02439857,
      "epoch": 0.47808573318102565,
      "flos": 19500428344320.0,
      "grad_norm": 2.1692501315324764,
      "language_loss": 0.80703193,
      "learning_rate": 2.2385748420939013e-06,
      "loss": 0.82906604,
      "num_input_tokens_seen": 85523425,
      "step": 3976,
      "time_per_iteration": 2.6487655639648438
    },
    {
      "auxiliary_loss_clip": 0.01184951,
      "auxiliary_loss_mlp": 0.01029888,
      "balance_loss_clip": 1.05501246,
      "balance_loss_mlp": 1.02249086,
      "epoch": 0.47820597607166476,
      "flos": 22601135013120.0,
      "grad_norm": 1.7530449769399241,
      "language_loss": 0.72219914,
      "learning_rate": 2.2378014144116583e-06,
      "loss": 0.74434751,
      "num_input_tokens_seen": 85542235,
      "step": 3977,
      "time_per_iteration": 2.5624608993530273
    },
    {
      "auxiliary_loss_clip": 0.01187614,
      "auxiliary_loss_mlp": 0.01034714,
      "balance_loss_clip": 1.05328894,
      "balance_loss_mlp": 1.02633929,
      "epoch": 0.4783262189623039,
      "flos": 23003011353600.0,
      "grad_norm": 2.751280837987753,
      "language_loss": 0.79615599,
      "learning_rate": 2.23702795065513e-06,
      "loss": 0.81837928,
      "num_input_tokens_seen": 85561815,
      "step": 3978,
      "time_per_iteration": 2.655606985092163
    },
    {
      "auxiliary_loss_clip": 0.01077564,
      "auxiliary_loss_mlp": 0.01001803,
      "balance_loss_clip": 1.01915193,
      "balance_loss_mlp": 1.00001454,
      "epoch": 0.47844646185294293,
      "flos": 49772801226240.0,
      "grad_norm": 0.980017192752197,
      "language_loss": 0.67464799,
      "learning_rate": 2.2362544509416493e-06,
      "loss": 0.69544172,
      "num_input_tokens_seen": 85613930,
      "step": 3979,
      "time_per_iteration": 3.1027910709381104
    },
    {
      "auxiliary_loss_clip": 0.01147939,
      "auxiliary_loss_mlp": 0.0102483,
      "balance_loss_clip": 1.04466128,
      "balance_loss_mlp": 1.01717114,
      "epoch": 0.47856670474358204,
      "flos": 20229558520320.0,
      "grad_norm": 5.058814830068448,
      "language_loss": 0.82787633,
      "learning_rate": 2.2354809153885572e-06,
      "loss": 0.84960407,
      "num_input_tokens_seen": 85631000,
      "step": 3980,
      "time_per_iteration": 2.757737398147583
    },
    {
      "auxiliary_loss_clip": 0.01167614,
      "auxiliary_loss_mlp": 0.01026049,
      "balance_loss_clip": 1.04881454,
      "balance_loss_mlp": 1.01822329,
      "epoch": 0.47868694763422115,
      "flos": 20990936131200.0,
      "grad_norm": 2.672648321257097,
      "language_loss": 0.83217156,
      "learning_rate": 2.234707344113197e-06,
      "loss": 0.85410815,
      "num_input_tokens_seen": 85649095,
      "step": 3981,
      "time_per_iteration": 2.656545877456665
    },
    {
      "auxiliary_loss_clip": 0.01181481,
      "auxiliary_loss_mlp": 0.01025556,
      "balance_loss_clip": 1.05206263,
      "balance_loss_mlp": 1.01776826,
      "epoch": 0.4788071905248602,
      "flos": 19026551191680.0,
      "grad_norm": 1.9699259707428296,
      "language_loss": 0.77855366,
      "learning_rate": 2.233933737232919e-06,
      "loss": 0.80062401,
      "num_input_tokens_seen": 85666875,
      "step": 3982,
      "time_per_iteration": 2.6217846870422363
    },
    {
      "auxiliary_loss_clip": 0.01119425,
      "auxiliary_loss_mlp": 0.00762681,
      "balance_loss_clip": 1.04392695,
      "balance_loss_mlp": 1.00018632,
      "epoch": 0.4789274334154993,
      "flos": 23002221254400.0,
      "grad_norm": 1.6698070940386518,
      "language_loss": 0.78023398,
      "learning_rate": 2.2331600948650793e-06,
      "loss": 0.79905504,
      "num_input_tokens_seen": 85687020,
      "step": 3983,
      "time_per_iteration": 2.7054078578948975
    },
    {
      "auxiliary_loss_clip": 0.01131929,
      "auxiliary_loss_mlp": 0.00764043,
      "balance_loss_clip": 1.04641318,
      "balance_loss_mlp": 1.0002867,
      "epoch": 0.4790476763061384,
      "flos": 23075586783360.0,
      "grad_norm": 2.6743920430907537,
      "language_loss": 0.80032361,
      "learning_rate": 2.2323864171270386e-06,
      "loss": 0.81928331,
      "num_input_tokens_seen": 85708290,
      "step": 3984,
      "time_per_iteration": 2.742558717727661
    },
    {
      "auxiliary_loss_clip": 0.01143536,
      "auxiliary_loss_mlp": 0.0102948,
      "balance_loss_clip": 1.04518557,
      "balance_loss_mlp": 1.02111125,
      "epoch": 0.4791679191967775,
      "flos": 21179288073600.0,
      "grad_norm": 1.9244381340507446,
      "language_loss": 0.73060048,
      "learning_rate": 2.231612704136164e-06,
      "loss": 0.75233054,
      "num_input_tokens_seen": 85728660,
      "step": 3985,
      "time_per_iteration": 2.728558301925659
    },
    {
      "auxiliary_loss_clip": 0.0116403,
      "auxiliary_loss_mlp": 0.01028821,
      "balance_loss_clip": 1.0484364,
      "balance_loss_mlp": 1.02010715,
      "epoch": 0.4792881620874166,
      "flos": 22301495758080.0,
      "grad_norm": 3.829164013013086,
      "language_loss": 0.75314379,
      "learning_rate": 2.2308389560098253e-06,
      "loss": 0.77507228,
      "num_input_tokens_seen": 85745035,
      "step": 3986,
      "time_per_iteration": 2.6348764896392822
    },
    {
      "auxiliary_loss_clip": 0.01148059,
      "auxiliary_loss_mlp": 0.01027598,
      "balance_loss_clip": 1.05017591,
      "balance_loss_mlp": 1.01953959,
      "epoch": 0.47940840497805565,
      "flos": 17420877423360.0,
      "grad_norm": 2.750734560541953,
      "language_loss": 0.7718268,
      "learning_rate": 2.2300651728654008e-06,
      "loss": 0.79358339,
      "num_input_tokens_seen": 85760295,
      "step": 3987,
      "time_per_iteration": 2.7001709938049316
    },
    {
      "auxiliary_loss_clip": 0.01068292,
      "auxiliary_loss_mlp": 0.00754718,
      "balance_loss_clip": 1.01676297,
      "balance_loss_mlp": 1.0004065,
      "epoch": 0.47952864786869476,
      "flos": 65358175708800.0,
      "grad_norm": 0.7268700870409951,
      "language_loss": 0.60194296,
      "learning_rate": 2.229291354820272e-06,
      "loss": 0.6201731,
      "num_input_tokens_seen": 85821305,
      "step": 3988,
      "time_per_iteration": 3.2397308349609375
    },
    {
      "auxiliary_loss_clip": 0.01170038,
      "auxiliary_loss_mlp": 0.01029497,
      "balance_loss_clip": 1.05021763,
      "balance_loss_mlp": 1.02085423,
      "epoch": 0.47964889075933387,
      "flos": 16799802336000.0,
      "grad_norm": 2.1381829544750994,
      "language_loss": 0.76049715,
      "learning_rate": 2.228517501991828e-06,
      "loss": 0.78249252,
      "num_input_tokens_seen": 85840105,
      "step": 3989,
      "time_per_iteration": 3.608168840408325
    },
    {
      "auxiliary_loss_clip": 0.01058869,
      "auxiliary_loss_mlp": 0.01002994,
      "balance_loss_clip": 1.01711607,
      "balance_loss_mlp": 1.00132549,
      "epoch": 0.4797691336499729,
      "flos": 70079244808320.0,
      "grad_norm": 0.8109597628611122,
      "language_loss": 0.61052966,
      "learning_rate": 2.22774361449746e-06,
      "loss": 0.63114834,
      "num_input_tokens_seen": 85896585,
      "step": 3990,
      "time_per_iteration": 4.1977458000183105
    },
    {
      "auxiliary_loss_clip": 0.01111621,
      "auxiliary_loss_mlp": 0.01026588,
      "balance_loss_clip": 1.04346871,
      "balance_loss_mlp": 1.01833868,
      "epoch": 0.47988937654061203,
      "flos": 18953329317120.0,
      "grad_norm": 5.23016769848664,
      "language_loss": 0.70943284,
      "learning_rate": 2.2269696924545668e-06,
      "loss": 0.73081493,
      "num_input_tokens_seen": 85914415,
      "step": 3991,
      "time_per_iteration": 2.739497661590576
    },
    {
      "auxiliary_loss_clip": 0.01142637,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.0499649,
      "balance_loss_mlp": 1.018821,
      "epoch": 0.48000961943125114,
      "flos": 14461981649280.0,
      "grad_norm": 2.14403107632961,
      "language_loss": 0.78333652,
      "learning_rate": 2.2261957359805523e-06,
      "loss": 0.80502772,
      "num_input_tokens_seen": 85931650,
      "step": 3992,
      "time_per_iteration": 3.634110927581787
    },
    {
      "auxiliary_loss_clip": 0.01184436,
      "auxiliary_loss_mlp": 0.01029684,
      "balance_loss_clip": 1.05243218,
      "balance_loss_mlp": 1.02126753,
      "epoch": 0.4801298623218902,
      "flos": 27051149105280.0,
      "grad_norm": 1.973302859308407,
      "language_loss": 0.73703575,
      "learning_rate": 2.225421745192823e-06,
      "loss": 0.75917697,
      "num_input_tokens_seen": 85951805,
      "step": 3993,
      "time_per_iteration": 2.7539656162261963
    },
    {
      "auxiliary_loss_clip": 0.01168167,
      "auxiliary_loss_mlp": 0.01025787,
      "balance_loss_clip": 1.05122495,
      "balance_loss_mlp": 1.01758504,
      "epoch": 0.4802501052125293,
      "flos": 26355236031360.0,
      "grad_norm": 2.1761348384607913,
      "language_loss": 0.78555661,
      "learning_rate": 2.2246477202087955e-06,
      "loss": 0.80749619,
      "num_input_tokens_seen": 85972485,
      "step": 3994,
      "time_per_iteration": 3.605707883834839
    },
    {
      "auxiliary_loss_clip": 0.01153964,
      "auxiliary_loss_mlp": 0.0102613,
      "balance_loss_clip": 1.04631472,
      "balance_loss_mlp": 1.01794589,
      "epoch": 0.4803703481031684,
      "flos": 20993916960000.0,
      "grad_norm": 1.61016377580868,
      "language_loss": 0.83101672,
      "learning_rate": 2.223873661145887e-06,
      "loss": 0.85281765,
      "num_input_tokens_seen": 85992540,
      "step": 3995,
      "time_per_iteration": 2.804819107055664
    },
    {
      "auxiliary_loss_clip": 0.01154935,
      "auxiliary_loss_mlp": 0.00762709,
      "balance_loss_clip": 1.05184591,
      "balance_loss_mlp": 1.00023627,
      "epoch": 0.4804905909938075,
      "flos": 20703722981760.0,
      "grad_norm": 1.6291853471689601,
      "language_loss": 0.71206868,
      "learning_rate": 2.2230995681215226e-06,
      "loss": 0.7312451,
      "num_input_tokens_seen": 86012065,
      "step": 3996,
      "time_per_iteration": 2.6791675090789795
    },
    {
      "auxiliary_loss_clip": 0.01139133,
      "auxiliary_loss_mlp": 0.01027505,
      "balance_loss_clip": 1.04786348,
      "balance_loss_mlp": 1.01867783,
      "epoch": 0.4806108338844466,
      "flos": 16654831044480.0,
      "grad_norm": 2.390346104680326,
      "language_loss": 0.78125226,
      "learning_rate": 2.2223254412531305e-06,
      "loss": 0.80291861,
      "num_input_tokens_seen": 86029435,
      "step": 3997,
      "time_per_iteration": 2.6908318996429443
    },
    {
      "auxiliary_loss_clip": 0.01140905,
      "auxiliary_loss_mlp": 0.01023098,
      "balance_loss_clip": 1.04474521,
      "balance_loss_mlp": 1.01552272,
      "epoch": 0.4807310767750857,
      "flos": 20011329440640.0,
      "grad_norm": 1.8011850013812505,
      "language_loss": 0.82283044,
      "learning_rate": 2.221551280658146e-06,
      "loss": 0.84447044,
      "num_input_tokens_seen": 86048495,
      "step": 3998,
      "time_per_iteration": 2.680685520172119
    },
    {
      "auxiliary_loss_clip": 0.01122723,
      "auxiliary_loss_mlp": 0.01028304,
      "balance_loss_clip": 1.04557693,
      "balance_loss_mlp": 1.01973891,
      "epoch": 0.48085131966572475,
      "flos": 23185257984000.0,
      "grad_norm": 1.6444677118650723,
      "language_loss": 0.7424823,
      "learning_rate": 2.2207770864540085e-06,
      "loss": 0.76399267,
      "num_input_tokens_seen": 86067470,
      "step": 3999,
      "time_per_iteration": 2.7773759365081787
    },
    {
      "auxiliary_loss_clip": 0.01149662,
      "auxiliary_loss_mlp": 0.01031644,
      "balance_loss_clip": 1.04711819,
      "balance_loss_mlp": 1.02319252,
      "epoch": 0.48097156255636386,
      "flos": 20558643949440.0,
      "grad_norm": 1.7493508689121668,
      "language_loss": 0.73431826,
      "learning_rate": 2.220002858758162e-06,
      "loss": 0.75613129,
      "num_input_tokens_seen": 86085460,
      "step": 4000,
      "time_per_iteration": 2.7083816528320312
    },
    {
      "auxiliary_loss_clip": 0.01073497,
      "auxiliary_loss_mlp": 0.01003648,
      "balance_loss_clip": 1.01593125,
      "balance_loss_mlp": 1.00190783,
      "epoch": 0.481091805447003,
      "flos": 70511608817280.0,
      "grad_norm": 0.8872232156697745,
      "language_loss": 0.60895783,
      "learning_rate": 2.2192285976880573e-06,
      "loss": 0.62972927,
      "num_input_tokens_seen": 86149715,
      "step": 4001,
      "time_per_iteration": 3.266961097717285
    },
    {
      "auxiliary_loss_clip": 0.0114032,
      "auxiliary_loss_mlp": 0.00762086,
      "balance_loss_clip": 1.04431581,
      "balance_loss_mlp": 1.00026989,
      "epoch": 0.48121204833764203,
      "flos": 36428214839040.0,
      "grad_norm": 1.5587059273603099,
      "language_loss": 0.80702293,
      "learning_rate": 2.2184543033611485e-06,
      "loss": 0.826047,
      "num_input_tokens_seen": 86170795,
      "step": 4002,
      "time_per_iteration": 2.802813768386841
    },
    {
      "auxiliary_loss_clip": 0.01169445,
      "auxiliary_loss_mlp": 0.01022261,
      "balance_loss_clip": 1.04850149,
      "balance_loss_mlp": 1.01427388,
      "epoch": 0.48133229122828114,
      "flos": 27490264871040.0,
      "grad_norm": 2.5294114712240767,
      "language_loss": 0.81753689,
      "learning_rate": 2.2176799758948957e-06,
      "loss": 0.83945394,
      "num_input_tokens_seen": 86190955,
      "step": 4003,
      "time_per_iteration": 2.6529674530029297
    },
    {
      "auxiliary_loss_clip": 0.01151055,
      "auxiliary_loss_mlp": 0.01031331,
      "balance_loss_clip": 1.04929805,
      "balance_loss_mlp": 1.02317131,
      "epoch": 0.4814525341189202,
      "flos": 43072802179200.0,
      "grad_norm": 1.7445863543665725,
      "language_loss": 0.73163056,
      "learning_rate": 2.2169056154067635e-06,
      "loss": 0.75345445,
      "num_input_tokens_seen": 86214875,
      "step": 4004,
      "time_per_iteration": 2.8670263290405273
    },
    {
      "auxiliary_loss_clip": 0.01170415,
      "auxiliary_loss_mlp": 0.00762708,
      "balance_loss_clip": 1.051373,
      "balance_loss_mlp": 1.0002569,
      "epoch": 0.4815727770095593,
      "flos": 24236901400320.0,
      "grad_norm": 2.123498959401769,
      "language_loss": 0.82638252,
      "learning_rate": 2.216131222014222e-06,
      "loss": 0.84571379,
      "num_input_tokens_seen": 86232950,
      "step": 4005,
      "time_per_iteration": 2.5868144035339355
    },
    {
      "auxiliary_loss_clip": 0.01133225,
      "auxiliary_loss_mlp": 0.01026684,
      "balance_loss_clip": 1.04617095,
      "balance_loss_mlp": 1.01844072,
      "epoch": 0.4816930199001984,
      "flos": 18113630100480.0,
      "grad_norm": 1.8986730814434014,
      "language_loss": 0.80653071,
      "learning_rate": 2.2153567958347455e-06,
      "loss": 0.82812983,
      "num_input_tokens_seen": 86249160,
      "step": 4006,
      "time_per_iteration": 2.5640811920166016
    },
    {
      "auxiliary_loss_clip": 0.01155019,
      "auxiliary_loss_mlp": 0.01028073,
      "balance_loss_clip": 1.05075264,
      "balance_loss_mlp": 1.02035451,
      "epoch": 0.48181326279083747,
      "flos": 17274720983040.0,
      "grad_norm": 2.6420638123502944,
      "language_loss": 0.79467738,
      "learning_rate": 2.214582336985815e-06,
      "loss": 0.81650829,
      "num_input_tokens_seen": 86267060,
      "step": 4007,
      "time_per_iteration": 2.6288905143737793
    },
    {
      "auxiliary_loss_clip": 0.01144023,
      "auxiliary_loss_mlp": 0.01027914,
      "balance_loss_clip": 1.04482698,
      "balance_loss_mlp": 1.01980209,
      "epoch": 0.4819335056814766,
      "flos": 14903252231040.0,
      "grad_norm": 2.1000046605165297,
      "language_loss": 0.66474211,
      "learning_rate": 2.2138078455849142e-06,
      "loss": 0.68646145,
      "num_input_tokens_seen": 86285055,
      "step": 4008,
      "time_per_iteration": 2.5176854133605957
    },
    {
      "auxiliary_loss_clip": 0.01175884,
      "auxiliary_loss_mlp": 0.01030294,
      "balance_loss_clip": 1.05224943,
      "balance_loss_mlp": 1.0228374,
      "epoch": 0.4820537485721157,
      "flos": 19244888012160.0,
      "grad_norm": 1.880428823616472,
      "language_loss": 0.78999841,
      "learning_rate": 2.2130333217495334e-06,
      "loss": 0.81206024,
      "num_input_tokens_seen": 86304225,
      "step": 4009,
      "time_per_iteration": 2.668553590774536
    },
    {
      "auxiliary_loss_clip": 0.01151427,
      "auxiliary_loss_mlp": 0.01026201,
      "balance_loss_clip": 1.04801369,
      "balance_loss_mlp": 1.0180943,
      "epoch": 0.48217399146275475,
      "flos": 16033791870720.0,
      "grad_norm": 2.644887090549291,
      "language_loss": 0.68096864,
      "learning_rate": 2.2122587655971665e-06,
      "loss": 0.7027449,
      "num_input_tokens_seen": 86319170,
      "step": 4010,
      "time_per_iteration": 2.6296064853668213
    },
    {
      "auxiliary_loss_clip": 0.01157034,
      "auxiliary_loss_mlp": 0.01028644,
      "balance_loss_clip": 1.04918635,
      "balance_loss_mlp": 1.02071691,
      "epoch": 0.48229423435339386,
      "flos": 24134197438080.0,
      "grad_norm": 1.816593394125063,
      "language_loss": 0.64175367,
      "learning_rate": 2.211484177245314e-06,
      "loss": 0.66361046,
      "num_input_tokens_seen": 86338760,
      "step": 4011,
      "time_per_iteration": 2.6761059761047363
    },
    {
      "auxiliary_loss_clip": 0.01184775,
      "auxiliary_loss_mlp": 0.01025944,
      "balance_loss_clip": 1.0511173,
      "balance_loss_mlp": 1.01739073,
      "epoch": 0.48241447724403297,
      "flos": 23805435231360.0,
      "grad_norm": 2.0857295090690062,
      "language_loss": 0.72368407,
      "learning_rate": 2.21070955681148e-06,
      "loss": 0.74579126,
      "num_input_tokens_seen": 86357865,
      "step": 4012,
      "time_per_iteration": 2.5849668979644775
    },
    {
      "auxiliary_loss_clip": 0.01132483,
      "auxiliary_loss_mlp": 0.01031909,
      "balance_loss_clip": 1.04665709,
      "balance_loss_mlp": 1.02424335,
      "epoch": 0.482534720134672,
      "flos": 23110312256640.0,
      "grad_norm": 1.5864890525494755,
      "language_loss": 0.78041464,
      "learning_rate": 2.209934904413174e-06,
      "loss": 0.80205858,
      "num_input_tokens_seen": 86379470,
      "step": 4013,
      "time_per_iteration": 2.688859224319458
    },
    {
      "auxiliary_loss_clip": 0.01109082,
      "auxiliary_loss_mlp": 0.01028358,
      "balance_loss_clip": 1.03779888,
      "balance_loss_mlp": 1.02011514,
      "epoch": 0.48265496302531113,
      "flos": 20923819568640.0,
      "grad_norm": 2.148816583160264,
      "language_loss": 0.71608949,
      "learning_rate": 2.2091602201679095e-06,
      "loss": 0.73746389,
      "num_input_tokens_seen": 86399080,
      "step": 4014,
      "time_per_iteration": 2.77426815032959
    },
    {
      "auxiliary_loss_clip": 0.01145963,
      "auxiliary_loss_mlp": 0.01028677,
      "balance_loss_clip": 1.04844892,
      "balance_loss_mlp": 1.02033818,
      "epoch": 0.48277520591595025,
      "flos": 15231152511360.0,
      "grad_norm": 2.2333877824979167,
      "language_loss": 0.83488876,
      "learning_rate": 2.208385504193206e-06,
      "loss": 0.85663509,
      "num_input_tokens_seen": 86416580,
      "step": 4015,
      "time_per_iteration": 3.628854274749756
    },
    {
      "auxiliary_loss_clip": 0.01183848,
      "auxiliary_loss_mlp": 0.01023553,
      "balance_loss_clip": 1.04995871,
      "balance_loss_mlp": 1.01628768,
      "epoch": 0.4828954488065893,
      "flos": 17858664385920.0,
      "grad_norm": 2.583067409931057,
      "language_loss": 0.81409872,
      "learning_rate": 2.2076107566065873e-06,
      "loss": 0.8361727,
      "num_input_tokens_seen": 86434365,
      "step": 4016,
      "time_per_iteration": 3.478276252746582
    },
    {
      "auxiliary_loss_clip": 0.01173135,
      "auxiliary_loss_mlp": 0.01030578,
      "balance_loss_clip": 1.0499928,
      "balance_loss_mlp": 1.02276385,
      "epoch": 0.4830156916972284,
      "flos": 32087405070720.0,
      "grad_norm": 2.216864701325059,
      "language_loss": 0.75245148,
      "learning_rate": 2.2068359775255816e-06,
      "loss": 0.77448863,
      "num_input_tokens_seen": 86452675,
      "step": 4017,
      "time_per_iteration": 2.682450294494629
    },
    {
      "auxiliary_loss_clip": 0.01123514,
      "auxiliary_loss_mlp": 0.01025812,
      "balance_loss_clip": 1.04416704,
      "balance_loss_mlp": 1.01799202,
      "epoch": 0.48313593458786747,
      "flos": 21871717528320.0,
      "grad_norm": 4.301178459590197,
      "language_loss": 0.78650141,
      "learning_rate": 2.206061167067723e-06,
      "loss": 0.8079946,
      "num_input_tokens_seen": 86470785,
      "step": 4018,
      "time_per_iteration": 2.784757137298584
    },
    {
      "auxiliary_loss_clip": 0.01135636,
      "auxiliary_loss_mlp": 0.01030051,
      "balance_loss_clip": 1.04248333,
      "balance_loss_mlp": 1.02162921,
      "epoch": 0.4832561774785066,
      "flos": 22601206840320.0,
      "grad_norm": 2.854682435732161,
      "language_loss": 0.80168176,
      "learning_rate": 2.205286325350549e-06,
      "loss": 0.82333869,
      "num_input_tokens_seen": 86489850,
      "step": 4019,
      "time_per_iteration": 3.64235258102417
    },
    {
      "auxiliary_loss_clip": 0.0112618,
      "auxiliary_loss_mlp": 0.01030847,
      "balance_loss_clip": 1.04608572,
      "balance_loss_mlp": 1.02312207,
      "epoch": 0.4833764203691457,
      "flos": 13437342282240.0,
      "grad_norm": 2.087837826919939,
      "language_loss": 0.72712815,
      "learning_rate": 2.204511452491603e-06,
      "loss": 0.74869847,
      "num_input_tokens_seen": 86506475,
      "step": 4020,
      "time_per_iteration": 3.5980746746063232
    },
    {
      "auxiliary_loss_clip": 0.01181829,
      "auxiliary_loss_mlp": 0.01027554,
      "balance_loss_clip": 1.05286002,
      "balance_loss_mlp": 1.01978731,
      "epoch": 0.48349666325978474,
      "flos": 44128036955520.0,
      "grad_norm": 1.7214661952583437,
      "language_loss": 0.74762595,
      "learning_rate": 2.2037365486084316e-06,
      "loss": 0.76971984,
      "num_input_tokens_seen": 86529715,
      "step": 4021,
      "time_per_iteration": 2.8566391468048096
    },
    {
      "auxiliary_loss_clip": 0.01149325,
      "auxiliary_loss_mlp": 0.01027313,
      "balance_loss_clip": 1.04504895,
      "balance_loss_mlp": 1.01912344,
      "epoch": 0.48361690615042385,
      "flos": 26028377245440.0,
      "grad_norm": 2.5046338339307934,
      "language_loss": 0.78371829,
      "learning_rate": 2.2029616138185886e-06,
      "loss": 0.80548465,
      "num_input_tokens_seen": 86548715,
      "step": 4022,
      "time_per_iteration": 2.7638494968414307
    },
    {
      "auxiliary_loss_clip": 0.01138271,
      "auxiliary_loss_mlp": 0.01024333,
      "balance_loss_clip": 1.04861593,
      "balance_loss_mlp": 1.01652503,
      "epoch": 0.48373714904106296,
      "flos": 22273306560000.0,
      "grad_norm": 1.8705726760247465,
      "language_loss": 0.83069181,
      "learning_rate": 2.202186648239629e-06,
      "loss": 0.85231793,
      "num_input_tokens_seen": 86568650,
      "step": 4023,
      "time_per_iteration": 2.734017848968506
    },
    {
      "auxiliary_loss_clip": 0.01166767,
      "auxiliary_loss_mlp": 0.01023059,
      "balance_loss_clip": 1.04962552,
      "balance_loss_mlp": 1.01555502,
      "epoch": 0.483857391931702,
      "flos": 28292293699200.0,
      "grad_norm": 3.5257314723457016,
      "language_loss": 0.71644694,
      "learning_rate": 2.201411651989117e-06,
      "loss": 0.73834521,
      "num_input_tokens_seen": 86590630,
      "step": 4024,
      "time_per_iteration": 2.6969945430755615
    },
    {
      "auxiliary_loss_clip": 0.01153938,
      "auxiliary_loss_mlp": 0.00762983,
      "balance_loss_clip": 1.04872775,
      "balance_loss_mlp": 1.00028372,
      "epoch": 0.48397763482234113,
      "flos": 27418048577280.0,
      "grad_norm": 1.7917837161341128,
      "language_loss": 0.78060699,
      "learning_rate": 2.2006366251846167e-06,
      "loss": 0.7997762,
      "num_input_tokens_seen": 86611270,
      "step": 4025,
      "time_per_iteration": 2.7299904823303223
    },
    {
      "auxiliary_loss_clip": 0.01156554,
      "auxiliary_loss_mlp": 0.01033388,
      "balance_loss_clip": 1.05237484,
      "balance_loss_mlp": 1.02566338,
      "epoch": 0.48409787771298024,
      "flos": 16797252470400.0,
      "grad_norm": 1.8080034993995413,
      "language_loss": 0.75508255,
      "learning_rate": 2.1998615679436997e-06,
      "loss": 0.77698195,
      "num_input_tokens_seen": 86628810,
      "step": 4026,
      "time_per_iteration": 2.658656358718872
    },
    {
      "auxiliary_loss_clip": 0.01161764,
      "auxiliary_loss_mlp": 0.01026232,
      "balance_loss_clip": 1.04856467,
      "balance_loss_mlp": 1.01775575,
      "epoch": 0.4842181206036193,
      "flos": 25083496028160.0,
      "grad_norm": 2.537386682040976,
      "language_loss": 0.77197433,
      "learning_rate": 2.199086480383942e-06,
      "loss": 0.7938543,
      "num_input_tokens_seen": 86648185,
      "step": 4027,
      "time_per_iteration": 2.7470242977142334
    },
    {
      "auxiliary_loss_clip": 0.01171786,
      "auxiliary_loss_mlp": 0.01030089,
      "balance_loss_clip": 1.05187488,
      "balance_loss_mlp": 1.02081442,
      "epoch": 0.4843383634942584,
      "flos": 30372311496960.0,
      "grad_norm": 2.6421168180202974,
      "language_loss": 0.68155146,
      "learning_rate": 2.1983113626229234e-06,
      "loss": 0.70357025,
      "num_input_tokens_seen": 86667435,
      "step": 4028,
      "time_per_iteration": 2.7181034088134766
    },
    {
      "auxiliary_loss_clip": 0.01131626,
      "auxiliary_loss_mlp": 0.00763559,
      "balance_loss_clip": 1.04280436,
      "balance_loss_mlp": 1.00027466,
      "epoch": 0.4844586063848975,
      "flos": 20413564917120.0,
      "grad_norm": 1.6596570122240712,
      "language_loss": 0.78284335,
      "learning_rate": 2.1975362147782293e-06,
      "loss": 0.80179524,
      "num_input_tokens_seen": 86686630,
      "step": 4029,
      "time_per_iteration": 2.739306688308716
    },
    {
      "auxiliary_loss_clip": 0.01074319,
      "auxiliary_loss_mlp": 0.01004561,
      "balance_loss_clip": 1.02954292,
      "balance_loss_mlp": 1.00290382,
      "epoch": 0.48457884927553657,
      "flos": 70303722854400.0,
      "grad_norm": 0.6930981595139819,
      "language_loss": 0.54107738,
      "learning_rate": 2.196761036967448e-06,
      "loss": 0.56186616,
      "num_input_tokens_seen": 86754595,
      "step": 4030,
      "time_per_iteration": 3.435328483581543
    },
    {
      "auxiliary_loss_clip": 0.01163024,
      "auxiliary_loss_mlp": 0.01030579,
      "balance_loss_clip": 1.04795265,
      "balance_loss_mlp": 1.02317941,
      "epoch": 0.4846990921661757,
      "flos": 19934516206080.0,
      "grad_norm": 1.6577480761146417,
      "language_loss": 0.77309442,
      "learning_rate": 2.1959858293081743e-06,
      "loss": 0.79503036,
      "num_input_tokens_seen": 86773730,
      "step": 4031,
      "time_per_iteration": 2.5975069999694824
    },
    {
      "auxiliary_loss_clip": 0.01138498,
      "auxiliary_loss_mlp": 0.0103273,
      "balance_loss_clip": 1.04771006,
      "balance_loss_mlp": 1.02427173,
      "epoch": 0.4848193350568148,
      "flos": 23075945919360.0,
      "grad_norm": 1.5805565529417993,
      "language_loss": 0.75720906,
      "learning_rate": 2.1952105919180056e-06,
      "loss": 0.77892137,
      "num_input_tokens_seen": 86792985,
      "step": 4032,
      "time_per_iteration": 2.713076114654541
    },
    {
      "auxiliary_loss_clip": 0.01154359,
      "auxiliary_loss_mlp": 0.01027989,
      "balance_loss_clip": 1.05120957,
      "balance_loss_mlp": 1.01951325,
      "epoch": 0.48493957794745385,
      "flos": 22455481363200.0,
      "grad_norm": 2.930879580591001,
      "language_loss": 0.68032241,
      "learning_rate": 2.1944353249145456e-06,
      "loss": 0.70214593,
      "num_input_tokens_seen": 86812095,
      "step": 4033,
      "time_per_iteration": 2.7157299518585205
    },
    {
      "auxiliary_loss_clip": 0.01184976,
      "auxiliary_loss_mlp": 0.0102747,
      "balance_loss_clip": 1.05496073,
      "balance_loss_mlp": 1.01978731,
      "epoch": 0.48505982083809296,
      "flos": 25046112948480.0,
      "grad_norm": 1.6198906433400748,
      "language_loss": 0.74641109,
      "learning_rate": 2.193660028415401e-06,
      "loss": 0.76853549,
      "num_input_tokens_seen": 86832875,
      "step": 4034,
      "time_per_iteration": 2.64768385887146
    },
    {
      "auxiliary_loss_clip": 0.011466,
      "auxiliary_loss_mlp": 0.01024941,
      "balance_loss_clip": 1.04765022,
      "balance_loss_mlp": 1.01713252,
      "epoch": 0.485180063728732,
      "flos": 26761386090240.0,
      "grad_norm": 2.4037668526284928,
      "language_loss": 0.82192504,
      "learning_rate": 2.1928847025381852e-06,
      "loss": 0.84364045,
      "num_input_tokens_seen": 86853480,
      "step": 4035,
      "time_per_iteration": 2.7970049381256104
    },
    {
      "auxiliary_loss_clip": 0.01166532,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 1.04614067,
      "balance_loss_mlp": 1.02360404,
      "epoch": 0.4853003066193711,
      "flos": 24059143969920.0,
      "grad_norm": 1.7428120733805368,
      "language_loss": 0.83818257,
      "learning_rate": 2.192109347400512e-06,
      "loss": 0.86016881,
      "num_input_tokens_seen": 86873695,
      "step": 4036,
      "time_per_iteration": 2.659339666366577
    },
    {
      "auxiliary_loss_clip": 0.01157156,
      "auxiliary_loss_mlp": 0.01031771,
      "balance_loss_clip": 1.04878485,
      "balance_loss_mlp": 1.02308011,
      "epoch": 0.48542054951001024,
      "flos": 23076376882560.0,
      "grad_norm": 1.9160880132097506,
      "language_loss": 0.79171813,
      "learning_rate": 2.191333963120004e-06,
      "loss": 0.81360739,
      "num_input_tokens_seen": 86892675,
      "step": 4037,
      "time_per_iteration": 2.692756414413452
    },
    {
      "auxiliary_loss_clip": 0.01151619,
      "auxiliary_loss_mlp": 0.01024837,
      "balance_loss_clip": 1.04755783,
      "balance_loss_mlp": 1.01716614,
      "epoch": 0.4855407924006493,
      "flos": 25664889565440.0,
      "grad_norm": 3.853883549504259,
      "language_loss": 0.70284164,
      "learning_rate": 2.190558549814286e-06,
      "loss": 0.72460616,
      "num_input_tokens_seen": 86912835,
      "step": 4038,
      "time_per_iteration": 2.6974172592163086
    },
    {
      "auxiliary_loss_clip": 0.01151641,
      "auxiliary_loss_mlp": 0.01030727,
      "balance_loss_clip": 1.04630065,
      "balance_loss_mlp": 1.02282906,
      "epoch": 0.4856610352912884,
      "flos": 23987933256960.0,
      "grad_norm": 1.7107355179224146,
      "language_loss": 0.79485536,
      "learning_rate": 2.1897831076009872e-06,
      "loss": 0.816679,
      "num_input_tokens_seen": 86932475,
      "step": 4039,
      "time_per_iteration": 2.6743619441986084
    },
    {
      "auxiliary_loss_clip": 0.01171385,
      "auxiliary_loss_mlp": 0.01024287,
      "balance_loss_clip": 1.05165291,
      "balance_loss_mlp": 1.01706564,
      "epoch": 0.4857812781819275,
      "flos": 24096814358400.0,
      "grad_norm": 2.172559458236493,
      "language_loss": 0.79621327,
      "learning_rate": 2.1890076365977426e-06,
      "loss": 0.81816995,
      "num_input_tokens_seen": 86952300,
      "step": 4040,
      "time_per_iteration": 2.7648849487304688
    },
    {
      "auxiliary_loss_clip": 0.0105524,
      "auxiliary_loss_mlp": 0.01000906,
      "balance_loss_clip": 1.01220798,
      "balance_loss_mlp": 0.99926108,
      "epoch": 0.48590152107256657,
      "flos": 56266635185280.0,
      "grad_norm": 0.8508634600409317,
      "language_loss": 0.52804589,
      "learning_rate": 2.188232136922189e-06,
      "loss": 0.54860735,
      "num_input_tokens_seen": 87010420,
      "step": 4041,
      "time_per_iteration": 4.139225006103516
    },
    {
      "auxiliary_loss_clip": 0.01104808,
      "auxiliary_loss_mlp": 0.0102651,
      "balance_loss_clip": 1.04241312,
      "balance_loss_mlp": 1.01818943,
      "epoch": 0.4860217639632057,
      "flos": 20046988667520.0,
      "grad_norm": 2.8871550260839935,
      "language_loss": 0.75552833,
      "learning_rate": 2.187456608691971e-06,
      "loss": 0.77684152,
      "num_input_tokens_seen": 87029295,
      "step": 4042,
      "time_per_iteration": 3.672436237335205
    },
    {
      "auxiliary_loss_clip": 0.0114352,
      "auxiliary_loss_mlp": 0.01030332,
      "balance_loss_clip": 1.04946256,
      "balance_loss_mlp": 1.02271485,
      "epoch": 0.4861420068538448,
      "flos": 17822143232640.0,
      "grad_norm": 2.15221582687372,
      "language_loss": 0.87751949,
      "learning_rate": 2.1866810520247334e-06,
      "loss": 0.89925802,
      "num_input_tokens_seen": 87048165,
      "step": 4043,
      "time_per_iteration": 2.6838765144348145
    },
    {
      "auxiliary_loss_clip": 0.01171221,
      "auxiliary_loss_mlp": 0.01024632,
      "balance_loss_clip": 1.04741716,
      "balance_loss_mlp": 1.01635909,
      "epoch": 0.48626224974448384,
      "flos": 26250125857920.0,
      "grad_norm": 1.8724644620978044,
      "language_loss": 0.65077388,
      "learning_rate": 2.185905467038129e-06,
      "loss": 0.67273241,
      "num_input_tokens_seen": 87067070,
      "step": 4044,
      "time_per_iteration": 3.5988035202026367
    },
    {
      "auxiliary_loss_clip": 0.01181624,
      "auxiliary_loss_mlp": 0.01030943,
      "balance_loss_clip": 1.05307865,
      "balance_loss_mlp": 1.02295041,
      "epoch": 0.48638249263512295,
      "flos": 22054502862720.0,
      "grad_norm": 1.7471524914703611,
      "language_loss": 0.77643627,
      "learning_rate": 2.1851298538498127e-06,
      "loss": 0.79856193,
      "num_input_tokens_seen": 87086785,
      "step": 4045,
      "time_per_iteration": 3.494997262954712
    },
    {
      "auxiliary_loss_clip": 0.01178681,
      "auxiliary_loss_mlp": 0.00763714,
      "balance_loss_clip": 1.05436838,
      "balance_loss_mlp": 1.00034046,
      "epoch": 0.48650273552576206,
      "flos": 25119945354240.0,
      "grad_norm": 2.132679526766794,
      "language_loss": 0.80038488,
      "learning_rate": 2.184354212577446e-06,
      "loss": 0.81980884,
      "num_input_tokens_seen": 87107090,
      "step": 4046,
      "time_per_iteration": 2.6700541973114014
    },
    {
      "auxiliary_loss_clip": 0.01186008,
      "auxiliary_loss_mlp": 0.0103876,
      "balance_loss_clip": 1.05081666,
      "balance_loss_mlp": 1.02995634,
      "epoch": 0.4866229784164011,
      "flos": 17456931699840.0,
      "grad_norm": 3.0611790694552843,
      "language_loss": 0.62983227,
      "learning_rate": 2.1835785433386907e-06,
      "loss": 0.65207994,
      "num_input_tokens_seen": 87125905,
      "step": 4047,
      "time_per_iteration": 2.5410852432250977
    },
    {
      "auxiliary_loss_clip": 0.01131198,
      "auxiliary_loss_mlp": 0.01028811,
      "balance_loss_clip": 1.04712653,
      "balance_loss_mlp": 1.02106214,
      "epoch": 0.48674322130704023,
      "flos": 23331127115520.0,
      "grad_norm": 1.8452014127932634,
      "language_loss": 0.65748721,
      "learning_rate": 2.182802846251216e-06,
      "loss": 0.67908728,
      "num_input_tokens_seen": 87146175,
      "step": 4048,
      "time_per_iteration": 2.7243688106536865
    },
    {
      "auxiliary_loss_clip": 0.01142269,
      "auxiliary_loss_mlp": 0.01032323,
      "balance_loss_clip": 1.04423761,
      "balance_loss_mlp": 1.02371013,
      "epoch": 0.4868634641976793,
      "flos": 28804344030720.0,
      "grad_norm": 2.10343445917185,
      "language_loss": 0.72273695,
      "learning_rate": 2.182027121432696e-06,
      "loss": 0.74448287,
      "num_input_tokens_seen": 87166800,
      "step": 4049,
      "time_per_iteration": 2.7725679874420166
    },
    {
      "auxiliary_loss_clip": 0.01186364,
      "auxiliary_loss_mlp": 0.01032213,
      "balance_loss_clip": 1.05266011,
      "balance_loss_mlp": 1.0240171,
      "epoch": 0.4869837070883184,
      "flos": 19025976574080.0,
      "grad_norm": 2.387930326486471,
      "language_loss": 0.8237133,
      "learning_rate": 2.1812513690008054e-06,
      "loss": 0.84589899,
      "num_input_tokens_seen": 87185920,
      "step": 4050,
      "time_per_iteration": 2.645840883255005
    },
    {
      "auxiliary_loss_clip": 0.01175041,
      "auxiliary_loss_mlp": 0.01038297,
      "balance_loss_clip": 1.05133915,
      "balance_loss_mlp": 1.03019714,
      "epoch": 0.4871039499789575,
      "flos": 15121409483520.0,
      "grad_norm": 2.238659325994885,
      "language_loss": 0.80004877,
      "learning_rate": 2.180475589073227e-06,
      "loss": 0.82218218,
      "num_input_tokens_seen": 87203620,
      "step": 4051,
      "time_per_iteration": 2.67335844039917
    },
    {
      "auxiliary_loss_clip": 0.01158423,
      "auxiliary_loss_mlp": 0.01023688,
      "balance_loss_clip": 1.04617977,
      "balance_loss_mlp": 1.01591253,
      "epoch": 0.48722419286959656,
      "flos": 26174066808960.0,
      "grad_norm": 1.6307540998970063,
      "language_loss": 0.73240912,
      "learning_rate": 2.1796997817676456e-06,
      "loss": 0.75423026,
      "num_input_tokens_seen": 87224630,
      "step": 4052,
      "time_per_iteration": 2.661839246749878
    },
    {
      "auxiliary_loss_clip": 0.01169181,
      "auxiliary_loss_mlp": 0.00762449,
      "balance_loss_clip": 1.0495286,
      "balance_loss_mlp": 1.00037515,
      "epoch": 0.4873444357602357,
      "flos": 24026142349440.0,
      "grad_norm": 1.6739859958805676,
      "language_loss": 0.67673653,
      "learning_rate": 2.1789239472017494e-06,
      "loss": 0.69605279,
      "num_input_tokens_seen": 87246280,
      "step": 4053,
      "time_per_iteration": 2.6774613857269287
    },
    {
      "auxiliary_loss_clip": 0.01139192,
      "auxiliary_loss_mlp": 0.01031123,
      "balance_loss_clip": 1.04511666,
      "balance_loss_mlp": 1.02318394,
      "epoch": 0.4874646786508748,
      "flos": 22820441500800.0,
      "grad_norm": 2.2481442452957756,
      "language_loss": 0.73449916,
      "learning_rate": 2.1781480854932326e-06,
      "loss": 0.75620228,
      "num_input_tokens_seen": 87266045,
      "step": 4054,
      "time_per_iteration": 2.713327646255493
    },
    {
      "auxiliary_loss_clip": 0.01125139,
      "auxiliary_loss_mlp": 0.0102685,
      "balance_loss_clip": 1.04831004,
      "balance_loss_mlp": 1.01925039,
      "epoch": 0.48758492154151384,
      "flos": 21287594557440.0,
      "grad_norm": 1.870174536632044,
      "language_loss": 0.79251474,
      "learning_rate": 2.1773721967597933e-06,
      "loss": 0.81403458,
      "num_input_tokens_seen": 87284495,
      "step": 4055,
      "time_per_iteration": 2.7886545658111572
    },
    {
      "auxiliary_loss_clip": 0.01050493,
      "auxiliary_loss_mlp": 0.01004138,
      "balance_loss_clip": 1.0123229,
      "balance_loss_mlp": 1.00274372,
      "epoch": 0.48770516443215295,
      "flos": 62244109180800.0,
      "grad_norm": 2.0032302265915627,
      "language_loss": 0.57399118,
      "learning_rate": 2.1765962811191322e-06,
      "loss": 0.5945375,
      "num_input_tokens_seen": 87338960,
      "step": 4056,
      "time_per_iteration": 3.194542407989502
    },
    {
      "auxiliary_loss_clip": 0.01032762,
      "auxiliary_loss_mlp": 0.01009705,
      "balance_loss_clip": 1.01383853,
      "balance_loss_mlp": 1.00814342,
      "epoch": 0.48782540732279206,
      "flos": 66133451882880.0,
      "grad_norm": 0.8260342643057988,
      "language_loss": 0.61973536,
      "learning_rate": 2.1758203386889566e-06,
      "loss": 0.64015996,
      "num_input_tokens_seen": 87401730,
      "step": 4057,
      "time_per_iteration": 3.317780017852783
    },
    {
      "auxiliary_loss_clip": 0.01140614,
      "auxiliary_loss_mlp": 0.00763285,
      "balance_loss_clip": 1.04717565,
      "balance_loss_mlp": 1.00033641,
      "epoch": 0.4879456502134311,
      "flos": 14607922608000.0,
      "grad_norm": 2.0748690192236094,
      "language_loss": 0.84047055,
      "learning_rate": 2.1750443695869746e-06,
      "loss": 0.85950953,
      "num_input_tokens_seen": 87417300,
      "step": 4058,
      "time_per_iteration": 2.683527946472168
    },
    {
      "auxiliary_loss_clip": 0.01171917,
      "auxiliary_loss_mlp": 0.01027256,
      "balance_loss_clip": 1.04975867,
      "balance_loss_mlp": 1.01948309,
      "epoch": 0.4880658931040702,
      "flos": 19500464257920.0,
      "grad_norm": 4.2486047182720466,
      "language_loss": 0.8586669,
      "learning_rate": 2.174268373930901e-06,
      "loss": 0.88065869,
      "num_input_tokens_seen": 87434815,
      "step": 4059,
      "time_per_iteration": 2.7109627723693848
    },
    {
      "auxiliary_loss_clip": 0.01136301,
      "auxiliary_loss_mlp": 0.0076244,
      "balance_loss_clip": 1.04964387,
      "balance_loss_mlp": 1.00031102,
      "epoch": 0.48818613599470934,
      "flos": 16723060928640.0,
      "grad_norm": 2.424134868911688,
      "language_loss": 0.79826045,
      "learning_rate": 2.1734923518384537e-06,
      "loss": 0.81724787,
      "num_input_tokens_seen": 87451420,
      "step": 4060,
      "time_per_iteration": 2.6793675422668457
    },
    {
      "auxiliary_loss_clip": 0.01125834,
      "auxiliary_loss_mlp": 0.01026105,
      "balance_loss_clip": 1.04676449,
      "balance_loss_mlp": 1.01838613,
      "epoch": 0.4883063788853484,
      "flos": 26756932803840.0,
      "grad_norm": 1.9184589520619713,
      "language_loss": 0.82298797,
      "learning_rate": 2.1727163034273547e-06,
      "loss": 0.8445074,
      "num_input_tokens_seen": 87469585,
      "step": 4061,
      "time_per_iteration": 2.8239872455596924
    },
    {
      "auxiliary_loss_clip": 0.01169714,
      "auxiliary_loss_mlp": 0.01026837,
      "balance_loss_clip": 1.04814744,
      "balance_loss_mlp": 1.01856136,
      "epoch": 0.4884266217759875,
      "flos": 16763388923520.0,
      "grad_norm": 6.8719346175016245,
      "language_loss": 0.78837115,
      "learning_rate": 2.17194022881533e-06,
      "loss": 0.81033665,
      "num_input_tokens_seen": 87485675,
      "step": 4062,
      "time_per_iteration": 2.6239242553710938
    },
    {
      "auxiliary_loss_clip": 0.01158036,
      "auxiliary_loss_mlp": 0.01032839,
      "balance_loss_clip": 1.04695904,
      "balance_loss_mlp": 1.0241667,
      "epoch": 0.4885468646666266,
      "flos": 24207132003840.0,
      "grad_norm": 1.7972658523394556,
      "language_loss": 0.67880821,
      "learning_rate": 2.1711641281201092e-06,
      "loss": 0.70071697,
      "num_input_tokens_seen": 87505605,
      "step": 4063,
      "time_per_iteration": 2.725346565246582
    },
    {
      "auxiliary_loss_clip": 0.01168056,
      "auxiliary_loss_mlp": 0.01022814,
      "balance_loss_clip": 1.05154872,
      "balance_loss_mlp": 1.01516128,
      "epoch": 0.48866710755726567,
      "flos": 14610795696000.0,
      "grad_norm": 2.780857309557539,
      "language_loss": 0.79819584,
      "learning_rate": 2.1703880014594264e-06,
      "loss": 0.82010454,
      "num_input_tokens_seen": 87523195,
      "step": 4064,
      "time_per_iteration": 2.6397337913513184
    },
    {
      "auxiliary_loss_clip": 0.01122702,
      "auxiliary_loss_mlp": 0.01029489,
      "balance_loss_clip": 1.04771185,
      "balance_loss_mlp": 1.02193105,
      "epoch": 0.4887873504479048,
      "flos": 28804451771520.0,
      "grad_norm": 2.772972175880441,
      "language_loss": 0.73886847,
      "learning_rate": 2.1696118489510182e-06,
      "loss": 0.76039034,
      "num_input_tokens_seen": 87544125,
      "step": 4065,
      "time_per_iteration": 2.822624683380127
    },
    {
      "auxiliary_loss_clip": 0.01144688,
      "auxiliary_loss_mlp": 0.00762914,
      "balance_loss_clip": 1.04765344,
      "balance_loss_mlp": 1.00032198,
      "epoch": 0.48890759333854383,
      "flos": 22784387224320.0,
      "grad_norm": 1.7850552757384905,
      "language_loss": 0.72593641,
      "learning_rate": 2.1688356707126286e-06,
      "loss": 0.7450124,
      "num_input_tokens_seen": 87563745,
      "step": 4066,
      "time_per_iteration": 2.768155097961426
    },
    {
      "auxiliary_loss_clip": 0.01135444,
      "auxiliary_loss_mlp": 0.01032844,
      "balance_loss_clip": 1.04867554,
      "balance_loss_mlp": 1.02420151,
      "epoch": 0.48902783622918294,
      "flos": 17786088956160.0,
      "grad_norm": 2.0723441366417172,
      "language_loss": 0.69946212,
      "learning_rate": 2.168059466862001e-06,
      "loss": 0.72114503,
      "num_input_tokens_seen": 87581895,
      "step": 4067,
      "time_per_iteration": 3.64701771736145
    },
    {
      "auxiliary_loss_clip": 0.01156406,
      "auxiliary_loss_mlp": 0.01025681,
      "balance_loss_clip": 1.04686594,
      "balance_loss_mlp": 1.01837993,
      "epoch": 0.48914807911982205,
      "flos": 22310294590080.0,
      "grad_norm": 2.1219998122716723,
      "language_loss": 0.81901062,
      "learning_rate": 2.167283237516887e-06,
      "loss": 0.84083152,
      "num_input_tokens_seen": 87600170,
      "step": 4068,
      "time_per_iteration": 3.649001359939575
    },
    {
      "auxiliary_loss_clip": 0.01157933,
      "auxiliary_loss_mlp": 0.01030879,
      "balance_loss_clip": 1.04709888,
      "balance_loss_mlp": 1.02265406,
      "epoch": 0.4892683220104611,
      "flos": 16363020954240.0,
      "grad_norm": 2.0843105894751806,
      "language_loss": 0.74571669,
      "learning_rate": 2.1665069827950383e-06,
      "loss": 0.76760483,
      "num_input_tokens_seen": 87617455,
      "step": 4069,
      "time_per_iteration": 2.614064931869507
    },
    {
      "auxiliary_loss_clip": 0.0115021,
      "auxiliary_loss_mlp": 0.01032391,
      "balance_loss_clip": 1.04688334,
      "balance_loss_mlp": 1.02505946,
      "epoch": 0.4893885649011002,
      "flos": 15739144606080.0,
      "grad_norm": 4.776150042076431,
      "language_loss": 0.8672033,
      "learning_rate": 2.1657307028142126e-06,
      "loss": 0.88902926,
      "num_input_tokens_seen": 87634995,
      "step": 4070,
      "time_per_iteration": 3.5018348693847656
    },
    {
      "auxiliary_loss_clip": 0.01158114,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.05038381,
      "balance_loss_mlp": 1.02149808,
      "epoch": 0.48950880779173933,
      "flos": 28581984887040.0,
      "grad_norm": 2.660467727163867,
      "language_loss": 0.67480755,
      "learning_rate": 2.164954397692171e-06,
      "loss": 0.69668442,
      "num_input_tokens_seen": 87654420,
      "step": 4071,
      "time_per_iteration": 2.776988983154297
    },
    {
      "auxiliary_loss_clip": 0.01063568,
      "auxiliary_loss_mlp": 0.010024,
      "balance_loss_clip": 1.01569402,
      "balance_loss_mlp": 1.00092745,
      "epoch": 0.4896290506823784,
      "flos": 66186310746240.0,
      "grad_norm": 1.0706712495271482,
      "language_loss": 0.77326202,
      "learning_rate": 2.164178067546678e-06,
      "loss": 0.79392171,
      "num_input_tokens_seen": 87713585,
      "step": 4072,
      "time_per_iteration": 4.24094820022583
    },
    {
      "auxiliary_loss_clip": 0.01159782,
      "auxiliary_loss_mlp": 0.01027301,
      "balance_loss_clip": 1.04669738,
      "balance_loss_mlp": 1.01969528,
      "epoch": 0.4897492935730175,
      "flos": 12531065207040.0,
      "grad_norm": 1.8638707318715113,
      "language_loss": 0.90844709,
      "learning_rate": 2.163401712495504e-06,
      "loss": 0.93031794,
      "num_input_tokens_seen": 87731280,
      "step": 4073,
      "time_per_iteration": 2.6344895362854004
    },
    {
      "auxiliary_loss_clip": 0.01132353,
      "auxiliary_loss_mlp": 0.0103026,
      "balance_loss_clip": 1.04846847,
      "balance_loss_mlp": 1.02235055,
      "epoch": 0.4898695364636566,
      "flos": 23476816679040.0,
      "grad_norm": 1.710853189510822,
      "language_loss": 0.79091293,
      "learning_rate": 2.1626253326564194e-06,
      "loss": 0.81253904,
      "num_input_tokens_seen": 87750230,
      "step": 4074,
      "time_per_iteration": 2.7451934814453125
    },
    {
      "auxiliary_loss_clip": 0.01152877,
      "auxiliary_loss_mlp": 0.01029501,
      "balance_loss_clip": 1.04649282,
      "balance_loss_mlp": 1.02090561,
      "epoch": 0.48998977935429566,
      "flos": 27160209774720.0,
      "grad_norm": 1.9130167945045482,
      "language_loss": 0.7686376,
      "learning_rate": 2.161848928147201e-06,
      "loss": 0.79046136,
      "num_input_tokens_seen": 87770500,
      "step": 4075,
      "time_per_iteration": 2.6714491844177246
    },
    {
      "auxiliary_loss_clip": 0.01170404,
      "auxiliary_loss_mlp": 0.01030441,
      "balance_loss_clip": 1.05107105,
      "balance_loss_mlp": 1.02219176,
      "epoch": 0.4901100222449348,
      "flos": 20339588856960.0,
      "grad_norm": 2.034902183705985,
      "language_loss": 0.80558813,
      "learning_rate": 2.161072499085629e-06,
      "loss": 0.82759655,
      "num_input_tokens_seen": 87789495,
      "step": 4076,
      "time_per_iteration": 2.6522562503814697
    },
    {
      "auxiliary_loss_clip": 0.01143228,
      "auxiliary_loss_mlp": 0.01038525,
      "balance_loss_clip": 1.04693568,
      "balance_loss_mlp": 1.03016281,
      "epoch": 0.4902302651355739,
      "flos": 30446359384320.0,
      "grad_norm": 1.6363658860335009,
      "language_loss": 0.83440924,
      "learning_rate": 2.160296045589487e-06,
      "loss": 0.85622674,
      "num_input_tokens_seen": 87812955,
      "step": 4077,
      "time_per_iteration": 2.7512543201446533
    },
    {
      "auxiliary_loss_clip": 0.01170685,
      "auxiliary_loss_mlp": 0.01032587,
      "balance_loss_clip": 1.05168092,
      "balance_loss_mlp": 1.02388465,
      "epoch": 0.49035050802621294,
      "flos": 19174180089600.0,
      "grad_norm": 3.3759790351846024,
      "language_loss": 0.70004767,
      "learning_rate": 2.159519567776562e-06,
      "loss": 0.72208041,
      "num_input_tokens_seen": 87832605,
      "step": 4078,
      "time_per_iteration": 2.691570281982422
    },
    {
      "auxiliary_loss_clip": 0.0112854,
      "auxiliary_loss_mlp": 0.01031798,
      "balance_loss_clip": 1.04148269,
      "balance_loss_mlp": 1.02312589,
      "epoch": 0.49047075091685205,
      "flos": 22228489365120.0,
      "grad_norm": 2.8256192291519913,
      "language_loss": 0.71258569,
      "learning_rate": 2.1587430657646463e-06,
      "loss": 0.73418903,
      "num_input_tokens_seen": 87846040,
      "step": 4079,
      "time_per_iteration": 2.760190963745117
    },
    {
      "auxiliary_loss_clip": 0.01153701,
      "auxiliary_loss_mlp": 0.01028954,
      "balance_loss_clip": 1.04955506,
      "balance_loss_mlp": 1.02108061,
      "epoch": 0.4905909938074911,
      "flos": 20156516213760.0,
      "grad_norm": 2.5059736149971266,
      "language_loss": 0.77987838,
      "learning_rate": 2.157966539671533e-06,
      "loss": 0.80170494,
      "num_input_tokens_seen": 87865680,
      "step": 4080,
      "time_per_iteration": 2.7427022457122803
    },
    {
      "auxiliary_loss_clip": 0.01142696,
      "auxiliary_loss_mlp": 0.01029615,
      "balance_loss_clip": 1.04582715,
      "balance_loss_mlp": 1.02194357,
      "epoch": 0.4907112366981302,
      "flos": 17202217380480.0,
      "grad_norm": 2.12984616130428,
      "language_loss": 0.67401123,
      "learning_rate": 2.157189989615021e-06,
      "loss": 0.69573432,
      "num_input_tokens_seen": 87884270,
      "step": 4081,
      "time_per_iteration": 2.6889097690582275
    },
    {
      "auxiliary_loss_clip": 0.01172684,
      "auxiliary_loss_mlp": 0.0076413,
      "balance_loss_clip": 1.05009949,
      "balance_loss_mlp": 1.00033212,
      "epoch": 0.4908314795887693,
      "flos": 21688968107520.0,
      "grad_norm": 1.8568779859971267,
      "language_loss": 0.75280428,
      "learning_rate": 2.156413415712913e-06,
      "loss": 0.77217245,
      "num_input_tokens_seen": 87906320,
      "step": 4082,
      "time_per_iteration": 2.6974568367004395
    },
    {
      "auxiliary_loss_clip": 0.01162247,
      "auxiliary_loss_mlp": 0.00763191,
      "balance_loss_clip": 1.0506587,
      "balance_loss_mlp": 1.00033545,
      "epoch": 0.4909517224794084,
      "flos": 26213676531840.0,
      "grad_norm": 1.730208412106505,
      "language_loss": 0.78495967,
      "learning_rate": 2.155636818083014e-06,
      "loss": 0.804214,
      "num_input_tokens_seen": 87927690,
      "step": 4083,
      "time_per_iteration": 2.761655807495117
    },
    {
      "auxiliary_loss_clip": 0.01151961,
      "auxiliary_loss_mlp": 0.01032021,
      "balance_loss_clip": 1.05004764,
      "balance_loss_mlp": 1.02439451,
      "epoch": 0.4910719653700475,
      "flos": 23148377694720.0,
      "grad_norm": 1.7901627369877156,
      "language_loss": 0.84180713,
      "learning_rate": 2.154860196843134e-06,
      "loss": 0.86364698,
      "num_input_tokens_seen": 87946885,
      "step": 4084,
      "time_per_iteration": 2.696606159210205
    },
    {
      "auxiliary_loss_clip": 0.01184547,
      "auxiliary_loss_mlp": 0.01029311,
      "balance_loss_clip": 1.05186617,
      "balance_loss_mlp": 1.02088273,
      "epoch": 0.4911922082606866,
      "flos": 23331845387520.0,
      "grad_norm": 2.2982421754714935,
      "language_loss": 0.76897347,
      "learning_rate": 2.154083552111085e-06,
      "loss": 0.79111207,
      "num_input_tokens_seen": 87966055,
      "step": 4085,
      "time_per_iteration": 2.5587785243988037
    },
    {
      "auxiliary_loss_clip": 0.01182075,
      "auxiliary_loss_mlp": 0.01034172,
      "balance_loss_clip": 1.04827619,
      "balance_loss_mlp": 1.02608371,
      "epoch": 0.49131245115132566,
      "flos": 29203239542400.0,
      "grad_norm": 1.7360014315407375,
      "language_loss": 0.81835556,
      "learning_rate": 2.1533068840046834e-06,
      "loss": 0.840518,
      "num_input_tokens_seen": 87986320,
      "step": 4086,
      "time_per_iteration": 2.6501095294952393
    },
    {
      "auxiliary_loss_clip": 0.01148733,
      "auxiliary_loss_mlp": 0.00763144,
      "balance_loss_clip": 1.04698801,
      "balance_loss_mlp": 1.00035107,
      "epoch": 0.49143269404196477,
      "flos": 20147465986560.0,
      "grad_norm": 2.375628441513896,
      "language_loss": 0.61247587,
      "learning_rate": 2.152530192641749e-06,
      "loss": 0.63159466,
      "num_input_tokens_seen": 88001230,
      "step": 4087,
      "time_per_iteration": 2.6306891441345215
    },
    {
      "auxiliary_loss_clip": 0.01173331,
      "auxiliary_loss_mlp": 0.0102664,
      "balance_loss_clip": 1.05061734,
      "balance_loss_mlp": 1.01833737,
      "epoch": 0.4915529369326039,
      "flos": 24389809597440.0,
      "grad_norm": 1.9440137445605954,
      "language_loss": 0.72867507,
      "learning_rate": 2.1517534781401068e-06,
      "loss": 0.75067478,
      "num_input_tokens_seen": 88019110,
      "step": 4088,
      "time_per_iteration": 2.6390931606292725
    },
    {
      "auxiliary_loss_clip": 0.01166256,
      "auxiliary_loss_mlp": 0.01025407,
      "balance_loss_clip": 1.04957414,
      "balance_loss_mlp": 1.01737189,
      "epoch": 0.49167317982324293,
      "flos": 10524305197440.0,
      "grad_norm": 2.449773637474804,
      "language_loss": 0.69400454,
      "learning_rate": 2.150976740617581e-06,
      "loss": 0.71592116,
      "num_input_tokens_seen": 88035670,
      "step": 4089,
      "time_per_iteration": 2.6393792629241943
    },
    {
      "auxiliary_loss_clip": 0.01161179,
      "auxiliary_loss_mlp": 0.01032173,
      "balance_loss_clip": 1.05125523,
      "balance_loss_mlp": 1.02431703,
      "epoch": 0.49179342271388204,
      "flos": 25593427457280.0,
      "grad_norm": 1.8800139399571998,
      "language_loss": 0.71559882,
      "learning_rate": 2.150199980192006e-06,
      "loss": 0.73753232,
      "num_input_tokens_seen": 88054790,
      "step": 4090,
      "time_per_iteration": 2.6972641944885254
    },
    {
      "auxiliary_loss_clip": 0.01150665,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.04709995,
      "balance_loss_mlp": 1.01849198,
      "epoch": 0.49191366560452116,
      "flos": 21102043875840.0,
      "grad_norm": 2.1088631241190656,
      "language_loss": 0.80847597,
      "learning_rate": 2.1494231969812114e-06,
      "loss": 0.83024633,
      "num_input_tokens_seen": 88073780,
      "step": 4091,
      "time_per_iteration": 2.6933693885803223
    },
    {
      "auxiliary_loss_clip": 0.01146995,
      "auxiliary_loss_mlp": 0.01026095,
      "balance_loss_clip": 1.04856014,
      "balance_loss_mlp": 1.01836467,
      "epoch": 0.4920339084951602,
      "flos": 26067520091520.0,
      "grad_norm": 2.868231388971745,
      "language_loss": 0.80834234,
      "learning_rate": 2.1486463911030372e-06,
      "loss": 0.83007324,
      "num_input_tokens_seen": 88094430,
      "step": 4092,
      "time_per_iteration": 2.8007242679595947
    },
    {
      "auxiliary_loss_clip": 0.01153592,
      "auxiliary_loss_mlp": 0.0102721,
      "balance_loss_clip": 1.04745364,
      "balance_loss_mlp": 1.0193063,
      "epoch": 0.4921541513857993,
      "flos": 25081269384960.0,
      "grad_norm": 1.8997663034763077,
      "language_loss": 0.74524403,
      "learning_rate": 2.147869562675324e-06,
      "loss": 0.76705205,
      "num_input_tokens_seen": 88113400,
      "step": 4093,
      "time_per_iteration": 3.6766157150268555
    },
    {
      "auxiliary_loss_clip": 0.01173189,
      "auxiliary_loss_mlp": 0.01032337,
      "balance_loss_clip": 1.05245864,
      "balance_loss_mlp": 1.0239867,
      "epoch": 0.49227439427643843,
      "flos": 24389809597440.0,
      "grad_norm": 1.9531436105029472,
      "language_loss": 0.72161031,
      "learning_rate": 2.147092711815915e-06,
      "loss": 0.74366552,
      "num_input_tokens_seen": 88132750,
      "step": 4094,
      "time_per_iteration": 3.6719284057617188
    },
    {
      "auxiliary_loss_clip": 0.01136811,
      "auxiliary_loss_mlp": 0.0102926,
      "balance_loss_clip": 1.04805553,
      "balance_loss_mlp": 1.0208919,
      "epoch": 0.4923946371670775,
      "flos": 11363753018880.0,
      "grad_norm": 2.404865874371887,
      "language_loss": 0.86751288,
      "learning_rate": 2.1463158386426593e-06,
      "loss": 0.88917363,
      "num_input_tokens_seen": 88150560,
      "step": 4095,
      "time_per_iteration": 2.683340311050415
    },
    {
      "auxiliary_loss_clip": 0.01162571,
      "auxiliary_loss_mlp": 0.01026884,
      "balance_loss_clip": 1.05012369,
      "balance_loss_mlp": 1.0184412,
      "epoch": 0.4925148800577166,
      "flos": 30445964334720.0,
      "grad_norm": 2.3295843227400215,
      "language_loss": 0.77586412,
      "learning_rate": 2.145538943273407e-06,
      "loss": 0.7977587,
      "num_input_tokens_seen": 88170835,
      "step": 4096,
      "time_per_iteration": 2.772714376449585
    },
    {
      "auxiliary_loss_clip": 0.01188493,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.0555625,
      "balance_loss_mlp": 1.01989818,
      "epoch": 0.49263512294835565,
      "flos": 20850454039680.0,
      "grad_norm": 2.0921426172300897,
      "language_loss": 0.72061694,
      "learning_rate": 2.144762025826013e-06,
      "loss": 0.74277842,
      "num_input_tokens_seen": 88189925,
      "step": 4097,
      "time_per_iteration": 4.410549879074097
    },
    {
      "auxiliary_loss_clip": 0.0117341,
      "auxiliary_loss_mlp": 0.01024052,
      "balance_loss_clip": 1.0487628,
      "balance_loss_mlp": 1.01594603,
      "epoch": 0.49275536583899476,
      "flos": 23767477534080.0,
      "grad_norm": 2.1707610089838227,
      "language_loss": 0.86893791,
      "learning_rate": 2.143985086418334e-06,
      "loss": 0.89091253,
      "num_input_tokens_seen": 88205105,
      "step": 4098,
      "time_per_iteration": 2.6296422481536865
    },
    {
      "auxiliary_loss_clip": 0.01155232,
      "auxiliary_loss_mlp": 0.01029013,
      "balance_loss_clip": 1.04794836,
      "balance_loss_mlp": 1.02158034,
      "epoch": 0.4928756087296339,
      "flos": 22273522041600.0,
      "grad_norm": 1.331473670637252,
      "language_loss": 0.76684403,
      "learning_rate": 2.1432081251682324e-06,
      "loss": 0.78868645,
      "num_input_tokens_seen": 88225475,
      "step": 4099,
      "time_per_iteration": 2.684675931930542
    },
    {
      "auxiliary_loss_clip": 0.01168327,
      "auxiliary_loss_mlp": 0.01034698,
      "balance_loss_clip": 1.05156302,
      "balance_loss_mlp": 1.02676511,
      "epoch": 0.49299585162027293,
      "flos": 19645471463040.0,
      "grad_norm": 1.669869483234081,
      "language_loss": 0.87121487,
      "learning_rate": 2.142431142193572e-06,
      "loss": 0.8932451,
      "num_input_tokens_seen": 88243255,
      "step": 4100,
      "time_per_iteration": 2.5751090049743652
    },
    {
      "auxiliary_loss_clip": 0.01184298,
      "auxiliary_loss_mlp": 0.01026966,
      "balance_loss_clip": 1.05269432,
      "balance_loss_mlp": 1.01883602,
      "epoch": 0.49311609451091204,
      "flos": 38837138497920.0,
      "grad_norm": 2.6088800958001253,
      "language_loss": 0.71678621,
      "learning_rate": 2.1416541376122207e-06,
      "loss": 0.73889887,
      "num_input_tokens_seen": 88263435,
      "step": 4101,
      "time_per_iteration": 2.7626793384552
    },
    {
      "auxiliary_loss_clip": 0.01180541,
      "auxiliary_loss_mlp": 0.0102477,
      "balance_loss_clip": 1.04883456,
      "balance_loss_mlp": 1.01637197,
      "epoch": 0.49323633740155115,
      "flos": 28329102161280.0,
      "grad_norm": 2.381386229599501,
      "language_loss": 0.72925907,
      "learning_rate": 2.1408771115420496e-06,
      "loss": 0.7513122,
      "num_input_tokens_seen": 88283295,
      "step": 4102,
      "time_per_iteration": 2.6249895095825195
    },
    {
      "auxiliary_loss_clip": 0.01131568,
      "auxiliary_loss_mlp": 0.01027462,
      "balance_loss_clip": 1.05114698,
      "balance_loss_mlp": 1.01974893,
      "epoch": 0.4933565802921902,
      "flos": 21135584200320.0,
      "grad_norm": 11.304618949356206,
      "language_loss": 0.64868879,
      "learning_rate": 2.140100064100932e-06,
      "loss": 0.67027903,
      "num_input_tokens_seen": 88299270,
      "step": 4103,
      "time_per_iteration": 2.748246431350708
    },
    {
      "auxiliary_loss_clip": 0.01169092,
      "auxiliary_loss_mlp": 0.01031006,
      "balance_loss_clip": 1.05214405,
      "balance_loss_mlp": 1.02324522,
      "epoch": 0.4934768231828293,
      "flos": 18039007595520.0,
      "grad_norm": 1.8701455087462335,
      "language_loss": 0.76028866,
      "learning_rate": 2.139322995406746e-06,
      "loss": 0.78228962,
      "num_input_tokens_seen": 88316905,
      "step": 4104,
      "time_per_iteration": 2.6235668659210205
    },
    {
      "auxiliary_loss_clip": 0.01185977,
      "auxiliary_loss_mlp": 0.01034584,
      "balance_loss_clip": 1.05475068,
      "balance_loss_mlp": 1.02577174,
      "epoch": 0.4935970660734684,
      "flos": 23469957181440.0,
      "grad_norm": 2.4098479545284834,
      "language_loss": 0.79656112,
      "learning_rate": 2.1385459055773727e-06,
      "loss": 0.81876677,
      "num_input_tokens_seen": 88335095,
      "step": 4105,
      "time_per_iteration": 2.630225419998169
    },
    {
      "auxiliary_loss_clip": 0.01112821,
      "auxiliary_loss_mlp": 0.00762343,
      "balance_loss_clip": 1.04274297,
      "balance_loss_mlp": 1.00029659,
      "epoch": 0.4937173089641075,
      "flos": 64479258840960.0,
      "grad_norm": 2.0517292677830823,
      "language_loss": 0.73960066,
      "learning_rate": 2.137768794730696e-06,
      "loss": 0.75835228,
      "num_input_tokens_seen": 88358545,
      "step": 4106,
      "time_per_iteration": 3.2441718578338623
    },
    {
      "auxiliary_loss_clip": 0.01160863,
      "auxiliary_loss_mlp": 0.01027982,
      "balance_loss_clip": 1.05070603,
      "balance_loss_mlp": 1.0200367,
      "epoch": 0.4938375518547466,
      "flos": 22346025644160.0,
      "grad_norm": 1.7906919532505554,
      "language_loss": 0.80322134,
      "learning_rate": 2.1369916629846026e-06,
      "loss": 0.82510978,
      "num_input_tokens_seen": 88378295,
      "step": 4107,
      "time_per_iteration": 2.686285972595215
    },
    {
      "auxiliary_loss_clip": 0.01158142,
      "auxiliary_loss_mlp": 0.01031275,
      "balance_loss_clip": 1.04958606,
      "balance_loss_mlp": 1.02312088,
      "epoch": 0.4939577947453857,
      "flos": 17858700299520.0,
      "grad_norm": 1.8561989515385042,
      "language_loss": 0.75189698,
      "learning_rate": 2.136214510456983e-06,
      "loss": 0.77379119,
      "num_input_tokens_seen": 88396750,
      "step": 4108,
      "time_per_iteration": 2.732663631439209
    },
    {
      "auxiliary_loss_clip": 0.01045805,
      "auxiliary_loss_mlp": 0.0075438,
      "balance_loss_clip": 1.01581335,
      "balance_loss_mlp": 1.00030363,
      "epoch": 0.49407803763602476,
      "flos": 70066746875520.0,
      "grad_norm": 0.8892520658895272,
      "language_loss": 0.63100636,
      "learning_rate": 2.1354373372657296e-06,
      "loss": 0.64900815,
      "num_input_tokens_seen": 88455190,
      "step": 4109,
      "time_per_iteration": 3.344489097595215
    },
    {
      "auxiliary_loss_clip": 0.01185234,
      "auxiliary_loss_mlp": 0.01029891,
      "balance_loss_clip": 1.05351865,
      "balance_loss_mlp": 1.02161765,
      "epoch": 0.49419828052666387,
      "flos": 24317485562880.0,
      "grad_norm": 1.490218012607458,
      "language_loss": 0.71228987,
      "learning_rate": 2.1346601435287404e-06,
      "loss": 0.7344411,
      "num_input_tokens_seen": 88477460,
      "step": 4110,
      "time_per_iteration": 2.6385204792022705
    },
    {
      "auxiliary_loss_clip": 0.01153896,
      "auxiliary_loss_mlp": 0.01027967,
      "balance_loss_clip": 1.0473659,
      "balance_loss_mlp": 1.01969373,
      "epoch": 0.494318523417303,
      "flos": 29386060790400.0,
      "grad_norm": 1.8043204520351457,
      "language_loss": 0.80026066,
      "learning_rate": 2.1338829293639144e-06,
      "loss": 0.8220793,
      "num_input_tokens_seen": 88497820,
      "step": 4111,
      "time_per_iteration": 2.709716796875
    },
    {
      "auxiliary_loss_clip": 0.01125842,
      "auxiliary_loss_mlp": 0.01027986,
      "balance_loss_clip": 1.04519916,
      "balance_loss_mlp": 1.01980865,
      "epoch": 0.49443876630794203,
      "flos": 15268284195840.0,
      "grad_norm": 2.2363337158099106,
      "language_loss": 0.8325671,
      "learning_rate": 2.1331056948891547e-06,
      "loss": 0.85410535,
      "num_input_tokens_seen": 88514920,
      "step": 4112,
      "time_per_iteration": 2.722691774368286
    },
    {
      "auxiliary_loss_clip": 0.01150664,
      "auxiliary_loss_mlp": 0.01025535,
      "balance_loss_clip": 1.04771626,
      "balance_loss_mlp": 1.0171603,
      "epoch": 0.49455900919858115,
      "flos": 12347453859840.0,
      "grad_norm": 2.3500485954076185,
      "language_loss": 0.76494336,
      "learning_rate": 2.1323284402223666e-06,
      "loss": 0.78670526,
      "num_input_tokens_seen": 88530910,
      "step": 4113,
      "time_per_iteration": 2.5932061672210693
    },
    {
      "auxiliary_loss_clip": 0.01184368,
      "auxiliary_loss_mlp": 0.00762195,
      "balance_loss_clip": 1.05610955,
      "balance_loss_mlp": 1.0002811,
      "epoch": 0.4946792520892202,
      "flos": 22779610715520.0,
      "grad_norm": 1.847822677149436,
      "language_loss": 0.882128,
      "learning_rate": 2.1315511654814597e-06,
      "loss": 0.90159357,
      "num_input_tokens_seen": 88549320,
      "step": 4114,
      "time_per_iteration": 2.670654535293579
    },
    {
      "auxiliary_loss_clip": 0.01147064,
      "auxiliary_loss_mlp": 0.01024047,
      "balance_loss_clip": 1.04894948,
      "balance_loss_mlp": 1.01671004,
      "epoch": 0.4947994949798593,
      "flos": 23148126299520.0,
      "grad_norm": 1.9344431933000936,
      "language_loss": 0.78470993,
      "learning_rate": 2.1307738707843456e-06,
      "loss": 0.80642104,
      "num_input_tokens_seen": 88568985,
      "step": 4115,
      "time_per_iteration": 2.7016329765319824
    },
    {
      "auxiliary_loss_clip": 0.01177245,
      "auxiliary_loss_mlp": 0.01030549,
      "balance_loss_clip": 1.05352461,
      "balance_loss_mlp": 1.02185309,
      "epoch": 0.4949197378704984,
      "flos": 23659997063040.0,
      "grad_norm": 1.9981652842888369,
      "language_loss": 0.69469088,
      "learning_rate": 2.1299965562489385e-06,
      "loss": 0.71676886,
      "num_input_tokens_seen": 88588790,
      "step": 4116,
      "time_per_iteration": 2.6574652194976807
    },
    {
      "auxiliary_loss_clip": 0.01167965,
      "auxiliary_loss_mlp": 0.01023854,
      "balance_loss_clip": 1.04925251,
      "balance_loss_mlp": 1.01598024,
      "epoch": 0.4950399807611375,
      "flos": 26911493026560.0,
      "grad_norm": 1.5714116964184477,
      "language_loss": 0.79169607,
      "learning_rate": 2.129219221993158e-06,
      "loss": 0.81361425,
      "num_input_tokens_seen": 88613575,
      "step": 4117,
      "time_per_iteration": 2.7980785369873047
    },
    {
      "auxiliary_loss_clip": 0.01047248,
      "auxiliary_loss_mlp": 0.01003682,
      "balance_loss_clip": 1.0129602,
      "balance_loss_mlp": 1.00219786,
      "epoch": 0.4951602236517766,
      "flos": 67315270187520.0,
      "grad_norm": 0.8041599280679855,
      "language_loss": 0.59985292,
      "learning_rate": 2.128441868134924e-06,
      "loss": 0.62036216,
      "num_input_tokens_seen": 88675510,
      "step": 4118,
      "time_per_iteration": 3.2767646312713623
    },
    {
      "auxiliary_loss_clip": 0.01144108,
      "auxiliary_loss_mlp": 0.01031176,
      "balance_loss_clip": 1.04797006,
      "balance_loss_mlp": 1.02389836,
      "epoch": 0.4952804665424157,
      "flos": 19901442758400.0,
      "grad_norm": 2.0731981528984167,
      "language_loss": 0.82945895,
      "learning_rate": 2.1276644947921606e-06,
      "loss": 0.85121179,
      "num_input_tokens_seen": 88694425,
      "step": 4119,
      "time_per_iteration": 3.7208800315856934
    },
    {
      "auxiliary_loss_clip": 0.01169331,
      "auxiliary_loss_mlp": 0.01025335,
      "balance_loss_clip": 1.04958892,
      "balance_loss_mlp": 1.01623344,
      "epoch": 0.49540070943305475,
      "flos": 18806813740800.0,
      "grad_norm": 2.9288430585415,
      "language_loss": 0.82257468,
      "learning_rate": 2.126887102082795e-06,
      "loss": 0.84452128,
      "num_input_tokens_seen": 88714450,
      "step": 4120,
      "time_per_iteration": 3.619898796081543
    },
    {
      "auxiliary_loss_clip": 0.01141019,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.04374158,
      "balance_loss_mlp": 1.02048659,
      "epoch": 0.49552095232369386,
      "flos": 24934179191040.0,
      "grad_norm": 2.22454144235084,
      "language_loss": 0.70327747,
      "learning_rate": 2.126109690124757e-06,
      "loss": 0.72496712,
      "num_input_tokens_seen": 88735265,
      "step": 4121,
      "time_per_iteration": 2.8440003395080566
    },
    {
      "auxiliary_loss_clip": 0.01127625,
      "auxiliary_loss_mlp": 0.01030154,
      "balance_loss_clip": 1.0462184,
      "balance_loss_mlp": 1.02212512,
      "epoch": 0.495641195214333,
      "flos": 22857249962880.0,
      "grad_norm": 1.7034075690877302,
      "language_loss": 0.71123517,
      "learning_rate": 2.1253322590359786e-06,
      "loss": 0.732813,
      "num_input_tokens_seen": 88754600,
      "step": 4122,
      "time_per_iteration": 3.6207664012908936
    },
    {
      "auxiliary_loss_clip": 0.01166419,
      "auxiliary_loss_mlp": 0.01034411,
      "balance_loss_clip": 1.04917312,
      "balance_loss_mlp": 1.02631116,
      "epoch": 0.49576143810497203,
      "flos": 25769748343680.0,
      "grad_norm": 1.693893121970706,
      "language_loss": 0.7412222,
      "learning_rate": 2.124554808934397e-06,
      "loss": 0.76323044,
      "num_input_tokens_seen": 88775180,
      "step": 4123,
      "time_per_iteration": 3.6027958393096924
    },
    {
      "auxiliary_loss_clip": 0.01118748,
      "auxiliary_loss_mlp": 0.01024752,
      "balance_loss_clip": 1.04098845,
      "balance_loss_mlp": 1.01635408,
      "epoch": 0.49588168099561114,
      "flos": 22128838058880.0,
      "grad_norm": 1.8068187518465213,
      "language_loss": 0.72951543,
      "learning_rate": 2.1237773399379496e-06,
      "loss": 0.75095046,
      "num_input_tokens_seen": 88796145,
      "step": 4124,
      "time_per_iteration": 2.7604830265045166
    },
    {
      "auxiliary_loss_clip": 0.01158182,
      "auxiliary_loss_mlp": 0.01034498,
      "balance_loss_clip": 1.04438519,
      "balance_loss_mlp": 1.02677357,
      "epoch": 0.49600192388625025,
      "flos": 24387331559040.0,
      "grad_norm": 1.724789406572805,
      "language_loss": 0.86896372,
      "learning_rate": 2.122999852164578e-06,
      "loss": 0.89089048,
      "num_input_tokens_seen": 88816765,
      "step": 4125,
      "time_per_iteration": 2.7239248752593994
    },
    {
      "auxiliary_loss_clip": 0.0112345,
      "auxiliary_loss_mlp": 0.01025626,
      "balance_loss_clip": 1.0465157,
      "balance_loss_mlp": 1.0180738,
      "epoch": 0.4961221667768893,
      "flos": 22857429530880.0,
      "grad_norm": 2.253252959821838,
      "language_loss": 0.58345854,
      "learning_rate": 2.122222345732227e-06,
      "loss": 0.60494936,
      "num_input_tokens_seen": 88836680,
      "step": 4126,
      "time_per_iteration": 2.7814407348632812
    },
    {
      "auxiliary_loss_clip": 0.01142435,
      "auxiliary_loss_mlp": 0.01032125,
      "balance_loss_clip": 1.04801595,
      "balance_loss_mlp": 1.02371526,
      "epoch": 0.4962424096675284,
      "flos": 17858089768320.0,
      "grad_norm": 1.98822415818956,
      "language_loss": 0.83065629,
      "learning_rate": 2.121444820758843e-06,
      "loss": 0.85240191,
      "num_input_tokens_seen": 88855320,
      "step": 4127,
      "time_per_iteration": 2.7356035709381104
    },
    {
      "auxiliary_loss_clip": 0.01128978,
      "auxiliary_loss_mlp": 0.01037388,
      "balance_loss_clip": 1.04795361,
      "balance_loss_mlp": 1.02882862,
      "epoch": 0.49636265255816747,
      "flos": 21793611404160.0,
      "grad_norm": 1.9130037154146204,
      "language_loss": 0.78495699,
      "learning_rate": 2.120667277362376e-06,
      "loss": 0.80662066,
      "num_input_tokens_seen": 88874035,
      "step": 4128,
      "time_per_iteration": 2.678474187850952
    },
    {
      "auxiliary_loss_clip": 0.01187913,
      "auxiliary_loss_mlp": 0.01029497,
      "balance_loss_clip": 1.05517423,
      "balance_loss_mlp": 1.02111053,
      "epoch": 0.4964828954488066,
      "flos": 16358603581440.0,
      "grad_norm": 2.1228242837549813,
      "language_loss": 0.84747052,
      "learning_rate": 2.1198897156607796e-06,
      "loss": 0.86964464,
      "num_input_tokens_seen": 88891390,
      "step": 4129,
      "time_per_iteration": 2.5810461044311523
    },
    {
      "auxiliary_loss_clip": 0.01172821,
      "auxiliary_loss_mlp": 0.01024893,
      "balance_loss_clip": 1.04895592,
      "balance_loss_mlp": 1.01696014,
      "epoch": 0.4966031383394457,
      "flos": 24711101775360.0,
      "grad_norm": 2.0430825167533615,
      "language_loss": 0.73699498,
      "learning_rate": 2.1191121357720085e-06,
      "loss": 0.75897211,
      "num_input_tokens_seen": 88909450,
      "step": 4130,
      "time_per_iteration": 2.6485555171966553
    },
    {
      "auxiliary_loss_clip": 0.01118939,
      "auxiliary_loss_mlp": 0.01028717,
      "balance_loss_clip": 1.04429197,
      "balance_loss_mlp": 1.02052689,
      "epoch": 0.49672338123008475,
      "flos": 22930615491840.0,
      "grad_norm": 1.6946415783410982,
      "language_loss": 0.74953175,
      "learning_rate": 2.1183345378140206e-06,
      "loss": 0.77100837,
      "num_input_tokens_seen": 88929195,
      "step": 4131,
      "time_per_iteration": 2.7105612754821777
    },
    {
      "auxiliary_loss_clip": 0.01072935,
      "auxiliary_loss_mlp": 0.01001334,
      "balance_loss_clip": 1.01489091,
      "balance_loss_mlp": 0.99972516,
      "epoch": 0.49684362412072386,
      "flos": 65976736844160.0,
      "grad_norm": 0.8532066866492722,
      "language_loss": 0.61935806,
      "learning_rate": 2.1175569219047783e-06,
      "loss": 0.64010078,
      "num_input_tokens_seen": 88990635,
      "step": 4132,
      "time_per_iteration": 3.3240840435028076
    },
    {
      "auxiliary_loss_clip": 0.01184481,
      "auxiliary_loss_mlp": 0.010343,
      "balance_loss_clip": 1.05268133,
      "balance_loss_mlp": 1.02647412,
      "epoch": 0.49696386701136297,
      "flos": 19971288754560.0,
      "grad_norm": 1.5349231611445002,
      "language_loss": 0.73317599,
      "learning_rate": 2.1167792881622437e-06,
      "loss": 0.75536382,
      "num_input_tokens_seen": 89009655,
      "step": 4133,
      "time_per_iteration": 2.553307056427002
    },
    {
      "auxiliary_loss_clip": 0.01151527,
      "auxiliary_loss_mlp": 0.01028289,
      "balance_loss_clip": 1.04891539,
      "balance_loss_mlp": 1.02070165,
      "epoch": 0.497084109902002,
      "flos": 24750819239040.0,
      "grad_norm": 1.669046828320635,
      "language_loss": 0.81043762,
      "learning_rate": 2.116001636704384e-06,
      "loss": 0.83223581,
      "num_input_tokens_seen": 89030040,
      "step": 4134,
      "time_per_iteration": 2.734523296356201
    },
    {
      "auxiliary_loss_clip": 0.01135591,
      "auxiliary_loss_mlp": 0.01034417,
      "balance_loss_clip": 1.04701567,
      "balance_loss_mlp": 1.02666235,
      "epoch": 0.49720435279264114,
      "flos": 21871825269120.0,
      "grad_norm": 1.8379272816105148,
      "language_loss": 0.80255038,
      "learning_rate": 2.1152239676491685e-06,
      "loss": 0.82425046,
      "num_input_tokens_seen": 89048145,
      "step": 4135,
      "time_per_iteration": 2.7451138496398926
    },
    {
      "auxiliary_loss_clip": 0.01160341,
      "auxiliary_loss_mlp": 0.01029276,
      "balance_loss_clip": 1.04834509,
      "balance_loss_mlp": 1.02206111,
      "epoch": 0.49732459568328025,
      "flos": 23805794367360.0,
      "grad_norm": 2.362006028466471,
      "language_loss": 0.73669553,
      "learning_rate": 2.114446281114569e-06,
      "loss": 0.75859177,
      "num_input_tokens_seen": 89067165,
      "step": 4136,
      "time_per_iteration": 2.711787223815918
    },
    {
      "auxiliary_loss_clip": 0.01146488,
      "auxiliary_loss_mlp": 0.01027282,
      "balance_loss_clip": 1.04770374,
      "balance_loss_mlp": 1.01937842,
      "epoch": 0.4974448385739193,
      "flos": 20047742853120.0,
      "grad_norm": 1.864077435653133,
      "language_loss": 0.76158333,
      "learning_rate": 2.1136685772185587e-06,
      "loss": 0.78332102,
      "num_input_tokens_seen": 89086190,
      "step": 4137,
      "time_per_iteration": 2.697575807571411
    },
    {
      "auxiliary_loss_clip": 0.01153819,
      "auxiliary_loss_mlp": 0.00763665,
      "balance_loss_clip": 1.04264033,
      "balance_loss_mlp": 1.00022888,
      "epoch": 0.4975650814645584,
      "flos": 24821347593600.0,
      "grad_norm": 1.6234469831281209,
      "language_loss": 0.78013515,
      "learning_rate": 2.1128908560791163e-06,
      "loss": 0.79930997,
      "num_input_tokens_seen": 89106020,
      "step": 4138,
      "time_per_iteration": 2.766541004180908
    },
    {
      "auxiliary_loss_clip": 0.01182174,
      "auxiliary_loss_mlp": 0.01031447,
      "balance_loss_clip": 1.05120981,
      "balance_loss_mlp": 1.0228343,
      "epoch": 0.4976853243551975,
      "flos": 19829477859840.0,
      "grad_norm": 5.276923367354073,
      "language_loss": 0.78259593,
      "learning_rate": 2.1121131178142203e-06,
      "loss": 0.8047322,
      "num_input_tokens_seen": 89125385,
      "step": 4139,
      "time_per_iteration": 2.6261820793151855
    },
    {
      "auxiliary_loss_clip": 0.01151157,
      "auxiliary_loss_mlp": 0.01025314,
      "balance_loss_clip": 1.04424405,
      "balance_loss_mlp": 1.01801884,
      "epoch": 0.4978055672458366,
      "flos": 23142990654720.0,
      "grad_norm": 1.466620322438816,
      "language_loss": 0.82298654,
      "learning_rate": 2.1113353625418544e-06,
      "loss": 0.84475124,
      "num_input_tokens_seen": 89143935,
      "step": 4140,
      "time_per_iteration": 2.7037353515625
    },
    {
      "auxiliary_loss_clip": 0.01162682,
      "auxiliary_loss_mlp": 0.01027907,
      "balance_loss_clip": 1.05176997,
      "balance_loss_mlp": 1.02005076,
      "epoch": 0.4979258101364757,
      "flos": 15559914718080.0,
      "grad_norm": 1.7659175170905714,
      "language_loss": 0.79045534,
      "learning_rate": 2.1105575903800017e-06,
      "loss": 0.81236118,
      "num_input_tokens_seen": 89162655,
      "step": 4141,
      "time_per_iteration": 2.653367280960083
    },
    {
      "auxiliary_loss_clip": 0.01171719,
      "auxiliary_loss_mlp": 0.01037627,
      "balance_loss_clip": 1.04850841,
      "balance_loss_mlp": 1.02874637,
      "epoch": 0.4980460530271148,
      "flos": 26356169784960.0,
      "grad_norm": 2.1191618418759437,
      "language_loss": 0.84992719,
      "learning_rate": 2.1097798014466502e-06,
      "loss": 0.8720206,
      "num_input_tokens_seen": 89182255,
      "step": 4142,
      "time_per_iteration": 2.682246446609497
    },
    {
      "auxiliary_loss_clip": 0.01174761,
      "auxiliary_loss_mlp": 0.0102879,
      "balance_loss_clip": 1.05123329,
      "balance_loss_mlp": 1.02066255,
      "epoch": 0.49816629591775385,
      "flos": 17274541415040.0,
      "grad_norm": 2.7531281231478766,
      "language_loss": 0.59096885,
      "learning_rate": 2.109001995859791e-06,
      "loss": 0.61300439,
      "num_input_tokens_seen": 89201155,
      "step": 4143,
      "time_per_iteration": 2.597564697265625
    },
    {
      "auxiliary_loss_clip": 0.01060778,
      "auxiliary_loss_mlp": 0.01002222,
      "balance_loss_clip": 1.01620376,
      "balance_loss_mlp": 1.00047016,
      "epoch": 0.49828653880839296,
      "flos": 64930947344640.0,
      "grad_norm": 0.7890878730322277,
      "language_loss": 0.59993899,
      "learning_rate": 2.108224173737415e-06,
      "loss": 0.62056899,
      "num_input_tokens_seen": 89264455,
      "step": 4144,
      "time_per_iteration": 3.270787477493286
    },
    {
      "auxiliary_loss_clip": 0.01154391,
      "auxiliary_loss_mlp": 0.01029356,
      "balance_loss_clip": 1.04761541,
      "balance_loss_mlp": 1.02090418,
      "epoch": 0.498406781699032,
      "flos": 27484806003840.0,
      "grad_norm": 1.9128306623219133,
      "language_loss": 0.76483285,
      "learning_rate": 2.1074463351975183e-06,
      "loss": 0.78667033,
      "num_input_tokens_seen": 89283340,
      "step": 4145,
      "time_per_iteration": 3.6887989044189453
    },
    {
      "auxiliary_loss_clip": 0.0114795,
      "auxiliary_loss_mlp": 0.0102616,
      "balance_loss_clip": 1.04773796,
      "balance_loss_mlp": 1.01898062,
      "epoch": 0.49852702458967113,
      "flos": 31499870307840.0,
      "grad_norm": 2.1924228883175534,
      "language_loss": 0.71676558,
      "learning_rate": 2.106668480358098e-06,
      "loss": 0.73850667,
      "num_input_tokens_seen": 89303565,
      "step": 4146,
      "time_per_iteration": 3.7092318534851074
    },
    {
      "auxiliary_loss_clip": 0.01150471,
      "auxiliary_loss_mlp": 0.01032706,
      "balance_loss_clip": 1.04567719,
      "balance_loss_mlp": 1.02396834,
      "epoch": 0.49864726748031024,
      "flos": 22852868503680.0,
      "grad_norm": 1.6171487995231857,
      "language_loss": 0.71018374,
      "learning_rate": 2.105890609337154e-06,
      "loss": 0.73201549,
      "num_input_tokens_seen": 89322080,
      "step": 4147,
      "time_per_iteration": 2.684776544570923
    },
    {
      "auxiliary_loss_clip": 0.0108241,
      "auxiliary_loss_mlp": 0.01002219,
      "balance_loss_clip": 1.01567578,
      "balance_loss_mlp": 1.00063324,
      "epoch": 0.4987675103709493,
      "flos": 70405708544640.0,
      "grad_norm": 0.6913399730511093,
      "language_loss": 0.63762987,
      "learning_rate": 2.1051127222526883e-06,
      "loss": 0.65847611,
      "num_input_tokens_seen": 89394195,
      "step": 4148,
      "time_per_iteration": 4.229222297668457
    },
    {
      "auxiliary_loss_clip": 0.01171325,
      "auxiliary_loss_mlp": 0.0102894,
      "balance_loss_clip": 1.05511522,
      "balance_loss_mlp": 1.02146578,
      "epoch": 0.4988877532615884,
      "flos": 28767571482240.0,
      "grad_norm": 1.6897640103667007,
      "language_loss": 0.80697227,
      "learning_rate": 2.1043348192227067e-06,
      "loss": 0.82897496,
      "num_input_tokens_seen": 89414565,
      "step": 4149,
      "time_per_iteration": 3.769097089767456
    },
    {
      "auxiliary_loss_clip": 0.01132054,
      "auxiliary_loss_mlp": 0.01028432,
      "balance_loss_clip": 1.04610157,
      "balance_loss_mlp": 1.02106488,
      "epoch": 0.4990079961522275,
      "flos": 16872700988160.0,
      "grad_norm": 1.640237289477909,
      "language_loss": 0.61582917,
      "learning_rate": 2.1035569003652156e-06,
      "loss": 0.63743401,
      "num_input_tokens_seen": 89433195,
      "step": 4150,
      "time_per_iteration": 2.656050205230713
    },
    {
      "auxiliary_loss_clip": 0.01123935,
      "auxiliary_loss_mlp": 0.01036152,
      "balance_loss_clip": 1.04483581,
      "balance_loss_mlp": 1.02687716,
      "epoch": 0.4991282390428666,
      "flos": 13291042187520.0,
      "grad_norm": 2.299880548381162,
      "language_loss": 0.81880075,
      "learning_rate": 2.1027789657982255e-06,
      "loss": 0.84040159,
      "num_input_tokens_seen": 89447410,
      "step": 4151,
      "time_per_iteration": 2.7204132080078125
    },
    {
      "auxiliary_loss_clip": 0.01125344,
      "auxiliary_loss_mlp": 0.01031419,
      "balance_loss_clip": 1.04604363,
      "balance_loss_mlp": 1.02355146,
      "epoch": 0.4992484819335057,
      "flos": 21537496454400.0,
      "grad_norm": 2.248182486268081,
      "language_loss": 0.77384055,
      "learning_rate": 2.1020010156397482e-06,
      "loss": 0.79540813,
      "num_input_tokens_seen": 89464630,
      "step": 4152,
      "time_per_iteration": 2.736994743347168
    },
    {
      "auxiliary_loss_clip": 0.01168637,
      "auxiliary_loss_mlp": 0.01029052,
      "balance_loss_clip": 1.048208,
      "balance_loss_mlp": 1.02098131,
      "epoch": 0.4993687248241448,
      "flos": 24860095390080.0,
      "grad_norm": 1.6643205762803908,
      "language_loss": 0.77704865,
      "learning_rate": 2.101223050007797e-06,
      "loss": 0.79902554,
      "num_input_tokens_seen": 89483180,
      "step": 4153,
      "time_per_iteration": 2.64463472366333
    },
    {
      "auxiliary_loss_clip": 0.01081314,
      "auxiliary_loss_mlp": 0.0100216,
      "balance_loss_clip": 1.01501977,
      "balance_loss_mlp": 1.00062203,
      "epoch": 0.49948896771478385,
      "flos": 62941602453120.0,
      "grad_norm": 0.8270829298885973,
      "language_loss": 0.5379945,
      "learning_rate": 2.1004450690203904e-06,
      "loss": 0.55882919,
      "num_input_tokens_seen": 89539260,
      "step": 4154,
      "time_per_iteration": 3.226736545562744
    },
    {
      "auxiliary_loss_clip": 0.01080853,
      "auxiliary_loss_mlp": 0.01001258,
      "balance_loss_clip": 1.0148468,
      "balance_loss_mlp": 0.99972051,
      "epoch": 0.49960921060542296,
      "flos": 68284213516800.0,
      "grad_norm": 0.8511115716072202,
      "language_loss": 0.63316166,
      "learning_rate": 2.099667072795546e-06,
      "loss": 0.65398276,
      "num_input_tokens_seen": 89601380,
      "step": 4155,
      "time_per_iteration": 3.2274184226989746
    },
    {
      "auxiliary_loss_clip": 0.01165985,
      "auxiliary_loss_mlp": 0.01023924,
      "balance_loss_clip": 1.04810202,
      "balance_loss_mlp": 1.01596081,
      "epoch": 0.49972945349606207,
      "flos": 23659350618240.0,
      "grad_norm": 1.808042519103185,
      "language_loss": 0.79895514,
      "learning_rate": 2.0988890614512864e-06,
      "loss": 0.82085419,
      "num_input_tokens_seen": 89621270,
      "step": 4156,
      "time_per_iteration": 2.6885881423950195
    },
    {
      "auxiliary_loss_clip": 0.01158748,
      "auxiliary_loss_mlp": 0.01025466,
      "balance_loss_clip": 1.05113173,
      "balance_loss_mlp": 1.01777685,
      "epoch": 0.4998496963867011,
      "flos": 19755825022080.0,
      "grad_norm": 2.461082191365096,
      "language_loss": 0.83966863,
      "learning_rate": 2.098111035105635e-06,
      "loss": 0.86151081,
      "num_input_tokens_seen": 89639695,
      "step": 4157,
      "time_per_iteration": 2.6376068592071533
    },
    {
      "auxiliary_loss_clip": 0.01126954,
      "auxiliary_loss_mlp": 0.0102244,
      "balance_loss_clip": 1.04748511,
      "balance_loss_mlp": 1.01472092,
      "epoch": 0.49996993927734024,
      "flos": 22265728790400.0,
      "grad_norm": 1.6619227598615467,
      "language_loss": 0.73223007,
      "learning_rate": 2.0973329938766176e-06,
      "loss": 0.75372398,
      "num_input_tokens_seen": 89657125,
      "step": 4158,
      "time_per_iteration": 2.7323129177093506
    },
    {
      "auxiliary_loss_clip": 0.011762,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.05144787,
      "balance_loss_mlp": 1.02282333,
      "epoch": 0.5000901821679793,
      "flos": 23327212533120.0,
      "grad_norm": 1.8494821861107642,
      "language_loss": 0.79150236,
      "learning_rate": 2.0965549378822618e-06,
      "loss": 0.81357932,
      "num_input_tokens_seen": 89678415,
      "step": 4159,
      "time_per_iteration": 2.686502456665039
    },
    {
      "auxiliary_loss_clip": 0.01082774,
      "auxiliary_loss_mlp": 0.01032881,
      "balance_loss_clip": 1.04126978,
      "balance_loss_mlp": 1.02386308,
      "epoch": 0.5002104250586185,
      "flos": 20339014239360.0,
      "grad_norm": 2.024192513036943,
      "language_loss": 0.84326291,
      "learning_rate": 2.095776867240599e-06,
      "loss": 0.86441946,
      "num_input_tokens_seen": 89695405,
      "step": 4160,
      "time_per_iteration": 2.904794454574585
    },
    {
      "auxiliary_loss_clip": 0.01133451,
      "auxiliary_loss_mlp": 0.0103203,
      "balance_loss_clip": 1.04552686,
      "balance_loss_mlp": 1.0243113,
      "epoch": 0.5003306679492575,
      "flos": 13991372634240.0,
      "grad_norm": 1.9084147902697803,
      "language_loss": 0.82609308,
      "learning_rate": 2.094998782069661e-06,
      "loss": 0.84774792,
      "num_input_tokens_seen": 89713110,
      "step": 4161,
      "time_per_iteration": 3.0508813858032227
    },
    {
      "auxiliary_loss_clip": 0.01181044,
      "auxiliary_loss_mlp": 0.01033108,
      "balance_loss_clip": 1.05087996,
      "balance_loss_mlp": 1.02613986,
      "epoch": 0.5004509108398966,
      "flos": 27672762896640.0,
      "grad_norm": 2.15606363449673,
      "language_loss": 0.75440538,
      "learning_rate": 2.0942206824874845e-06,
      "loss": 0.77654696,
      "num_input_tokens_seen": 89735885,
      "step": 4162,
      "time_per_iteration": 2.6663448810577393
    },
    {
      "auxiliary_loss_clip": 0.01165395,
      "auxiliary_loss_mlp": 0.01029669,
      "balance_loss_clip": 1.04874802,
      "balance_loss_mlp": 1.0213542,
      "epoch": 0.5005711537305357,
      "flos": 14976186796800.0,
      "grad_norm": 2.0761966375045864,
      "language_loss": 0.7883929,
      "learning_rate": 2.093442568612105e-06,
      "loss": 0.8103435,
      "num_input_tokens_seen": 89753690,
      "step": 4163,
      "time_per_iteration": 2.6545279026031494
    },
    {
      "auxiliary_loss_clip": 0.0118563,
      "auxiliary_loss_mlp": 0.01029581,
      "balance_loss_clip": 1.05276918,
      "balance_loss_mlp": 1.02123046,
      "epoch": 0.5006913966211748,
      "flos": 26503259978880.0,
      "grad_norm": 1.4500630218557733,
      "language_loss": 0.84948397,
      "learning_rate": 2.0926644405615613e-06,
      "loss": 0.87163609,
      "num_input_tokens_seen": 89774590,
      "step": 4164,
      "time_per_iteration": 2.6310176849365234
    },
    {
      "auxiliary_loss_clip": 0.01136994,
      "auxiliary_loss_mlp": 0.01027543,
      "balance_loss_clip": 1.04649711,
      "balance_loss_mlp": 1.02018154,
      "epoch": 0.5008116395118138,
      "flos": 20449295971200.0,
      "grad_norm": 1.7313599828817534,
      "language_loss": 0.81139815,
      "learning_rate": 2.091886298453897e-06,
      "loss": 0.83304358,
      "num_input_tokens_seen": 89792775,
      "step": 4165,
      "time_per_iteration": 2.779615640640259
    },
    {
      "auxiliary_loss_clip": 0.01164322,
      "auxiliary_loss_mlp": 0.01027495,
      "balance_loss_clip": 1.04729486,
      "balance_loss_mlp": 1.01953816,
      "epoch": 0.500931882402453,
      "flos": 21579871524480.0,
      "grad_norm": 1.9985382828567275,
      "language_loss": 0.72407418,
      "learning_rate": 2.091108142407153e-06,
      "loss": 0.7459923,
      "num_input_tokens_seen": 89811515,
      "step": 4166,
      "time_per_iteration": 2.6181044578552246
    },
    {
      "auxiliary_loss_clip": 0.01065218,
      "auxiliary_loss_mlp": 0.01002943,
      "balance_loss_clip": 1.0224936,
      "balance_loss_mlp": 1.00130963,
      "epoch": 0.5010521252930921,
      "flos": 57785011925760.0,
      "grad_norm": 0.8416519672079961,
      "language_loss": 0.62400335,
      "learning_rate": 2.090329972539377e-06,
      "loss": 0.64468497,
      "num_input_tokens_seen": 89870080,
      "step": 4167,
      "time_per_iteration": 3.2728874683380127
    },
    {
      "auxiliary_loss_clip": 0.01073619,
      "auxiliary_loss_mlp": 0.01025154,
      "balance_loss_clip": 1.04024673,
      "balance_loss_mlp": 1.01720858,
      "epoch": 0.5011723681837311,
      "flos": 18625500864000.0,
      "grad_norm": 1.9485641665977913,
      "language_loss": 0.68765759,
      "learning_rate": 2.089551788968616e-06,
      "loss": 0.70864522,
      "num_input_tokens_seen": 89888045,
      "step": 4168,
      "time_per_iteration": 3.0246427059173584
    },
    {
      "auxiliary_loss_clip": 0.01079529,
      "auxiliary_loss_mlp": 0.0100193,
      "balance_loss_clip": 1.0137856,
      "balance_loss_mlp": 1.00033307,
      "epoch": 0.5012926110743702,
      "flos": 55883146608000.0,
      "grad_norm": 0.8401174593415519,
      "language_loss": 0.61073768,
      "learning_rate": 2.08877359181292e-06,
      "loss": 0.63155234,
      "num_input_tokens_seen": 89944610,
      "step": 4169,
      "time_per_iteration": 3.427090883255005
    },
    {
      "auxiliary_loss_clip": 0.01144372,
      "auxiliary_loss_mlp": 0.01028333,
      "balance_loss_clip": 1.04359949,
      "balance_loss_mlp": 1.02017343,
      "epoch": 0.5014128539650093,
      "flos": 24238266117120.0,
      "grad_norm": 2.85410478072569,
      "language_loss": 0.85753995,
      "learning_rate": 2.0879953811903396e-06,
      "loss": 0.87926698,
      "num_input_tokens_seen": 89959495,
      "step": 4170,
      "time_per_iteration": 2.7821552753448486
    },
    {
      "auxiliary_loss_clip": 0.01167086,
      "auxiliary_loss_mlp": 0.01027398,
      "balance_loss_clip": 1.05057847,
      "balance_loss_mlp": 1.01938105,
      "epoch": 0.5015330968556484,
      "flos": 27527468382720.0,
      "grad_norm": 2.0082013332474484,
      "language_loss": 0.78652883,
      "learning_rate": 2.08721715721893e-06,
      "loss": 0.80847359,
      "num_input_tokens_seen": 89978820,
      "step": 4171,
      "time_per_iteration": 4.076817989349365
    },
    {
      "auxiliary_loss_clip": 0.01164069,
      "auxiliary_loss_mlp": 0.01024575,
      "balance_loss_clip": 1.04790854,
      "balance_loss_mlp": 1.01689816,
      "epoch": 0.5016533397462875,
      "flos": 23800802376960.0,
      "grad_norm": 1.8411182860186295,
      "language_loss": 0.77069038,
      "learning_rate": 2.0864389200167477e-06,
      "loss": 0.79257679,
      "num_input_tokens_seen": 89997075,
      "step": 4172,
      "time_per_iteration": 3.622549295425415
    },
    {
      "auxiliary_loss_clip": 0.01171618,
      "auxiliary_loss_mlp": 0.00762622,
      "balance_loss_clip": 1.05156934,
      "balance_loss_mlp": 1.00029182,
      "epoch": 0.5017735826369266,
      "flos": 25295009264640.0,
      "grad_norm": 1.7759079661929333,
      "language_loss": 0.79145783,
      "learning_rate": 2.0856606697018504e-06,
      "loss": 0.81080019,
      "num_input_tokens_seen": 90015085,
      "step": 4173,
      "time_per_iteration": 2.875133514404297
    },
    {
      "auxiliary_loss_clip": 0.0114906,
      "auxiliary_loss_mlp": 0.01022289,
      "balance_loss_clip": 1.0460695,
      "balance_loss_mlp": 1.01399755,
      "epoch": 0.5018938255275657,
      "flos": 16873203778560.0,
      "grad_norm": 2.0826578663452366,
      "language_loss": 0.73306286,
      "learning_rate": 2.084882406392297e-06,
      "loss": 0.75477636,
      "num_input_tokens_seen": 90033045,
      "step": 4174,
      "time_per_iteration": 3.65132212638855
    },
    {
      "auxiliary_loss_clip": 0.01169301,
      "auxiliary_loss_mlp": 0.01020309,
      "balance_loss_clip": 1.0512166,
      "balance_loss_mlp": 1.01286471,
      "epoch": 0.5020140684182047,
      "flos": 25515429073920.0,
      "grad_norm": 5.101294750756256,
      "language_loss": 0.71419841,
      "learning_rate": 2.0841041302061496e-06,
      "loss": 0.73609447,
      "num_input_tokens_seen": 90052505,
      "step": 4175,
      "time_per_iteration": 2.6758530139923096
    },
    {
      "auxiliary_loss_clip": 0.01142854,
      "auxiliary_loss_mlp": 0.0102622,
      "balance_loss_clip": 1.04417467,
      "balance_loss_mlp": 1.01864457,
      "epoch": 0.5021343113088439,
      "flos": 23659278791040.0,
      "grad_norm": 2.1720853999616208,
      "language_loss": 0.75886285,
      "learning_rate": 2.083325841261473e-06,
      "loss": 0.78055364,
      "num_input_tokens_seen": 90071565,
      "step": 4176,
      "time_per_iteration": 3.606649398803711
    },
    {
      "auxiliary_loss_clip": 0.01148221,
      "auxiliary_loss_mlp": 0.01027483,
      "balance_loss_clip": 1.04543269,
      "balance_loss_mlp": 1.01915646,
      "epoch": 0.502254554199483,
      "flos": 24534673148160.0,
      "grad_norm": 2.115151919462895,
      "language_loss": 0.66928148,
      "learning_rate": 2.0825475396763322e-06,
      "loss": 0.69103849,
      "num_input_tokens_seen": 90092215,
      "step": 4177,
      "time_per_iteration": 2.692342758178711
    },
    {
      "auxiliary_loss_clip": 0.01076853,
      "auxiliary_loss_mlp": 0.0102899,
      "balance_loss_clip": 1.03926444,
      "balance_loss_mlp": 1.02063918,
      "epoch": 0.502374797090122,
      "flos": 34240285607040.0,
      "grad_norm": 1.5553888767649453,
      "language_loss": 0.65656286,
      "learning_rate": 2.081769225568796e-06,
      "loss": 0.67762125,
      "num_input_tokens_seen": 90114665,
      "step": 4178,
      "time_per_iteration": 3.0914196968078613
    },
    {
      "auxiliary_loss_clip": 0.01168602,
      "auxiliary_loss_mlp": 0.01030374,
      "balance_loss_clip": 1.04770446,
      "balance_loss_mlp": 1.02282214,
      "epoch": 0.5024950399807612,
      "flos": 26031106679040.0,
      "grad_norm": 1.611324435934562,
      "language_loss": 0.76205003,
      "learning_rate": 2.0809908990569327e-06,
      "loss": 0.78403974,
      "num_input_tokens_seen": 90136445,
      "step": 4179,
      "time_per_iteration": 3.021836757659912
    },
    {
      "auxiliary_loss_clip": 0.01150504,
      "auxiliary_loss_mlp": 0.01032866,
      "balance_loss_clip": 1.04673278,
      "balance_loss_mlp": 1.02533746,
      "epoch": 0.5026152828714002,
      "flos": 21252438120960.0,
      "grad_norm": 1.815330536111044,
      "language_loss": 0.79334056,
      "learning_rate": 2.0802125602588146e-06,
      "loss": 0.81517428,
      "num_input_tokens_seen": 90155710,
      "step": 4180,
      "time_per_iteration": 2.713195323944092
    },
    {
      "auxiliary_loss_clip": 0.01186318,
      "auxiliary_loss_mlp": 0.01024981,
      "balance_loss_clip": 1.05366898,
      "balance_loss_mlp": 1.01676798,
      "epoch": 0.5027355257620393,
      "flos": 30956111245440.0,
      "grad_norm": 3.4864962273628533,
      "language_loss": 0.66793501,
      "learning_rate": 2.0794342092925146e-06,
      "loss": 0.69004798,
      "num_input_tokens_seen": 90176845,
      "step": 4181,
      "time_per_iteration": 2.679734468460083
    },
    {
      "auxiliary_loss_clip": 0.01173341,
      "auxiliary_loss_mlp": 0.01030567,
      "balance_loss_clip": 1.05343735,
      "balance_loss_mlp": 1.02296185,
      "epoch": 0.5028557686526784,
      "flos": 24791147233920.0,
      "grad_norm": 1.9107429587290148,
      "language_loss": 0.68004799,
      "learning_rate": 2.078655846276108e-06,
      "loss": 0.70208704,
      "num_input_tokens_seen": 90197175,
      "step": 4182,
      "time_per_iteration": 2.657201051712036
    },
    {
      "auxiliary_loss_clip": 0.01149015,
      "auxiliary_loss_mlp": 0.01031978,
      "balance_loss_clip": 1.04682672,
      "balance_loss_mlp": 1.02359772,
      "epoch": 0.5029760115433175,
      "flos": 22966992990720.0,
      "grad_norm": 1.9810689739267082,
      "language_loss": 0.68829399,
      "learning_rate": 2.0778774713276727e-06,
      "loss": 0.71010393,
      "num_input_tokens_seen": 90216650,
      "step": 4183,
      "time_per_iteration": 2.720242738723755
    },
    {
      "auxiliary_loss_clip": 0.01165587,
      "auxiliary_loss_mlp": 0.01027417,
      "balance_loss_clip": 1.04706454,
      "balance_loss_mlp": 1.01940656,
      "epoch": 0.5030962544339566,
      "flos": 15305164485120.0,
      "grad_norm": 2.371071605341998,
      "language_loss": 0.67727637,
      "learning_rate": 2.077099084565287e-06,
      "loss": 0.69920641,
      "num_input_tokens_seen": 90234055,
      "step": 4184,
      "time_per_iteration": 2.599565267562866
    },
    {
      "auxiliary_loss_clip": 0.01147777,
      "auxiliary_loss_mlp": 0.01026985,
      "balance_loss_clip": 1.0446291,
      "balance_loss_mlp": 1.01849139,
      "epoch": 0.5032164973245957,
      "flos": 24494847943680.0,
      "grad_norm": 2.2661487330041608,
      "language_loss": 0.65069807,
      "learning_rate": 2.0763206861070313e-06,
      "loss": 0.67244571,
      "num_input_tokens_seen": 90253115,
      "step": 4185,
      "time_per_iteration": 2.7256710529327393
    },
    {
      "auxiliary_loss_clip": 0.01182704,
      "auxiliary_loss_mlp": 0.01029488,
      "balance_loss_clip": 1.05009592,
      "balance_loss_mlp": 1.02134025,
      "epoch": 0.5033367402152348,
      "flos": 16213452721920.0,
      "grad_norm": 1.9891549725504352,
      "language_loss": 0.75055361,
      "learning_rate": 2.0755422760709876e-06,
      "loss": 0.77267551,
      "num_input_tokens_seen": 90270515,
      "step": 4186,
      "time_per_iteration": 2.612501621246338
    },
    {
      "auxiliary_loss_clip": 0.01118138,
      "auxiliary_loss_mlp": 0.01023651,
      "balance_loss_clip": 1.04280448,
      "balance_loss_mlp": 1.01616502,
      "epoch": 0.5034569831058738,
      "flos": 21391375927680.0,
      "grad_norm": 1.8953053516344838,
      "language_loss": 0.76963735,
      "learning_rate": 2.0747638545752417e-06,
      "loss": 0.7910552,
      "num_input_tokens_seen": 90289075,
      "step": 4187,
      "time_per_iteration": 2.7648212909698486
    },
    {
      "auxiliary_loss_clip": 0.01150832,
      "auxiliary_loss_mlp": 0.01031836,
      "balance_loss_clip": 1.04811001,
      "balance_loss_mlp": 1.02381873,
      "epoch": 0.503577225996513,
      "flos": 20558751690240.0,
      "grad_norm": 2.2048373721697563,
      "language_loss": 0.83291066,
      "learning_rate": 2.073985421737878e-06,
      "loss": 0.85473734,
      "num_input_tokens_seen": 90306385,
      "step": 4188,
      "time_per_iteration": 2.725551128387451
    },
    {
      "auxiliary_loss_clip": 0.01172653,
      "auxiliary_loss_mlp": 0.01032955,
      "balance_loss_clip": 1.05143619,
      "balance_loss_mlp": 1.02441919,
      "epoch": 0.5036974688871521,
      "flos": 27229157930880.0,
      "grad_norm": 2.762331733466256,
      "language_loss": 0.74345297,
      "learning_rate": 2.0732069776769844e-06,
      "loss": 0.76550901,
      "num_input_tokens_seen": 90323795,
      "step": 4189,
      "time_per_iteration": 2.6788506507873535
    },
    {
      "auxiliary_loss_clip": 0.01185884,
      "auxiliary_loss_mlp": 0.01027899,
      "balance_loss_clip": 1.05404139,
      "balance_loss_mlp": 1.01951265,
      "epoch": 0.5038177117777911,
      "flos": 20412164286720.0,
      "grad_norm": 1.9776931605191572,
      "language_loss": 0.73219907,
      "learning_rate": 2.072428522510651e-06,
      "loss": 0.75433683,
      "num_input_tokens_seen": 90340360,
      "step": 4190,
      "time_per_iteration": 2.6441097259521484
    },
    {
      "auxiliary_loss_clip": 0.01137838,
      "auxiliary_loss_mlp": 0.01031637,
      "balance_loss_clip": 1.04736614,
      "balance_loss_mlp": 1.02329206,
      "epoch": 0.5039379546684303,
      "flos": 21907987286400.0,
      "grad_norm": 2.3819020572562697,
      "language_loss": 0.76073796,
      "learning_rate": 2.071650056356968e-06,
      "loss": 0.78243268,
      "num_input_tokens_seen": 90357900,
      "step": 4191,
      "time_per_iteration": 2.7456560134887695
    },
    {
      "auxiliary_loss_clip": 0.01183538,
      "auxiliary_loss_mlp": 0.01036449,
      "balance_loss_clip": 1.05217659,
      "balance_loss_mlp": 1.02814054,
      "epoch": 0.5040581975590693,
      "flos": 20010718909440.0,
      "grad_norm": 2.1482425177727915,
      "language_loss": 0.80078697,
      "learning_rate": 2.070871579334028e-06,
      "loss": 0.82298684,
      "num_input_tokens_seen": 90377010,
      "step": 4192,
      "time_per_iteration": 2.71087646484375
    },
    {
      "auxiliary_loss_clip": 0.01178815,
      "auxiliary_loss_mlp": 0.0103025,
      "balance_loss_clip": 1.04809928,
      "balance_loss_mlp": 1.0224061,
      "epoch": 0.5041784404497084,
      "flos": 20959837931520.0,
      "grad_norm": 3.0613769990018804,
      "language_loss": 0.71634221,
      "learning_rate": 2.0700930915599264e-06,
      "loss": 0.73843282,
      "num_input_tokens_seen": 90396740,
      "step": 4193,
      "time_per_iteration": 2.5917296409606934
    },
    {
      "auxiliary_loss_clip": 0.01181603,
      "auxiliary_loss_mlp": 0.01027026,
      "balance_loss_clip": 1.05084777,
      "balance_loss_mlp": 1.01962352,
      "epoch": 0.5042986833403476,
      "flos": 12495082757760.0,
      "grad_norm": 2.087851619360629,
      "language_loss": 0.78381133,
      "learning_rate": 2.0693145931527583e-06,
      "loss": 0.80589765,
      "num_input_tokens_seen": 90413220,
      "step": 4194,
      "time_per_iteration": 2.7896101474761963
    },
    {
      "auxiliary_loss_clip": 0.01150635,
      "auxiliary_loss_mlp": 0.01028189,
      "balance_loss_clip": 1.04835522,
      "balance_loss_mlp": 1.0203923,
      "epoch": 0.5044189262309866,
      "flos": 29202305788800.0,
      "grad_norm": 1.5224714572593459,
      "language_loss": 0.77984488,
      "learning_rate": 2.068536084230622e-06,
      "loss": 0.80163312,
      "num_input_tokens_seen": 90435085,
      "step": 4195,
      "time_per_iteration": 2.7925758361816406
    },
    {
      "auxiliary_loss_clip": 0.01169824,
      "auxiliary_loss_mlp": 0.01028026,
      "balance_loss_clip": 1.05130756,
      "balance_loss_mlp": 1.01928234,
      "epoch": 0.5045391691216257,
      "flos": 23873198238720.0,
      "grad_norm": 2.365633982538467,
      "language_loss": 0.89017838,
      "learning_rate": 2.067757564911616e-06,
      "loss": 0.91215694,
      "num_input_tokens_seen": 90453660,
      "step": 4196,
      "time_per_iteration": 2.708341360092163
    },
    {
      "auxiliary_loss_clip": 0.01163782,
      "auxiliary_loss_mlp": 0.00763246,
      "balance_loss_clip": 1.04948032,
      "balance_loss_mlp": 1.00026011,
      "epoch": 0.5046594120122648,
      "flos": 24644990793600.0,
      "grad_norm": 1.9052011721899877,
      "language_loss": 0.92615998,
      "learning_rate": 2.0669790353138407e-06,
      "loss": 0.94543028,
      "num_input_tokens_seen": 90472625,
      "step": 4197,
      "time_per_iteration": 3.6740736961364746
    },
    {
      "auxiliary_loss_clip": 0.01135106,
      "auxiliary_loss_mlp": 0.0076308,
      "balance_loss_clip": 1.04639912,
      "balance_loss_mlp": 1.00026917,
      "epoch": 0.5047796549029039,
      "flos": 23362835846400.0,
      "grad_norm": 2.3464352578273964,
      "language_loss": 0.73345006,
      "learning_rate": 2.0662004955553995e-06,
      "loss": 0.75243187,
      "num_input_tokens_seen": 90492325,
      "step": 4198,
      "time_per_iteration": 3.6061909198760986
    },
    {
      "auxiliary_loss_clip": 0.01151051,
      "auxiliary_loss_mlp": 0.01026173,
      "balance_loss_clip": 1.04735279,
      "balance_loss_mlp": 1.01865053,
      "epoch": 0.5048998977935429,
      "flos": 17304095329920.0,
      "grad_norm": 1.9213310851157395,
      "language_loss": 0.76631427,
      "learning_rate": 2.065421945754395e-06,
      "loss": 0.78808659,
      "num_input_tokens_seen": 90510055,
      "step": 4199,
      "time_per_iteration": 3.6012279987335205
    },
    {
      "auxiliary_loss_clip": 0.01128594,
      "auxiliary_loss_mlp": 0.01029984,
      "balance_loss_clip": 1.04696274,
      "balance_loss_mlp": 1.0226171,
      "epoch": 0.505020140684182,
      "flos": 34856979235200.0,
      "grad_norm": 1.7394713021140098,
      "language_loss": 0.78091085,
      "learning_rate": 2.0646433860289344e-06,
      "loss": 0.80249667,
      "num_input_tokens_seen": 90528980,
      "step": 4200,
      "time_per_iteration": 2.8401331901550293
    },
    {
      "auxiliary_loss_clip": 0.01173978,
      "auxiliary_loss_mlp": 0.00763338,
      "balance_loss_clip": 1.05033624,
      "balance_loss_mlp": 1.00025833,
      "epoch": 0.5051403835748212,
      "flos": 24863974058880.0,
      "grad_norm": 2.048517698415416,
      "language_loss": 0.82808638,
      "learning_rate": 2.0638648164971233e-06,
      "loss": 0.84745955,
      "num_input_tokens_seen": 90547445,
      "step": 4201,
      "time_per_iteration": 2.6215784549713135
    },
    {
      "auxiliary_loss_clip": 0.01153025,
      "auxiliary_loss_mlp": 0.01026425,
      "balance_loss_clip": 1.04839611,
      "balance_loss_mlp": 1.01908755,
      "epoch": 0.5052606264654602,
      "flos": 20959694277120.0,
      "grad_norm": 1.7309860575282148,
      "language_loss": 0.88424087,
      "learning_rate": 2.06308623727707e-06,
      "loss": 0.9060353,
      "num_input_tokens_seen": 90567545,
      "step": 4202,
      "time_per_iteration": 3.6678428649902344
    },
    {
      "auxiliary_loss_clip": 0.01167767,
      "auxiliary_loss_mlp": 0.01033012,
      "balance_loss_clip": 1.0504086,
      "balance_loss_mlp": 1.02381456,
      "epoch": 0.5053808693560993,
      "flos": 19642382893440.0,
      "grad_norm": 2.225811532672944,
      "language_loss": 0.76795447,
      "learning_rate": 2.0623076484868846e-06,
      "loss": 0.78996217,
      "num_input_tokens_seen": 90585000,
      "step": 4203,
      "time_per_iteration": 2.6137685775756836
    },
    {
      "auxiliary_loss_clip": 0.01051679,
      "auxiliary_loss_mlp": 0.01006888,
      "balance_loss_clip": 1.01127982,
      "balance_loss_mlp": 1.00532675,
      "epoch": 0.5055011122467384,
      "flos": 67504915019520.0,
      "grad_norm": 0.8311054218088607,
      "language_loss": 0.60642612,
      "learning_rate": 2.061529050244679e-06,
      "loss": 0.62701178,
      "num_input_tokens_seen": 90644745,
      "step": 4204,
      "time_per_iteration": 3.2554993629455566
    },
    {
      "auxiliary_loss_clip": 0.01147882,
      "auxiliary_loss_mlp": 0.01024327,
      "balance_loss_clip": 1.04707432,
      "balance_loss_mlp": 1.01631284,
      "epoch": 0.5056213551373775,
      "flos": 16872952383360.0,
      "grad_norm": 2.721617058483369,
      "language_loss": 0.7415899,
      "learning_rate": 2.060750442668565e-06,
      "loss": 0.76331198,
      "num_input_tokens_seen": 90662500,
      "step": 4205,
      "time_per_iteration": 2.721179246902466
    },
    {
      "auxiliary_loss_clip": 0.01170658,
      "auxiliary_loss_mlp": 0.01026199,
      "balance_loss_clip": 1.05385756,
      "balance_loss_mlp": 1.01824164,
      "epoch": 0.5057415980280165,
      "flos": 15334179696000.0,
      "grad_norm": 2.684867526835601,
      "language_loss": 0.63983607,
      "learning_rate": 2.059971825876657e-06,
      "loss": 0.66180456,
      "num_input_tokens_seen": 90677010,
      "step": 4206,
      "time_per_iteration": 2.6170902252197266
    },
    {
      "auxiliary_loss_clip": 0.0117163,
      "auxiliary_loss_mlp": 0.01024158,
      "balance_loss_clip": 1.05098903,
      "balance_loss_mlp": 1.01627207,
      "epoch": 0.5058618409186557,
      "flos": 19025976574080.0,
      "grad_norm": 2.0819972655753363,
      "language_loss": 0.76738554,
      "learning_rate": 2.0591931999870713e-06,
      "loss": 0.78934336,
      "num_input_tokens_seen": 90695935,
      "step": 4207,
      "time_per_iteration": 2.7252726554870605
    },
    {
      "auxiliary_loss_clip": 0.01061513,
      "auxiliary_loss_mlp": 0.01003424,
      "balance_loss_clip": 1.0118432,
      "balance_loss_mlp": 1.00181425,
      "epoch": 0.5059820838092948,
      "flos": 63453114080640.0,
      "grad_norm": 0.8291613101492425,
      "language_loss": 0.57600707,
      "learning_rate": 2.0584145651179234e-06,
      "loss": 0.59665644,
      "num_input_tokens_seen": 90751645,
      "step": 4208,
      "time_per_iteration": 3.241603374481201
    },
    {
      "auxiliary_loss_clip": 0.01153731,
      "auxiliary_loss_mlp": 0.00762731,
      "balance_loss_clip": 1.05026317,
      "balance_loss_mlp": 1.00032687,
      "epoch": 0.5061023266999338,
      "flos": 15441803821440.0,
      "grad_norm": 4.767999856636263,
      "language_loss": 0.80413616,
      "learning_rate": 2.0576359213873327e-06,
      "loss": 0.82330084,
      "num_input_tokens_seen": 90766795,
      "step": 4209,
      "time_per_iteration": 2.679694414138794
    },
    {
      "auxiliary_loss_clip": 0.01161757,
      "auxiliary_loss_mlp": 0.01026563,
      "balance_loss_clip": 1.04710221,
      "balance_loss_mlp": 1.01865911,
      "epoch": 0.506222569590573,
      "flos": 22451063990400.0,
      "grad_norm": 3.423705379713862,
      "language_loss": 0.70162678,
      "learning_rate": 2.056857268913419e-06,
      "loss": 0.72350997,
      "num_input_tokens_seen": 90786845,
      "step": 4210,
      "time_per_iteration": 2.680133581161499
    },
    {
      "auxiliary_loss_clip": 0.0117011,
      "auxiliary_loss_mlp": 0.01031419,
      "balance_loss_clip": 1.05205917,
      "balance_loss_mlp": 1.02315199,
      "epoch": 0.506342812481212,
      "flos": 17558665994880.0,
      "grad_norm": 2.4299161285522564,
      "language_loss": 0.84146369,
      "learning_rate": 2.056078607814303e-06,
      "loss": 0.8634789,
      "num_input_tokens_seen": 90802630,
      "step": 4211,
      "time_per_iteration": 2.6550557613372803
    },
    {
      "auxiliary_loss_clip": 0.01168368,
      "auxiliary_loss_mlp": 0.01033144,
      "balance_loss_clip": 1.05073667,
      "balance_loss_mlp": 1.02514493,
      "epoch": 0.5064630553718511,
      "flos": 23402050519680.0,
      "grad_norm": 2.117006350919607,
      "language_loss": 0.78300989,
      "learning_rate": 2.055299938208106e-06,
      "loss": 0.80502498,
      "num_input_tokens_seen": 90823620,
      "step": 4212,
      "time_per_iteration": 2.636911392211914
    },
    {
      "auxiliary_loss_clip": 0.01173655,
      "auxiliary_loss_mlp": 0.01032618,
      "balance_loss_clip": 1.05243528,
      "balance_loss_mlp": 1.02481592,
      "epoch": 0.5065832982624903,
      "flos": 23987035416960.0,
      "grad_norm": 1.6155203275529053,
      "language_loss": 0.86070669,
      "learning_rate": 2.0545212602129526e-06,
      "loss": 0.88276941,
      "num_input_tokens_seen": 90843475,
      "step": 4213,
      "time_per_iteration": 2.6521801948547363
    },
    {
      "auxiliary_loss_clip": 0.01149627,
      "auxiliary_loss_mlp": 0.01027832,
      "balance_loss_clip": 1.04679322,
      "balance_loss_mlp": 1.01963925,
      "epoch": 0.5067035411531293,
      "flos": 21503058289920.0,
      "grad_norm": 4.663001567306374,
      "language_loss": 0.66551018,
      "learning_rate": 2.0537425739469673e-06,
      "loss": 0.68728483,
      "num_input_tokens_seen": 90862410,
      "step": 4214,
      "time_per_iteration": 2.7132513523101807
    },
    {
      "auxiliary_loss_clip": 0.01068057,
      "auxiliary_loss_mlp": 0.01002848,
      "balance_loss_clip": 1.01238632,
      "balance_loss_mlp": 1.00128675,
      "epoch": 0.5068237840437684,
      "flos": 65934397687680.0,
      "grad_norm": 0.834543569772369,
      "language_loss": 0.59419799,
      "learning_rate": 2.052963879528276e-06,
      "loss": 0.61490703,
      "num_input_tokens_seen": 90922280,
      "step": 4215,
      "time_per_iteration": 3.1462759971618652
    },
    {
      "auxiliary_loss_clip": 0.01169314,
      "auxiliary_loss_mlp": 0.01029509,
      "balance_loss_clip": 1.05098259,
      "balance_loss_mlp": 1.02109861,
      "epoch": 0.5069440269344075,
      "flos": 27264206626560.0,
      "grad_norm": 2.096957256263172,
      "language_loss": 0.764,
      "learning_rate": 2.052185177075007e-06,
      "loss": 0.78598821,
      "num_input_tokens_seen": 90941850,
      "step": 4216,
      "time_per_iteration": 2.6854097843170166
    },
    {
      "auxiliary_loss_clip": 0.01170837,
      "auxiliary_loss_mlp": 0.01031429,
      "balance_loss_clip": 1.04981625,
      "balance_loss_mlp": 1.02272081,
      "epoch": 0.5070642698250466,
      "flos": 23366319465600.0,
      "grad_norm": 2.0454657097439877,
      "language_loss": 0.8280288,
      "learning_rate": 2.051406466705288e-06,
      "loss": 0.85005146,
      "num_input_tokens_seen": 90961390,
      "step": 4217,
      "time_per_iteration": 2.6286449432373047
    },
    {
      "auxiliary_loss_clip": 0.01180719,
      "auxiliary_loss_mlp": 0.01020828,
      "balance_loss_clip": 1.04968929,
      "balance_loss_mlp": 1.01329958,
      "epoch": 0.5071845127156857,
      "flos": 20340127560960.0,
      "grad_norm": 2.270308472242166,
      "language_loss": 0.81226552,
      "learning_rate": 2.0506277485372486e-06,
      "loss": 0.83428103,
      "num_input_tokens_seen": 90980215,
      "step": 4218,
      "time_per_iteration": 2.5990729331970215
    },
    {
      "auxiliary_loss_clip": 0.01161992,
      "auxiliary_loss_mlp": 0.01025293,
      "balance_loss_clip": 1.04863811,
      "balance_loss_mlp": 1.01716316,
      "epoch": 0.5073047556063248,
      "flos": 12092955022080.0,
      "grad_norm": 2.949332957722248,
      "language_loss": 0.66787112,
      "learning_rate": 2.04984902268902e-06,
      "loss": 0.689744,
      "num_input_tokens_seen": 90997415,
      "step": 4219,
      "time_per_iteration": 2.6079375743865967
    },
    {
      "auxiliary_loss_clip": 0.01175442,
      "auxiliary_loss_mlp": 0.01035279,
      "balance_loss_clip": 1.04986155,
      "balance_loss_mlp": 1.02634978,
      "epoch": 0.5074249984969639,
      "flos": 19682854542720.0,
      "grad_norm": 2.099156481898898,
      "language_loss": 0.75360185,
      "learning_rate": 2.0490702892787345e-06,
      "loss": 0.77570903,
      "num_input_tokens_seen": 91016475,
      "step": 4220,
      "time_per_iteration": 2.670884132385254
    },
    {
      "auxiliary_loss_clip": 0.01158816,
      "auxiliary_loss_mlp": 0.01029584,
      "balance_loss_clip": 1.04590034,
      "balance_loss_mlp": 1.02175808,
      "epoch": 0.5075452413876029,
      "flos": 28765703975040.0,
      "grad_norm": 1.5264643261504691,
      "language_loss": 0.62268585,
      "learning_rate": 2.0482915484245246e-06,
      "loss": 0.64456987,
      "num_input_tokens_seen": 91038095,
      "step": 4221,
      "time_per_iteration": 2.7097678184509277
    },
    {
      "auxiliary_loss_clip": 0.01121815,
      "auxiliary_loss_mlp": 0.0102915,
      "balance_loss_clip": 1.04657698,
      "balance_loss_mlp": 1.02019191,
      "epoch": 0.5076654842782421,
      "flos": 20339445202560.0,
      "grad_norm": 2.5321150910992993,
      "language_loss": 0.84112084,
      "learning_rate": 2.047512800244526e-06,
      "loss": 0.86263049,
      "num_input_tokens_seen": 91053360,
      "step": 4222,
      "time_per_iteration": 2.7460930347442627
    },
    {
      "auxiliary_loss_clip": 0.01168015,
      "auxiliary_loss_mlp": 0.01023536,
      "balance_loss_clip": 1.04994285,
      "balance_loss_mlp": 1.01514339,
      "epoch": 0.5077857271688812,
      "flos": 26359653404160.0,
      "grad_norm": 1.9381016844735832,
      "language_loss": 0.79083014,
      "learning_rate": 2.046734044856873e-06,
      "loss": 0.81274569,
      "num_input_tokens_seen": 91072770,
      "step": 4223,
      "time_per_iteration": 3.682727813720703
    },
    {
      "auxiliary_loss_clip": 0.01168261,
      "auxiliary_loss_mlp": 0.01029357,
      "balance_loss_clip": 1.05041146,
      "balance_loss_mlp": 1.02162039,
      "epoch": 0.5079059700595202,
      "flos": 21798962530560.0,
      "grad_norm": 2.011576481806382,
      "language_loss": 0.81113017,
      "learning_rate": 2.045955282379702e-06,
      "loss": 0.83310634,
      "num_input_tokens_seen": 91091430,
      "step": 4224,
      "time_per_iteration": 3.633483409881592
    },
    {
      "auxiliary_loss_clip": 0.01165101,
      "auxiliary_loss_mlp": 0.01030546,
      "balance_loss_clip": 1.04671836,
      "balance_loss_mlp": 1.02188516,
      "epoch": 0.5080262129501594,
      "flos": 13187943175680.0,
      "grad_norm": 3.604550938465008,
      "language_loss": 0.75323957,
      "learning_rate": 2.045176512931152e-06,
      "loss": 0.77519602,
      "num_input_tokens_seen": 91106060,
      "step": 4225,
      "time_per_iteration": 3.5453405380249023
    },
    {
      "auxiliary_loss_clip": 0.01143471,
      "auxiliary_loss_mlp": 0.01029414,
      "balance_loss_clip": 1.0465169,
      "balance_loss_mlp": 1.02180862,
      "epoch": 0.5081464558407984,
      "flos": 25301473712640.0,
      "grad_norm": 2.9211346177553517,
      "language_loss": 0.76338828,
      "learning_rate": 2.0443977366293604e-06,
      "loss": 0.78511709,
      "num_input_tokens_seen": 91124100,
      "step": 4226,
      "time_per_iteration": 2.7462334632873535
    },
    {
      "auxiliary_loss_clip": 0.01111123,
      "auxiliary_loss_mlp": 0.01028576,
      "balance_loss_clip": 1.04251766,
      "balance_loss_mlp": 1.02032661,
      "epoch": 0.5082666987314375,
      "flos": 30951226995840.0,
      "grad_norm": 2.256581258433177,
      "language_loss": 0.77009809,
      "learning_rate": 2.043618953592468e-06,
      "loss": 0.79149508,
      "num_input_tokens_seen": 91146555,
      "step": 4227,
      "time_per_iteration": 3.7615737915039062
    },
    {
      "auxiliary_loss_clip": 0.01156124,
      "auxiliary_loss_mlp": 0.01030895,
      "balance_loss_clip": 1.05103981,
      "balance_loss_mlp": 1.02214539,
      "epoch": 0.5083869416220766,
      "flos": 19682495406720.0,
      "grad_norm": 1.507943147901373,
      "language_loss": 0.81143141,
      "learning_rate": 2.0428401639386144e-06,
      "loss": 0.8333016,
      "num_input_tokens_seen": 91167120,
      "step": 4228,
      "time_per_iteration": 2.707005262374878
    },
    {
      "auxiliary_loss_clip": 0.01049131,
      "auxiliary_loss_mlp": 0.01001144,
      "balance_loss_clip": 1.01066899,
      "balance_loss_mlp": 0.99968398,
      "epoch": 0.5085071845127157,
      "flos": 71817535589760.0,
      "grad_norm": 0.8221350246816006,
      "language_loss": 0.58067518,
      "learning_rate": 2.042061367785943e-06,
      "loss": 0.60117793,
      "num_input_tokens_seen": 91220260,
      "step": 4229,
      "time_per_iteration": 3.201850175857544
    },
    {
      "auxiliary_loss_clip": 0.01140675,
      "auxiliary_loss_mlp": 0.01030989,
      "balance_loss_clip": 1.04289937,
      "balance_loss_mlp": 1.02231622,
      "epoch": 0.5086274274033548,
      "flos": 35951608252800.0,
      "grad_norm": 2.2674704311301874,
      "language_loss": 0.75350654,
      "learning_rate": 2.041282565252594e-06,
      "loss": 0.7752232,
      "num_input_tokens_seen": 91240425,
      "step": 4230,
      "time_per_iteration": 2.810241222381592
    },
    {
      "auxiliary_loss_clip": 0.01139042,
      "auxiliary_loss_mlp": 0.01025061,
      "balance_loss_clip": 1.04504919,
      "balance_loss_mlp": 1.01728857,
      "epoch": 0.5087476702939938,
      "flos": 23513732881920.0,
      "grad_norm": 1.7878090994965394,
      "language_loss": 0.77107334,
      "learning_rate": 2.040503756456714e-06,
      "loss": 0.79271448,
      "num_input_tokens_seen": 91259635,
      "step": 4231,
      "time_per_iteration": 2.7487435340881348
    },
    {
      "auxiliary_loss_clip": 0.01161135,
      "auxiliary_loss_mlp": 0.01031866,
      "balance_loss_clip": 1.04656458,
      "balance_loss_mlp": 1.02264464,
      "epoch": 0.508867913184633,
      "flos": 15122091841920.0,
      "grad_norm": 2.0470382258209954,
      "language_loss": 0.79175389,
      "learning_rate": 2.0397249415164456e-06,
      "loss": 0.81368387,
      "num_input_tokens_seen": 91276990,
      "step": 4232,
      "time_per_iteration": 2.602362632751465
    },
    {
      "auxiliary_loss_clip": 0.01143987,
      "auxiliary_loss_mlp": 0.01028278,
      "balance_loss_clip": 1.04245234,
      "balance_loss_mlp": 1.01982641,
      "epoch": 0.508988156075272,
      "flos": 25885309374720.0,
      "grad_norm": 1.7136886119577657,
      "language_loss": 0.80243319,
      "learning_rate": 2.0389461205499354e-06,
      "loss": 0.82415581,
      "num_input_tokens_seen": 91296125,
      "step": 4233,
      "time_per_iteration": 2.7469444274902344
    },
    {
      "auxiliary_loss_clip": 0.01143019,
      "auxiliary_loss_mlp": 0.01024384,
      "balance_loss_clip": 1.04567671,
      "balance_loss_mlp": 1.0165695,
      "epoch": 0.5091083989659111,
      "flos": 13844857057920.0,
      "grad_norm": 1.7901303361785612,
      "language_loss": 0.73562646,
      "learning_rate": 2.03816729367533e-06,
      "loss": 0.7573005,
      "num_input_tokens_seen": 91314280,
      "step": 4234,
      "time_per_iteration": 2.643343210220337
    },
    {
      "auxiliary_loss_clip": 0.0115789,
      "auxiliary_loss_mlp": 0.01028149,
      "balance_loss_clip": 1.04926968,
      "balance_loss_mlp": 1.02023327,
      "epoch": 0.5092286418565503,
      "flos": 21104881050240.0,
      "grad_norm": 3.8234780064007907,
      "language_loss": 0.71383798,
      "learning_rate": 2.0373884610107765e-06,
      "loss": 0.73569834,
      "num_input_tokens_seen": 91334595,
      "step": 4235,
      "time_per_iteration": 2.7530951499938965
    },
    {
      "auxiliary_loss_clip": 0.01171049,
      "auxiliary_loss_mlp": 0.01025056,
      "balance_loss_clip": 1.04745901,
      "balance_loss_mlp": 1.01751637,
      "epoch": 0.5093488847471893,
      "flos": 18621298972800.0,
      "grad_norm": 2.827427389745533,
      "language_loss": 0.69465226,
      "learning_rate": 2.0366096226744225e-06,
      "loss": 0.71661329,
      "num_input_tokens_seen": 91349790,
      "step": 4236,
      "time_per_iteration": 2.5716726779937744
    },
    {
      "auxiliary_loss_clip": 0.01159533,
      "auxiliary_loss_mlp": 0.01030529,
      "balance_loss_clip": 1.04710948,
      "balance_loss_mlp": 1.02255964,
      "epoch": 0.5094691276378284,
      "flos": 23803783205760.0,
      "grad_norm": 1.884613245947169,
      "language_loss": 0.76803505,
      "learning_rate": 2.035830778784418e-06,
      "loss": 0.78993565,
      "num_input_tokens_seen": 91370465,
      "step": 4237,
      "time_per_iteration": 2.666252374649048
    },
    {
      "auxiliary_loss_clip": 0.01159611,
      "auxiliary_loss_mlp": 0.01033369,
      "balance_loss_clip": 1.05236673,
      "balance_loss_mlp": 1.02573705,
      "epoch": 0.5095893705284675,
      "flos": 17420410546560.0,
      "grad_norm": 3.4601602087226313,
      "language_loss": 0.79902422,
      "learning_rate": 2.0350519294589134e-06,
      "loss": 0.82095408,
      "num_input_tokens_seen": 91388505,
      "step": 4238,
      "time_per_iteration": 2.6558332443237305
    },
    {
      "auxiliary_loss_clip": 0.0112326,
      "auxiliary_loss_mlp": 0.01030939,
      "balance_loss_clip": 1.04187131,
      "balance_loss_mlp": 1.02212381,
      "epoch": 0.5097096134191066,
      "flos": 25849362839040.0,
      "grad_norm": 1.6992148136169019,
      "language_loss": 0.82559937,
      "learning_rate": 2.0342730748160588e-06,
      "loss": 0.84714139,
      "num_input_tokens_seen": 91408970,
      "step": 4239,
      "time_per_iteration": 2.846330165863037
    },
    {
      "auxiliary_loss_clip": 0.0115084,
      "auxiliary_loss_mlp": 0.01030062,
      "balance_loss_clip": 1.04413962,
      "balance_loss_mlp": 1.02101374,
      "epoch": 0.5098298563097456,
      "flos": 27745122844800.0,
      "grad_norm": 2.0694718899715054,
      "language_loss": 0.70392418,
      "learning_rate": 2.033494214974006e-06,
      "loss": 0.72573316,
      "num_input_tokens_seen": 91430115,
      "step": 4240,
      "time_per_iteration": 2.7652196884155273
    },
    {
      "auxiliary_loss_clip": 0.0114402,
      "auxiliary_loss_mlp": 0.01031944,
      "balance_loss_clip": 1.04774833,
      "balance_loss_mlp": 1.02411199,
      "epoch": 0.5099500992003848,
      "flos": 21358913011200.0,
      "grad_norm": 1.794649802168663,
      "language_loss": 0.83729398,
      "learning_rate": 2.0327153500509067e-06,
      "loss": 0.85905361,
      "num_input_tokens_seen": 91449140,
      "step": 4241,
      "time_per_iteration": 2.647798538208008
    },
    {
      "auxiliary_loss_clip": 0.01158512,
      "auxiliary_loss_mlp": 0.01027116,
      "balance_loss_clip": 1.05075884,
      "balance_loss_mlp": 1.01874113,
      "epoch": 0.5100703420910239,
      "flos": 19865999013120.0,
      "grad_norm": 1.8614640784367724,
      "language_loss": 0.84557128,
      "learning_rate": 2.031936480164916e-06,
      "loss": 0.86742759,
      "num_input_tokens_seen": 91466880,
      "step": 4242,
      "time_per_iteration": 2.7399678230285645
    },
    {
      "auxiliary_loss_clip": 0.01154377,
      "auxiliary_loss_mlp": 0.0103076,
      "balance_loss_clip": 1.05145836,
      "balance_loss_mlp": 1.0228442,
      "epoch": 0.5101905849816629,
      "flos": 24648797635200.0,
      "grad_norm": 2.2683545006256565,
      "language_loss": 0.80711693,
      "learning_rate": 2.0311576054341857e-06,
      "loss": 0.82896829,
      "num_input_tokens_seen": 91487495,
      "step": 4243,
      "time_per_iteration": 2.7955117225646973
    },
    {
      "auxiliary_loss_clip": 0.0118789,
      "auxiliary_loss_mlp": 0.01023954,
      "balance_loss_clip": 1.05485356,
      "balance_loss_mlp": 1.01572871,
      "epoch": 0.5103108278723021,
      "flos": 22930076787840.0,
      "grad_norm": 1.7032508627711769,
      "language_loss": 0.62651908,
      "learning_rate": 2.0303787259768715e-06,
      "loss": 0.64863753,
      "num_input_tokens_seen": 91508395,
      "step": 4244,
      "time_per_iteration": 2.6356124877929688
    },
    {
      "auxiliary_loss_clip": 0.01156915,
      "auxiliary_loss_mlp": 0.01027332,
      "balance_loss_clip": 1.05094552,
      "balance_loss_mlp": 1.01904345,
      "epoch": 0.5104310707629411,
      "flos": 21506613736320.0,
      "grad_norm": 2.213187162513074,
      "language_loss": 0.68718481,
      "learning_rate": 2.0295998419111294e-06,
      "loss": 0.70902729,
      "num_input_tokens_seen": 91525685,
      "step": 4245,
      "time_per_iteration": 2.660735845565796
    },
    {
      "auxiliary_loss_clip": 0.01115358,
      "auxiliary_loss_mlp": 0.01034647,
      "balance_loss_clip": 1.04345536,
      "balance_loss_mlp": 1.02722585,
      "epoch": 0.5105513136535802,
      "flos": 14903180403840.0,
      "grad_norm": 3.027812203900172,
      "language_loss": 0.73523748,
      "learning_rate": 2.028820953355115e-06,
      "loss": 0.75673753,
      "num_input_tokens_seen": 91543785,
      "step": 4246,
      "time_per_iteration": 2.8027307987213135
    },
    {
      "auxiliary_loss_clip": 0.01162914,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 1.04992092,
      "balance_loss_mlp": 1.0181551,
      "epoch": 0.5106715565442194,
      "flos": 22602212421120.0,
      "grad_norm": 2.425300865988715,
      "language_loss": 0.78717911,
      "learning_rate": 2.0280420604269834e-06,
      "loss": 0.80907869,
      "num_input_tokens_seen": 91563325,
      "step": 4247,
      "time_per_iteration": 2.7292001247406006
    },
    {
      "auxiliary_loss_clip": 0.01066434,
      "auxiliary_loss_mlp": 0.01005766,
      "balance_loss_clip": 1.0131892,
      "balance_loss_mlp": 1.00419223,
      "epoch": 0.5107917994348584,
      "flos": 71027645558400.0,
      "grad_norm": 0.7044462418895434,
      "language_loss": 0.58905363,
      "learning_rate": 2.027263163244895e-06,
      "loss": 0.60977566,
      "num_input_tokens_seen": 91632450,
      "step": 4248,
      "time_per_iteration": 3.3702101707458496
    },
    {
      "auxiliary_loss_clip": 0.01166681,
      "auxiliary_loss_mlp": 0.01025452,
      "balance_loss_clip": 1.05047011,
      "balance_loss_mlp": 1.01773357,
      "epoch": 0.5109120423254975,
      "flos": 24827416992000.0,
      "grad_norm": 1.5559987202658407,
      "language_loss": 0.74623263,
      "learning_rate": 2.026484261927005e-06,
      "loss": 0.76815397,
      "num_input_tokens_seen": 91651945,
      "step": 4249,
      "time_per_iteration": 3.6684329509735107
    },
    {
      "auxiliary_loss_clip": 0.01177382,
      "auxiliary_loss_mlp": 0.01029518,
      "balance_loss_clip": 1.054039,
      "balance_loss_mlp": 1.02122676,
      "epoch": 0.5110322852161366,
      "flos": 21247661612160.0,
      "grad_norm": 2.0522762036906865,
      "language_loss": 0.73607242,
      "learning_rate": 2.025705356591475e-06,
      "loss": 0.75814146,
      "num_input_tokens_seen": 91669635,
      "step": 4250,
      "time_per_iteration": 3.6083621978759766
    },
    {
      "auxiliary_loss_clip": 0.01041524,
      "auxiliary_loss_mlp": 0.00754167,
      "balance_loss_clip": 1.01146543,
      "balance_loss_mlp": 1.00056577,
      "epoch": 0.5111525281067757,
      "flos": 66457114358400.0,
      "grad_norm": 0.7580895053274237,
      "language_loss": 0.57960117,
      "learning_rate": 2.024926447356462e-06,
      "loss": 0.59755814,
      "num_input_tokens_seen": 91731920,
      "step": 4251,
      "time_per_iteration": 4.156554222106934
    },
    {
      "auxiliary_loss_clip": 0.01171486,
      "auxiliary_loss_mlp": 0.01027407,
      "balance_loss_clip": 1.05207133,
      "balance_loss_mlp": 1.01897335,
      "epoch": 0.5112727709974147,
      "flos": 14866731077760.0,
      "grad_norm": 2.025099352185664,
      "language_loss": 0.79051709,
      "learning_rate": 2.024147534340127e-06,
      "loss": 0.81250602,
      "num_input_tokens_seen": 91749780,
      "step": 4252,
      "time_per_iteration": 2.684636116027832
    },
    {
      "auxiliary_loss_clip": 0.01149682,
      "auxiliary_loss_mlp": 0.01025875,
      "balance_loss_clip": 1.04406285,
      "balance_loss_mlp": 1.01788771,
      "epoch": 0.5113930138880539,
      "flos": 21177600134400.0,
      "grad_norm": 1.5745444452456543,
      "language_loss": 0.79480481,
      "learning_rate": 2.02336861766063e-06,
      "loss": 0.81656027,
      "num_input_tokens_seen": 91768840,
      "step": 4253,
      "time_per_iteration": 2.6903581619262695
    },
    {
      "auxiliary_loss_clip": 0.01180987,
      "auxiliary_loss_mlp": 0.01033267,
      "balance_loss_clip": 1.05561101,
      "balance_loss_mlp": 1.02468395,
      "epoch": 0.511513256778693,
      "flos": 20409111630720.0,
      "grad_norm": 1.8723186010908548,
      "language_loss": 0.7870307,
      "learning_rate": 2.0225896974361327e-06,
      "loss": 0.80917329,
      "num_input_tokens_seen": 91788945,
      "step": 4254,
      "time_per_iteration": 3.6285934448242188
    },
    {
      "auxiliary_loss_clip": 0.01042841,
      "auxiliary_loss_mlp": 0.01005618,
      "balance_loss_clip": 1.01201415,
      "balance_loss_mlp": 1.00399709,
      "epoch": 0.511633499669332,
      "flos": 69879975131520.0,
      "grad_norm": 0.8651359059432232,
      "language_loss": 0.59944272,
      "learning_rate": 2.0218107737847962e-06,
      "loss": 0.61992741,
      "num_input_tokens_seen": 91850990,
      "step": 4255,
      "time_per_iteration": 3.2887799739837646
    },
    {
      "auxiliary_loss_clip": 0.01185306,
      "auxiliary_loss_mlp": 0.01029474,
      "balance_loss_clip": 1.05386555,
      "balance_loss_mlp": 1.02108157,
      "epoch": 0.5117537425599712,
      "flos": 24097855852800.0,
      "grad_norm": 2.6342995059425514,
      "language_loss": 0.74327075,
      "learning_rate": 2.0210318468247826e-06,
      "loss": 0.76541853,
      "num_input_tokens_seen": 91869960,
      "step": 4256,
      "time_per_iteration": 2.632917881011963
    },
    {
      "auxiliary_loss_clip": 0.01157301,
      "auxiliary_loss_mlp": 0.01030208,
      "balance_loss_clip": 1.04865766,
      "balance_loss_mlp": 1.02174449,
      "epoch": 0.5118739854506102,
      "flos": 20959550622720.0,
      "grad_norm": 1.8295152265896288,
      "language_loss": 0.81785405,
      "learning_rate": 2.020252916674255e-06,
      "loss": 0.83972913,
      "num_input_tokens_seen": 91889075,
      "step": 4257,
      "time_per_iteration": 2.763042688369751
    },
    {
      "auxiliary_loss_clip": 0.0117242,
      "auxiliary_loss_mlp": 0.01028029,
      "balance_loss_clip": 1.05064988,
      "balance_loss_mlp": 1.02030396,
      "epoch": 0.5119942283412493,
      "flos": 17457326749440.0,
      "grad_norm": 1.9272404792603803,
      "language_loss": 0.80987304,
      "learning_rate": 2.019473983451375e-06,
      "loss": 0.83187753,
      "num_input_tokens_seen": 91907495,
      "step": 4258,
      "time_per_iteration": 2.6178691387176514
    },
    {
      "auxiliary_loss_clip": 0.0114696,
      "auxiliary_loss_mlp": 0.01026603,
      "balance_loss_clip": 1.04692769,
      "balance_loss_mlp": 1.01844335,
      "epoch": 0.5121144712318885,
      "flos": 21066743784960.0,
      "grad_norm": 1.7925899813285122,
      "language_loss": 0.71570003,
      "learning_rate": 2.0186950472743076e-06,
      "loss": 0.7374357,
      "num_input_tokens_seen": 91927400,
      "step": 4259,
      "time_per_iteration": 2.7691292762756348
    },
    {
      "auxiliary_loss_clip": 0.01186069,
      "auxiliary_loss_mlp": 0.01025641,
      "balance_loss_clip": 1.0533402,
      "balance_loss_mlp": 1.01753473,
      "epoch": 0.5122347141225275,
      "flos": 19860791541120.0,
      "grad_norm": 1.63305182996989,
      "language_loss": 0.7391718,
      "learning_rate": 2.0179161082612162e-06,
      "loss": 0.76128894,
      "num_input_tokens_seen": 91946790,
      "step": 4260,
      "time_per_iteration": 2.6032440662384033
    },
    {
      "auxiliary_loss_clip": 0.0114899,
      "auxiliary_loss_mlp": 0.01025605,
      "balance_loss_clip": 1.04504967,
      "balance_loss_mlp": 1.01705217,
      "epoch": 0.5123549570131666,
      "flos": 22528487756160.0,
      "grad_norm": 1.9810234116866055,
      "language_loss": 0.72685218,
      "learning_rate": 2.017137166530266e-06,
      "loss": 0.7485981,
      "num_input_tokens_seen": 91966325,
      "step": 4261,
      "time_per_iteration": 2.668117046356201
    },
    {
      "auxiliary_loss_clip": 0.01162472,
      "auxiliary_loss_mlp": 0.01028829,
      "balance_loss_clip": 1.05199933,
      "balance_loss_mlp": 1.02092516,
      "epoch": 0.5124751999038056,
      "flos": 20333375804160.0,
      "grad_norm": 5.1719832720519845,
      "language_loss": 0.80202305,
      "learning_rate": 2.0163582221996213e-06,
      "loss": 0.8239361,
      "num_input_tokens_seen": 91984700,
      "step": 4262,
      "time_per_iteration": 2.715456962585449
    },
    {
      "auxiliary_loss_clip": 0.01156224,
      "auxiliary_loss_mlp": 0.01025792,
      "balance_loss_clip": 1.04874611,
      "balance_loss_mlp": 1.01721513,
      "epoch": 0.5125954427944448,
      "flos": 39785970211200.0,
      "grad_norm": 2.1120652283810024,
      "language_loss": 0.68114227,
      "learning_rate": 2.015579275387446e-06,
      "loss": 0.70296246,
      "num_input_tokens_seen": 92010020,
      "step": 4263,
      "time_per_iteration": 2.794590711593628
    },
    {
      "auxiliary_loss_clip": 0.0114883,
      "auxiliary_loss_mlp": 0.01030601,
      "balance_loss_clip": 1.04952013,
      "balance_loss_mlp": 1.02246475,
      "epoch": 0.5127156856850839,
      "flos": 29205394358400.0,
      "grad_norm": 1.8700899212970121,
      "language_loss": 0.68795878,
      "learning_rate": 2.0148003262119085e-06,
      "loss": 0.7097531,
      "num_input_tokens_seen": 92030990,
      "step": 4264,
      "time_per_iteration": 2.7926645278930664
    },
    {
      "auxiliary_loss_clip": 0.0114204,
      "auxiliary_loss_mlp": 0.01026019,
      "balance_loss_clip": 1.0480423,
      "balance_loss_mlp": 1.01758456,
      "epoch": 0.5128359285757229,
      "flos": 13553693412480.0,
      "grad_norm": 1.9056512514797443,
      "language_loss": 0.7653532,
      "learning_rate": 2.0140213747911728e-06,
      "loss": 0.7870338,
      "num_input_tokens_seen": 92049525,
      "step": 4265,
      "time_per_iteration": 2.7287943363189697
    },
    {
      "auxiliary_loss_clip": 0.01142349,
      "auxiliary_loss_mlp": 0.01032716,
      "balance_loss_clip": 1.05145586,
      "balance_loss_mlp": 1.02441931,
      "epoch": 0.5129561714663621,
      "flos": 25192089820800.0,
      "grad_norm": 2.0986063084028856,
      "language_loss": 0.80794322,
      "learning_rate": 2.013242421243406e-06,
      "loss": 0.82969391,
      "num_input_tokens_seen": 92068430,
      "step": 4266,
      "time_per_iteration": 2.807281017303467
    },
    {
      "auxiliary_loss_clip": 0.01129523,
      "auxiliary_loss_mlp": 0.01029965,
      "balance_loss_clip": 1.04948401,
      "balance_loss_mlp": 1.02177811,
      "epoch": 0.5130764143570011,
      "flos": 18150223080960.0,
      "grad_norm": 1.7032575172625988,
      "language_loss": 0.789662,
      "learning_rate": 2.012463465686774e-06,
      "loss": 0.81125689,
      "num_input_tokens_seen": 92088180,
      "step": 4267,
      "time_per_iteration": 2.755438804626465
    },
    {
      "auxiliary_loss_clip": 0.01043825,
      "auxiliary_loss_mlp": 0.01002901,
      "balance_loss_clip": 1.01956868,
      "balance_loss_mlp": 1.00147653,
      "epoch": 0.5131966572476402,
      "flos": 59794896418560.0,
      "grad_norm": 0.766405452087811,
      "language_loss": 0.54804635,
      "learning_rate": 2.0116845082394446e-06,
      "loss": 0.56851363,
      "num_input_tokens_seen": 92153015,
      "step": 4268,
      "time_per_iteration": 3.319882392883301
    },
    {
      "auxiliary_loss_clip": 0.01170277,
      "auxiliary_loss_mlp": 0.01028815,
      "balance_loss_clip": 1.04777825,
      "balance_loss_mlp": 1.02083969,
      "epoch": 0.5133169001382794,
      "flos": 18515219132160.0,
      "grad_norm": 2.0464611180038585,
      "language_loss": 0.78935778,
      "learning_rate": 2.0109055490195836e-06,
      "loss": 0.81134868,
      "num_input_tokens_seen": 92171470,
      "step": 4269,
      "time_per_iteration": 2.6687822341918945
    },
    {
      "auxiliary_loss_clip": 0.0111896,
      "auxiliary_loss_mlp": 0.01029802,
      "balance_loss_clip": 1.04021096,
      "balance_loss_mlp": 1.0217073,
      "epoch": 0.5134371430289184,
      "flos": 15523537219200.0,
      "grad_norm": 2.1188305272805983,
      "language_loss": 0.64121878,
      "learning_rate": 2.0101265881453605e-06,
      "loss": 0.66270638,
      "num_input_tokens_seen": 92189945,
      "step": 4270,
      "time_per_iteration": 2.7287752628326416
    },
    {
      "auxiliary_loss_clip": 0.01148988,
      "auxiliary_loss_mlp": 0.01026841,
      "balance_loss_clip": 1.05000579,
      "balance_loss_mlp": 1.01932538,
      "epoch": 0.5135573859195575,
      "flos": 21433786911360.0,
      "grad_norm": 2.0910050434895533,
      "language_loss": 0.78196585,
      "learning_rate": 2.009347625734941e-06,
      "loss": 0.80372405,
      "num_input_tokens_seen": 92209855,
      "step": 4271,
      "time_per_iteration": 2.691169023513794
    },
    {
      "auxiliary_loss_clip": 0.01190299,
      "auxiliary_loss_mlp": 0.0103072,
      "balance_loss_clip": 1.05705631,
      "balance_loss_mlp": 1.02208352,
      "epoch": 0.5136776288101966,
      "flos": 17712651600000.0,
      "grad_norm": 2.645288233279411,
      "language_loss": 0.75510395,
      "learning_rate": 2.0085686619064954e-06,
      "loss": 0.77731407,
      "num_input_tokens_seen": 92226295,
      "step": 4272,
      "time_per_iteration": 2.5454154014587402
    },
    {
      "auxiliary_loss_clip": 0.01176812,
      "auxiliary_loss_mlp": 0.01037963,
      "balance_loss_clip": 1.05335832,
      "balance_loss_mlp": 1.02966011,
      "epoch": 0.5137978717008357,
      "flos": 16581680997120.0,
      "grad_norm": 2.083646639310859,
      "language_loss": 0.82764393,
      "learning_rate": 2.00778969677819e-06,
      "loss": 0.84979171,
      "num_input_tokens_seen": 92243330,
      "step": 4273,
      "time_per_iteration": 2.616377115249634
    },
    {
      "auxiliary_loss_clip": 0.01154381,
      "auxiliary_loss_mlp": 0.0102921,
      "balance_loss_clip": 1.04875863,
      "balance_loss_mlp": 1.0208658,
      "epoch": 0.5139181145914747,
      "flos": 20668243322880.0,
      "grad_norm": 1.7736662017803648,
      "language_loss": 0.63900268,
      "learning_rate": 2.0070107304681934e-06,
      "loss": 0.66083866,
      "num_input_tokens_seen": 92262285,
      "step": 4274,
      "time_per_iteration": 2.67352294921875
    },
    {
      "auxiliary_loss_clip": 0.0114056,
      "auxiliary_loss_mlp": 0.01024389,
      "balance_loss_clip": 1.04943752,
      "balance_loss_mlp": 1.017308,
      "epoch": 0.5140383574821139,
      "flos": 32926996546560.0,
      "grad_norm": 2.4687008746462564,
      "language_loss": 0.78343248,
      "learning_rate": 2.006231763094675e-06,
      "loss": 0.80508196,
      "num_input_tokens_seen": 92283305,
      "step": 4275,
      "time_per_iteration": 3.7122576236724854
    },
    {
      "auxiliary_loss_clip": 0.01155313,
      "auxiliary_loss_mlp": 0.01034527,
      "balance_loss_clip": 1.05520844,
      "balance_loss_mlp": 1.02665353,
      "epoch": 0.514158600372753,
      "flos": 19537093152000.0,
      "grad_norm": 2.3229218222063026,
      "language_loss": 0.88187015,
      "learning_rate": 2.0054527947758027e-06,
      "loss": 0.90376854,
      "num_input_tokens_seen": 92302105,
      "step": 4276,
      "time_per_iteration": 3.649789333343506
    },
    {
      "auxiliary_loss_clip": 0.01065533,
      "auxiliary_loss_mlp": 0.01002036,
      "balance_loss_clip": 1.01321971,
      "balance_loss_mlp": 1.00058126,
      "epoch": 0.514278843263392,
      "flos": 62523855279360.0,
      "grad_norm": 0.7242175184846665,
      "language_loss": 0.56000376,
      "learning_rate": 2.004673825629746e-06,
      "loss": 0.58067942,
      "num_input_tokens_seen": 92362885,
      "step": 4277,
      "time_per_iteration": 3.2330658435821533
    },
    {
      "auxiliary_loss_clip": 0.01148773,
      "auxiliary_loss_mlp": 0.01027744,
      "balance_loss_clip": 1.04590261,
      "balance_loss_mlp": 1.01944089,
      "epoch": 0.5143990861540312,
      "flos": 25882328545920.0,
      "grad_norm": 2.2620290086841344,
      "language_loss": 0.72838175,
      "learning_rate": 2.0038948557746744e-06,
      "loss": 0.75014687,
      "num_input_tokens_seen": 92384740,
      "step": 4278,
      "time_per_iteration": 2.771160364151001
    },
    {
      "auxiliary_loss_clip": 0.01166885,
      "auxiliary_loss_mlp": 0.01025159,
      "balance_loss_clip": 1.05053377,
      "balance_loss_mlp": 1.01723766,
      "epoch": 0.5145193290446702,
      "flos": 23330660238720.0,
      "grad_norm": 1.6727543139773986,
      "language_loss": 0.75154901,
      "learning_rate": 2.0031158853287558e-06,
      "loss": 0.77346945,
      "num_input_tokens_seen": 92405175,
      "step": 4279,
      "time_per_iteration": 2.7449469566345215
    },
    {
      "auxiliary_loss_clip": 0.01153678,
      "auxiliary_loss_mlp": 0.0102767,
      "balance_loss_clip": 1.05079222,
      "balance_loss_mlp": 1.01920605,
      "epoch": 0.5146395719353093,
      "flos": 22856603518080.0,
      "grad_norm": 2.4776488641643275,
      "language_loss": 0.70457232,
      "learning_rate": 2.0023369144101593e-06,
      "loss": 0.72638583,
      "num_input_tokens_seen": 92423345,
      "step": 4280,
      "time_per_iteration": 3.576746940612793
    },
    {
      "auxiliary_loss_clip": 0.01145833,
      "auxiliary_loss_mlp": 0.01027868,
      "balance_loss_clip": 1.04739356,
      "balance_loss_mlp": 1.02017856,
      "epoch": 0.5147598148259485,
      "flos": 26391577616640.0,
      "grad_norm": 2.057315456394534,
      "language_loss": 0.76820922,
      "learning_rate": 2.0015579431370555e-06,
      "loss": 0.78994626,
      "num_input_tokens_seen": 92445025,
      "step": 4281,
      "time_per_iteration": 2.7272138595581055
    },
    {
      "auxiliary_loss_clip": 0.01166879,
      "auxiliary_loss_mlp": 0.0102721,
      "balance_loss_clip": 1.05113935,
      "balance_loss_mlp": 1.01953876,
      "epoch": 0.5148800577165875,
      "flos": 29965694561280.0,
      "grad_norm": 2.380719225964806,
      "language_loss": 0.69996184,
      "learning_rate": 2.000778971627612e-06,
      "loss": 0.72190273,
      "num_input_tokens_seen": 92464490,
      "step": 4282,
      "time_per_iteration": 2.7115352153778076
    },
    {
      "auxiliary_loss_clip": 0.01149994,
      "auxiliary_loss_mlp": 0.01030496,
      "balance_loss_clip": 1.04831004,
      "balance_loss_mlp": 1.02262843,
      "epoch": 0.5150003006072266,
      "flos": 17931383470080.0,
      "grad_norm": 2.009446677089578,
      "language_loss": 0.89985871,
      "learning_rate": 2e-06,
      "loss": 0.92166364,
      "num_input_tokens_seen": 92482085,
      "step": 4283,
      "time_per_iteration": 2.6880173683166504
    },
    {
      "auxiliary_loss_clip": 0.01183246,
      "auxiliary_loss_mlp": 0.01029534,
      "balance_loss_clip": 1.05305302,
      "balance_loss_mlp": 1.02146983,
      "epoch": 0.5151205434978657,
      "flos": 18478733892480.0,
      "grad_norm": 1.9601885009184798,
      "language_loss": 0.85772777,
      "learning_rate": 1.9992210283723878e-06,
      "loss": 0.87985563,
      "num_input_tokens_seen": 92499325,
      "step": 4284,
      "time_per_iteration": 2.6190545558929443
    },
    {
      "auxiliary_loss_clip": 0.01182792,
      "auxiliary_loss_mlp": 0.01025867,
      "balance_loss_clip": 1.05268478,
      "balance_loss_mlp": 1.01837492,
      "epoch": 0.5152407863885048,
      "flos": 25341263003520.0,
      "grad_norm": 1.5385854226168054,
      "language_loss": 0.79308045,
      "learning_rate": 1.9984420568629448e-06,
      "loss": 0.81516701,
      "num_input_tokens_seen": 92522090,
      "step": 4285,
      "time_per_iteration": 2.717719793319702
    },
    {
      "auxiliary_loss_clip": 0.01171392,
      "auxiliary_loss_mlp": 0.01024383,
      "balance_loss_clip": 1.05194235,
      "balance_loss_mlp": 1.01690233,
      "epoch": 0.5153610292791438,
      "flos": 18329740277760.0,
      "grad_norm": 2.3473583069748356,
      "language_loss": 0.78541231,
      "learning_rate": 1.9976630855898405e-06,
      "loss": 0.80737001,
      "num_input_tokens_seen": 92539845,
      "step": 4286,
      "time_per_iteration": 2.619304656982422
    },
    {
      "auxiliary_loss_clip": 0.01148474,
      "auxiliary_loss_mlp": 0.01020987,
      "balance_loss_clip": 1.04450536,
      "balance_loss_mlp": 1.0143261,
      "epoch": 0.515481272169783,
      "flos": 30409945971840.0,
      "grad_norm": 2.2663077152240882,
      "language_loss": 0.74394208,
      "learning_rate": 1.9968841146712445e-06,
      "loss": 0.76563668,
      "num_input_tokens_seen": 92559460,
      "step": 4287,
      "time_per_iteration": 2.774848222732544
    },
    {
      "auxiliary_loss_clip": 0.01108183,
      "auxiliary_loss_mlp": 0.00762868,
      "balance_loss_clip": 1.04182935,
      "balance_loss_mlp": 1.00033283,
      "epoch": 0.5156015150604221,
      "flos": 23037305863680.0,
      "grad_norm": 1.693699981919337,
      "language_loss": 0.71627283,
      "learning_rate": 1.996105144225326e-06,
      "loss": 0.73498338,
      "num_input_tokens_seen": 92579695,
      "step": 4288,
      "time_per_iteration": 2.7832891941070557
    },
    {
      "auxiliary_loss_clip": 0.01166559,
      "auxiliary_loss_mlp": 0.01028194,
      "balance_loss_clip": 1.05139542,
      "balance_loss_mlp": 1.02072573,
      "epoch": 0.5157217579510611,
      "flos": 17858556645120.0,
      "grad_norm": 2.203241670002149,
      "language_loss": 0.78854704,
      "learning_rate": 1.995326174370254e-06,
      "loss": 0.8104946,
      "num_input_tokens_seen": 92598795,
      "step": 4289,
      "time_per_iteration": 2.73026180267334
    },
    {
      "auxiliary_loss_clip": 0.01165488,
      "auxiliary_loss_mlp": 0.00761947,
      "balance_loss_clip": 1.04841137,
      "balance_loss_mlp": 1.00019884,
      "epoch": 0.5158420008417003,
      "flos": 19171486569600.0,
      "grad_norm": 1.4380366010003816,
      "language_loss": 0.7294836,
      "learning_rate": 1.994547205224197e-06,
      "loss": 0.74875802,
      "num_input_tokens_seen": 92617700,
      "step": 4290,
      "time_per_iteration": 2.7143137454986572
    },
    {
      "auxiliary_loss_clip": 0.01150446,
      "auxiliary_loss_mlp": 0.01023819,
      "balance_loss_clip": 1.05000687,
      "balance_loss_mlp": 1.0162822,
      "epoch": 0.5159622437323393,
      "flos": 22419534827520.0,
      "grad_norm": 2.071379412065401,
      "language_loss": 0.67949325,
      "learning_rate": 1.993768236905325e-06,
      "loss": 0.70123589,
      "num_input_tokens_seen": 92638370,
      "step": 4291,
      "time_per_iteration": 2.7979423999786377
    },
    {
      "auxiliary_loss_clip": 0.01149913,
      "auxiliary_loss_mlp": 0.01030548,
      "balance_loss_clip": 1.04781842,
      "balance_loss_mlp": 1.02216792,
      "epoch": 0.5160824866229784,
      "flos": 24603010773120.0,
      "grad_norm": 3.8264167145017,
      "language_loss": 0.65815902,
      "learning_rate": 1.992989269531807e-06,
      "loss": 0.67996365,
      "num_input_tokens_seen": 92657180,
      "step": 4292,
      "time_per_iteration": 2.765042781829834
    },
    {
      "auxiliary_loss_clip": 0.01151664,
      "auxiliary_loss_mlp": 0.01022094,
      "balance_loss_clip": 1.04682732,
      "balance_loss_mlp": 1.01485193,
      "epoch": 0.5162027295136175,
      "flos": 18002737837440.0,
      "grad_norm": 3.3694371301829267,
      "language_loss": 0.6806519,
      "learning_rate": 1.99221030322181e-06,
      "loss": 0.70238948,
      "num_input_tokens_seen": 92673985,
      "step": 4293,
      "time_per_iteration": 2.665212392807007
    },
    {
      "auxiliary_loss_clip": 0.01157002,
      "auxiliary_loss_mlp": 0.01031287,
      "balance_loss_clip": 1.04738617,
      "balance_loss_mlp": 1.0238601,
      "epoch": 0.5163229724042566,
      "flos": 27344611221120.0,
      "grad_norm": 1.8073466057013952,
      "language_loss": 0.81111461,
      "learning_rate": 1.991431338093505e-06,
      "loss": 0.83299744,
      "num_input_tokens_seen": 92696340,
      "step": 4294,
      "time_per_iteration": 2.8028039932250977
    },
    {
      "auxiliary_loss_clip": 0.01154074,
      "auxiliary_loss_mlp": 0.01025265,
      "balance_loss_clip": 1.05084038,
      "balance_loss_mlp": 1.01848173,
      "epoch": 0.5164432152948957,
      "flos": 21762764599680.0,
      "grad_norm": 1.9014689419192339,
      "language_loss": 0.79511631,
      "learning_rate": 1.9906523742650587e-06,
      "loss": 0.81690967,
      "num_input_tokens_seen": 92715200,
      "step": 4295,
      "time_per_iteration": 2.685328245162964
    },
    {
      "auxiliary_loss_clip": 0.01183045,
      "auxiliary_loss_mlp": 0.01027933,
      "balance_loss_clip": 1.0509671,
      "balance_loss_mlp": 1.01949334,
      "epoch": 0.5165634581855347,
      "flos": 25550334115200.0,
      "grad_norm": 1.9879206144633594,
      "language_loss": 0.77237082,
      "learning_rate": 1.9898734118546397e-06,
      "loss": 0.79448056,
      "num_input_tokens_seen": 92735150,
      "step": 4296,
      "time_per_iteration": 2.6982617378234863
    },
    {
      "auxiliary_loss_clip": 0.01103422,
      "auxiliary_loss_mlp": 0.01029612,
      "balance_loss_clip": 1.04485273,
      "balance_loss_mlp": 1.0215776,
      "epoch": 0.5166837010761739,
      "flos": 19901191363200.0,
      "grad_norm": 1.530887177881446,
      "language_loss": 0.80446017,
      "learning_rate": 1.989094450980416e-06,
      "loss": 0.82579052,
      "num_input_tokens_seen": 92755250,
      "step": 4297,
      "time_per_iteration": 2.922832489013672
    },
    {
      "auxiliary_loss_clip": 0.01166279,
      "auxiliary_loss_mlp": 0.01021704,
      "balance_loss_clip": 1.04854882,
      "balance_loss_mlp": 1.01428354,
      "epoch": 0.516803943966813,
      "flos": 26646076454400.0,
      "grad_norm": 2.1565564190771274,
      "language_loss": 0.77001357,
      "learning_rate": 1.9883154917605556e-06,
      "loss": 0.79189342,
      "num_input_tokens_seen": 92774460,
      "step": 4298,
      "time_per_iteration": 3.5020172595977783
    },
    {
      "auxiliary_loss_clip": 0.01180732,
      "auxiliary_loss_mlp": 0.01031174,
      "balance_loss_clip": 1.05046129,
      "balance_loss_mlp": 1.02362549,
      "epoch": 0.516924186857452,
      "flos": 19682854542720.0,
      "grad_norm": 1.747453116700484,
      "language_loss": 0.83398557,
      "learning_rate": 1.9875365343132262e-06,
      "loss": 0.85610467,
      "num_input_tokens_seen": 92791580,
      "step": 4299,
      "time_per_iteration": 2.6476144790649414
    },
    {
      "auxiliary_loss_clip": 0.01169509,
      "auxiliary_loss_mlp": 0.00762452,
      "balance_loss_clip": 1.05179965,
      "balance_loss_mlp": 1.00024676,
      "epoch": 0.5170444297480912,
      "flos": 15956583586560.0,
      "grad_norm": 2.210165327965464,
      "language_loss": 0.8466984,
      "learning_rate": 1.9867575787565946e-06,
      "loss": 0.86601806,
      "num_input_tokens_seen": 92806240,
      "step": 4300,
      "time_per_iteration": 2.61966609954834
    },
    {
      "auxiliary_loss_clip": 0.01169294,
      "auxiliary_loss_mlp": 0.01035855,
      "balance_loss_clip": 1.05197835,
      "balance_loss_mlp": 1.02693784,
      "epoch": 0.5171646726387302,
      "flos": 14174157968640.0,
      "grad_norm": 2.2497316143525334,
      "language_loss": 0.8627699,
      "learning_rate": 1.9859786252088275e-06,
      "loss": 0.88482141,
      "num_input_tokens_seen": 92823420,
      "step": 4301,
      "time_per_iteration": 2.6327226161956787
    },
    {
      "auxiliary_loss_clip": 0.01145082,
      "auxiliary_loss_mlp": 0.01025552,
      "balance_loss_clip": 1.04860485,
      "balance_loss_mlp": 1.01705849,
      "epoch": 0.5172849155293693,
      "flos": 23578550974080.0,
      "grad_norm": 4.80721895216572,
      "language_loss": 0.66851074,
      "learning_rate": 1.9851996737880914e-06,
      "loss": 0.69021714,
      "num_input_tokens_seen": 92838605,
      "step": 4302,
      "time_per_iteration": 4.677529573440552
    },
    {
      "auxiliary_loss_clip": 0.01175581,
      "auxiliary_loss_mlp": 0.01031469,
      "balance_loss_clip": 1.05193019,
      "balance_loss_mlp": 1.02302265,
      "epoch": 0.5174051584200084,
      "flos": 14283541860480.0,
      "grad_norm": 2.764068710313437,
      "language_loss": 0.74488968,
      "learning_rate": 1.9844207246125537e-06,
      "loss": 0.76696014,
      "num_input_tokens_seen": 92855185,
      "step": 4303,
      "time_per_iteration": 2.5673739910125732
    },
    {
      "auxiliary_loss_clip": 0.01151678,
      "auxiliary_loss_mlp": 0.01029568,
      "balance_loss_clip": 1.04883265,
      "balance_loss_mlp": 1.02193236,
      "epoch": 0.5175254013106475,
      "flos": 37889384192640.0,
      "grad_norm": 1.7025420924983574,
      "language_loss": 0.68280017,
      "learning_rate": 1.983641777800379e-06,
      "loss": 0.70461267,
      "num_input_tokens_seen": 92877830,
      "step": 4304,
      "time_per_iteration": 2.850768804550171
    },
    {
      "auxiliary_loss_clip": 0.01058774,
      "auxiliary_loss_mlp": 0.01003957,
      "balance_loss_clip": 1.01201105,
      "balance_loss_mlp": 1.00253868,
      "epoch": 0.5176456442012866,
      "flos": 68549737829760.0,
      "grad_norm": 0.746463696035837,
      "language_loss": 0.5885644,
      "learning_rate": 1.9828628334697343e-06,
      "loss": 0.60919178,
      "num_input_tokens_seen": 92945040,
      "step": 4305,
      "time_per_iteration": 3.43418288230896
    },
    {
      "auxiliary_loss_clip": 0.01058337,
      "auxiliary_loss_mlp": 0.01002276,
      "balance_loss_clip": 1.01145315,
      "balance_loss_mlp": 1.00082743,
      "epoch": 0.5177658870919257,
      "flos": 64084137235200.0,
      "grad_norm": 0.7573182130819075,
      "language_loss": 0.54652345,
      "learning_rate": 1.982083891738784e-06,
      "loss": 0.56712961,
      "num_input_tokens_seen": 93005910,
      "step": 4306,
      "time_per_iteration": 4.546627759933472
    },
    {
      "auxiliary_loss_clip": 0.01149406,
      "auxiliary_loss_mlp": 0.01025831,
      "balance_loss_clip": 1.05183983,
      "balance_loss_mlp": 1.01866651,
      "epoch": 0.5178861299825648,
      "flos": 26651248012800.0,
      "grad_norm": 1.4791126017477254,
      "language_loss": 0.82854164,
      "learning_rate": 1.9813049527256923e-06,
      "loss": 0.85029399,
      "num_input_tokens_seen": 93026305,
      "step": 4307,
      "time_per_iteration": 2.794661283493042
    },
    {
      "auxiliary_loss_clip": 0.01138295,
      "auxiliary_loss_mlp": 0.01030477,
      "balance_loss_clip": 1.04468429,
      "balance_loss_mlp": 1.02276444,
      "epoch": 0.5180063728732038,
      "flos": 17931886260480.0,
      "grad_norm": 2.8471841404472817,
      "language_loss": 0.82032025,
      "learning_rate": 1.9805260165486252e-06,
      "loss": 0.84200799,
      "num_input_tokens_seen": 93045675,
      "step": 4308,
      "time_per_iteration": 2.731330633163452
    },
    {
      "auxiliary_loss_clip": 0.01170605,
      "auxiliary_loss_mlp": 0.01029754,
      "balance_loss_clip": 1.05362296,
      "balance_loss_mlp": 1.02166533,
      "epoch": 0.518126615763843,
      "flos": 19500895221120.0,
      "grad_norm": 2.05320601974293,
      "language_loss": 0.86155128,
      "learning_rate": 1.9797470833257457e-06,
      "loss": 0.88355482,
      "num_input_tokens_seen": 93065375,
      "step": 4309,
      "time_per_iteration": 2.7096712589263916
    },
    {
      "auxiliary_loss_clip": 0.01170917,
      "auxiliary_loss_mlp": 0.01032238,
      "balance_loss_clip": 1.05297089,
      "balance_loss_mlp": 1.0241611,
      "epoch": 0.5182468586544821,
      "flos": 20704082117760.0,
      "grad_norm": 2.015952670555153,
      "language_loss": 0.77566314,
      "learning_rate": 1.9789681531752177e-06,
      "loss": 0.79769468,
      "num_input_tokens_seen": 93085595,
      "step": 4310,
      "time_per_iteration": 2.6832869052886963
    },
    {
      "auxiliary_loss_clip": 0.01124798,
      "auxiliary_loss_mlp": 0.01029112,
      "balance_loss_clip": 1.0490123,
      "balance_loss_mlp": 1.02203107,
      "epoch": 0.5183671015451211,
      "flos": 23112107936640.0,
      "grad_norm": 1.7438015801676086,
      "language_loss": 0.72280329,
      "learning_rate": 1.978189226215204e-06,
      "loss": 0.74434245,
      "num_input_tokens_seen": 93106140,
      "step": 4311,
      "time_per_iteration": 2.7803261280059814
    },
    {
      "auxiliary_loss_clip": 0.01182343,
      "auxiliary_loss_mlp": 0.01031963,
      "balance_loss_clip": 1.05242205,
      "balance_loss_mlp": 1.02360058,
      "epoch": 0.5184873444357603,
      "flos": 17597090568960.0,
      "grad_norm": 1.709753032897899,
      "language_loss": 0.77218163,
      "learning_rate": 1.9774103025638675e-06,
      "loss": 0.79432464,
      "num_input_tokens_seen": 93124265,
      "step": 4312,
      "time_per_iteration": 2.6822755336761475
    },
    {
      "auxiliary_loss_clip": 0.01128756,
      "auxiliary_loss_mlp": 0.01030048,
      "balance_loss_clip": 1.05034888,
      "balance_loss_mlp": 1.02238917,
      "epoch": 0.5186075873263993,
      "flos": 24936800883840.0,
      "grad_norm": 1.8179613403278088,
      "language_loss": 0.76427615,
      "learning_rate": 1.9766313823393696e-06,
      "loss": 0.78586423,
      "num_input_tokens_seen": 93145130,
      "step": 4313,
      "time_per_iteration": 2.8908841609954834
    },
    {
      "auxiliary_loss_clip": 0.01117876,
      "auxiliary_loss_mlp": 0.0103459,
      "balance_loss_clip": 1.04202104,
      "balance_loss_mlp": 1.02634096,
      "epoch": 0.5187278302170384,
      "flos": 15190106244480.0,
      "grad_norm": 2.4885987604226325,
      "language_loss": 0.6930393,
      "learning_rate": 1.975852465659873e-06,
      "loss": 0.71456397,
      "num_input_tokens_seen": 93161110,
      "step": 4314,
      "time_per_iteration": 2.889552116394043
    },
    {
      "auxiliary_loss_clip": 0.01171871,
      "auxiliary_loss_mlp": 0.0103556,
      "balance_loss_clip": 1.05289745,
      "balance_loss_mlp": 1.02708375,
      "epoch": 0.5188480731076776,
      "flos": 25009412227200.0,
      "grad_norm": 2.20299208676975,
      "language_loss": 0.69978368,
      "learning_rate": 1.9750735526435377e-06,
      "loss": 0.72185802,
      "num_input_tokens_seen": 93178055,
      "step": 4315,
      "time_per_iteration": 2.708498477935791
    },
    {
      "auxiliary_loss_clip": 0.01154784,
      "auxiliary_loss_mlp": 0.01030183,
      "balance_loss_clip": 1.05182028,
      "balance_loss_mlp": 1.02254725,
      "epoch": 0.5189683159983166,
      "flos": 24790141653120.0,
      "grad_norm": 3.0159061290431795,
      "language_loss": 0.79643381,
      "learning_rate": 1.974294643408525e-06,
      "loss": 0.81828344,
      "num_input_tokens_seen": 93195850,
      "step": 4316,
      "time_per_iteration": 2.737149715423584
    },
    {
      "auxiliary_loss_clip": 0.01173919,
      "auxiliary_loss_mlp": 0.01025812,
      "balance_loss_clip": 1.05118871,
      "balance_loss_mlp": 1.01836109,
      "epoch": 0.5190885588889557,
      "flos": 24754266944640.0,
      "grad_norm": 3.3139351997769997,
      "language_loss": 0.67140168,
      "learning_rate": 1.9735157380729947e-06,
      "loss": 0.69339895,
      "num_input_tokens_seen": 93216260,
      "step": 4317,
      "time_per_iteration": 2.722264289855957
    },
    {
      "auxiliary_loss_clip": 0.0115436,
      "auxiliary_loss_mlp": 0.01024659,
      "balance_loss_clip": 1.04831576,
      "balance_loss_mlp": 1.01718163,
      "epoch": 0.5192088017795948,
      "flos": 24712646060160.0,
      "grad_norm": 1.8461758199598859,
      "language_loss": 0.84131515,
      "learning_rate": 1.9727368367551053e-06,
      "loss": 0.86310536,
      "num_input_tokens_seen": 93234810,
      "step": 4318,
      "time_per_iteration": 2.7310335636138916
    },
    {
      "auxiliary_loss_clip": 0.01143637,
      "auxiliary_loss_mlp": 0.01030053,
      "balance_loss_clip": 1.04699135,
      "balance_loss_mlp": 1.02148199,
      "epoch": 0.5193290446702339,
      "flos": 27229588894080.0,
      "grad_norm": 1.8255274639666053,
      "language_loss": 0.68338549,
      "learning_rate": 1.9719579395730164e-06,
      "loss": 0.70512241,
      "num_input_tokens_seen": 93254185,
      "step": 4319,
      "time_per_iteration": 2.806253433227539
    },
    {
      "auxiliary_loss_clip": 0.0118726,
      "auxiliary_loss_mlp": 0.01029105,
      "balance_loss_clip": 1.05721426,
      "balance_loss_mlp": 1.02136874,
      "epoch": 0.5194492875608729,
      "flos": 11473352392320.0,
      "grad_norm": 2.7463984416623513,
      "language_loss": 0.93919402,
      "learning_rate": 1.9711790466448854e-06,
      "loss": 0.96135759,
      "num_input_tokens_seen": 93268205,
      "step": 4320,
      "time_per_iteration": 2.645686626434326
    },
    {
      "auxiliary_loss_clip": 0.0113296,
      "auxiliary_loss_mlp": 0.01031342,
      "balance_loss_clip": 1.04853153,
      "balance_loss_mlp": 1.02287817,
      "epoch": 0.5195695304515121,
      "flos": 20338906498560.0,
      "grad_norm": 2.5002726231466137,
      "language_loss": 0.71550894,
      "learning_rate": 1.9704001580888704e-06,
      "loss": 0.73715186,
      "num_input_tokens_seen": 93286945,
      "step": 4321,
      "time_per_iteration": 2.86267352104187
    },
    {
      "auxiliary_loss_clip": 0.01147939,
      "auxiliary_loss_mlp": 0.0076269,
      "balance_loss_clip": 1.0453949,
      "balance_loss_mlp": 1.00021625,
      "epoch": 0.5196897733421512,
      "flos": 20048317470720.0,
      "grad_norm": 2.111072251528371,
      "language_loss": 0.87004125,
      "learning_rate": 1.9696212740231283e-06,
      "loss": 0.88914758,
      "num_input_tokens_seen": 93305595,
      "step": 4322,
      "time_per_iteration": 2.6753649711608887
    },
    {
      "auxiliary_loss_clip": 0.01175361,
      "auxiliary_loss_mlp": 0.01029223,
      "balance_loss_clip": 1.05108881,
      "balance_loss_mlp": 1.02101552,
      "epoch": 0.5198100162327902,
      "flos": 23805507058560.0,
      "grad_norm": 2.345426863982677,
      "language_loss": 0.82136631,
      "learning_rate": 1.9688423945658146e-06,
      "loss": 0.84341216,
      "num_input_tokens_seen": 93326460,
      "step": 4323,
      "time_per_iteration": 2.720966100692749
    },
    {
      "auxiliary_loss_clip": 0.01112175,
      "auxiliary_loss_mlp": 0.01033011,
      "balance_loss_clip": 1.03844762,
      "balance_loss_mlp": 1.02505374,
      "epoch": 0.5199302591234293,
      "flos": 24023951619840.0,
      "grad_norm": 2.3225183175107422,
      "language_loss": 0.71969616,
      "learning_rate": 1.9680635198350845e-06,
      "loss": 0.74114799,
      "num_input_tokens_seen": 93346170,
      "step": 4324,
      "time_per_iteration": 2.7868387699127197
    },
    {
      "auxiliary_loss_clip": 0.01170591,
      "auxiliary_loss_mlp": 0.01028561,
      "balance_loss_clip": 1.04966974,
      "balance_loss_mlp": 1.01951909,
      "epoch": 0.5200505020140684,
      "flos": 26359366095360.0,
      "grad_norm": 5.012314962184714,
      "language_loss": 0.72621071,
      "learning_rate": 1.967284649949093e-06,
      "loss": 0.7482022,
      "num_input_tokens_seen": 93365380,
      "step": 4325,
      "time_per_iteration": 2.737074136734009
    },
    {
      "auxiliary_loss_clip": 0.01136378,
      "auxiliary_loss_mlp": 0.01033084,
      "balance_loss_clip": 1.04363465,
      "balance_loss_mlp": 1.02503157,
      "epoch": 0.5201707449047075,
      "flos": 39604262284800.0,
      "grad_norm": 1.7904806334505594,
      "language_loss": 0.72611487,
      "learning_rate": 1.966505785025994e-06,
      "loss": 0.74780953,
      "num_input_tokens_seen": 93387285,
      "step": 4326,
      "time_per_iteration": 2.907452344894409
    },
    {
      "auxiliary_loss_clip": 0.01143885,
      "auxiliary_loss_mlp": 0.01032754,
      "balance_loss_clip": 1.04999483,
      "balance_loss_mlp": 1.02521396,
      "epoch": 0.5202909877953465,
      "flos": 53682788292480.0,
      "grad_norm": 1.8037083575201058,
      "language_loss": 0.76270509,
      "learning_rate": 1.965726925183941e-06,
      "loss": 0.78447151,
      "num_input_tokens_seen": 93410390,
      "step": 4327,
      "time_per_iteration": 3.934385299682617
    },
    {
      "auxiliary_loss_clip": 0.0118113,
      "auxiliary_loss_mlp": 0.0102987,
      "balance_loss_clip": 1.05100763,
      "balance_loss_mlp": 1.02183557,
      "epoch": 0.5204112306859857,
      "flos": 19537021324800.0,
      "grad_norm": 1.8865015122829725,
      "language_loss": 0.84989059,
      "learning_rate": 1.964948070541087e-06,
      "loss": 0.87200063,
      "num_input_tokens_seen": 93429050,
      "step": 4328,
      "time_per_iteration": 3.580063581466675
    },
    {
      "auxiliary_loss_clip": 0.01158431,
      "auxiliary_loss_mlp": 0.01029003,
      "balance_loss_clip": 1.04705024,
      "balance_loss_mlp": 1.02090263,
      "epoch": 0.5205314735766248,
      "flos": 15304697608320.0,
      "grad_norm": 2.3891939207417168,
      "language_loss": 0.69450378,
      "learning_rate": 1.9641692212155816e-06,
      "loss": 0.71637809,
      "num_input_tokens_seen": 93446815,
      "step": 4329,
      "time_per_iteration": 3.598076343536377
    },
    {
      "auxiliary_loss_clip": 0.01125171,
      "auxiliary_loss_mlp": 0.01029732,
      "balance_loss_clip": 1.04906058,
      "balance_loss_mlp": 1.02135468,
      "epoch": 0.5206517164672638,
      "flos": 59263701160320.0,
      "grad_norm": 2.1421411487096584,
      "language_loss": 0.72495884,
      "learning_rate": 1.9633903773255777e-06,
      "loss": 0.74650794,
      "num_input_tokens_seen": 93469130,
      "step": 4330,
      "time_per_iteration": 3.026902914047241
    },
    {
      "auxiliary_loss_clip": 0.01182787,
      "auxiliary_loss_mlp": 0.01030022,
      "balance_loss_clip": 1.05185056,
      "balance_loss_mlp": 1.02189744,
      "epoch": 0.520771959357903,
      "flos": 26871129118080.0,
      "grad_norm": 1.6255477718591331,
      "language_loss": 0.74670005,
      "learning_rate": 1.9626115389892237e-06,
      "loss": 0.76882815,
      "num_input_tokens_seen": 93489920,
      "step": 4331,
      "time_per_iteration": 2.6687910556793213
    },
    {
      "auxiliary_loss_clip": 0.01150195,
      "auxiliary_loss_mlp": 0.01035521,
      "balance_loss_clip": 1.05098319,
      "balance_loss_mlp": 1.02706659,
      "epoch": 0.520892202248542,
      "flos": 26907075653760.0,
      "grad_norm": 2.0392066531748396,
      "language_loss": 0.85562259,
      "learning_rate": 1.96183270632467e-06,
      "loss": 0.87747973,
      "num_input_tokens_seen": 93509770,
      "step": 4332,
      "time_per_iteration": 3.648815870285034
    },
    {
      "auxiliary_loss_clip": 0.01137319,
      "auxiliary_loss_mlp": 0.00763466,
      "balance_loss_clip": 1.0477457,
      "balance_loss_mlp": 1.00021482,
      "epoch": 0.5210124451391811,
      "flos": 25849434666240.0,
      "grad_norm": 1.6902652007666814,
      "language_loss": 0.79060745,
      "learning_rate": 1.9610538794500644e-06,
      "loss": 0.80961531,
      "num_input_tokens_seen": 93529320,
      "step": 4333,
      "time_per_iteration": 2.8012449741363525
    },
    {
      "auxiliary_loss_clip": 0.01046199,
      "auxiliary_loss_mlp": 0.01005182,
      "balance_loss_clip": 1.01138663,
      "balance_loss_mlp": 1.00375772,
      "epoch": 0.5211326880298203,
      "flos": 70553804319360.0,
      "grad_norm": 0.7742296792610236,
      "language_loss": 0.59436417,
      "learning_rate": 1.9602750584835542e-06,
      "loss": 0.614878,
      "num_input_tokens_seen": 93595255,
      "step": 4334,
      "time_per_iteration": 3.3547070026397705
    },
    {
      "auxiliary_loss_clip": 0.01150987,
      "auxiliary_loss_mlp": 0.01025737,
      "balance_loss_clip": 1.04618144,
      "balance_loss_mlp": 1.01836741,
      "epoch": 0.5212529309204593,
      "flos": 15628898787840.0,
      "grad_norm": 2.062345338681083,
      "language_loss": 0.83150995,
      "learning_rate": 1.959496243543286e-06,
      "loss": 0.85327721,
      "num_input_tokens_seen": 93613135,
      "step": 4335,
      "time_per_iteration": 2.639859676361084
    },
    {
      "auxiliary_loss_clip": 0.01174668,
      "auxiliary_loss_mlp": 0.01024683,
      "balance_loss_clip": 1.05596185,
      "balance_loss_mlp": 1.01701784,
      "epoch": 0.5213731738110984,
      "flos": 26242655829120.0,
      "grad_norm": 2.017832654779342,
      "language_loss": 0.79522741,
      "learning_rate": 1.9587174347474057e-06,
      "loss": 0.81722093,
      "num_input_tokens_seen": 93629645,
      "step": 4336,
      "time_per_iteration": 2.6603150367736816
    },
    {
      "auxiliary_loss_clip": 0.01109566,
      "auxiliary_loss_mlp": 0.01027815,
      "balance_loss_clip": 1.04176474,
      "balance_loss_mlp": 1.0190767,
      "epoch": 0.5214934167017375,
      "flos": 19418407637760.0,
      "grad_norm": 2.9674071410988767,
      "language_loss": 0.82009912,
      "learning_rate": 1.9579386322140574e-06,
      "loss": 0.84147292,
      "num_input_tokens_seen": 93645325,
      "step": 4337,
      "time_per_iteration": 2.690481424331665
    },
    {
      "auxiliary_loss_clip": 0.01187917,
      "auxiliary_loss_mlp": 0.00763831,
      "balance_loss_clip": 1.05446184,
      "balance_loss_mlp": 1.00034642,
      "epoch": 0.5216136595923766,
      "flos": 30955788023040.0,
      "grad_norm": 2.0843467730815646,
      "language_loss": 0.80960405,
      "learning_rate": 1.9571598360613854e-06,
      "loss": 0.82912159,
      "num_input_tokens_seen": 93668200,
      "step": 4338,
      "time_per_iteration": 2.7676608562469482
    },
    {
      "auxiliary_loss_clip": 0.01139934,
      "auxiliary_loss_mlp": 0.01028649,
      "balance_loss_clip": 1.04450655,
      "balance_loss_mlp": 1.02111459,
      "epoch": 0.5217339024830157,
      "flos": 21945047143680.0,
      "grad_norm": 5.202327789918528,
      "language_loss": 0.70378643,
      "learning_rate": 1.956381046407532e-06,
      "loss": 0.72547221,
      "num_input_tokens_seen": 93688495,
      "step": 4339,
      "time_per_iteration": 2.6835906505584717
    },
    {
      "auxiliary_loss_clip": 0.01138393,
      "auxiliary_loss_mlp": 0.01033857,
      "balance_loss_clip": 1.04643941,
      "balance_loss_mlp": 1.02585268,
      "epoch": 0.5218541453736548,
      "flos": 20923209037440.0,
      "grad_norm": 2.8551816841337168,
      "language_loss": 0.85940135,
      "learning_rate": 1.9556022633706394e-06,
      "loss": 0.8811239,
      "num_input_tokens_seen": 93707285,
      "step": 4340,
      "time_per_iteration": 2.737086296081543
    },
    {
      "auxiliary_loss_clip": 0.01147128,
      "auxiliary_loss_mlp": 0.01026259,
      "balance_loss_clip": 1.04916131,
      "balance_loss_mlp": 1.01785469,
      "epoch": 0.5219743882642939,
      "flos": 23951663498880.0,
      "grad_norm": 1.633871796446336,
      "language_loss": 0.7963289,
      "learning_rate": 1.954823487068848e-06,
      "loss": 0.81806278,
      "num_input_tokens_seen": 93727495,
      "step": 4341,
      "time_per_iteration": 2.6377642154693604
    },
    {
      "auxiliary_loss_clip": 0.0116859,
      "auxiliary_loss_mlp": 0.01028326,
      "balance_loss_clip": 1.05377078,
      "balance_loss_mlp": 1.02033854,
      "epoch": 0.5220946311549329,
      "flos": 28799280213120.0,
      "grad_norm": 1.630684455458344,
      "language_loss": 0.81293112,
      "learning_rate": 1.9540447176202976e-06,
      "loss": 0.8349002,
      "num_input_tokens_seen": 93748740,
      "step": 4342,
      "time_per_iteration": 2.721656560897827
    },
    {
      "auxiliary_loss_clip": 0.01065041,
      "auxiliary_loss_mlp": 0.01002506,
      "balance_loss_clip": 1.01378536,
      "balance_loss_mlp": 1.0010041,
      "epoch": 0.5222148740455721,
      "flos": 67189369017600.0,
      "grad_norm": 0.8791994028653938,
      "language_loss": 0.60699868,
      "learning_rate": 1.9532659551431272e-06,
      "loss": 0.62767422,
      "num_input_tokens_seen": 93815770,
      "step": 4343,
      "time_per_iteration": 3.38393497467041
    },
    {
      "auxiliary_loss_clip": 0.01172157,
      "auxiliary_loss_mlp": 0.01027308,
      "balance_loss_clip": 1.05186749,
      "balance_loss_mlp": 1.01953614,
      "epoch": 0.5223351169362112,
      "flos": 61856164339200.0,
      "grad_norm": 1.6187867247611154,
      "language_loss": 0.67715788,
      "learning_rate": 1.9524871997554744e-06,
      "loss": 0.69915253,
      "num_input_tokens_seen": 93843530,
      "step": 4344,
      "time_per_iteration": 3.065627098083496
    },
    {
      "auxiliary_loss_clip": 0.01169458,
      "auxiliary_loss_mlp": 0.01026217,
      "balance_loss_clip": 1.05166841,
      "balance_loss_mlp": 1.01815844,
      "epoch": 0.5224553598268502,
      "flos": 14647388676480.0,
      "grad_norm": 2.053701062657634,
      "language_loss": 0.80906737,
      "learning_rate": 1.951708451575475e-06,
      "loss": 0.83102417,
      "num_input_tokens_seen": 93860595,
      "step": 4345,
      "time_per_iteration": 2.6549623012542725
    },
    {
      "auxiliary_loss_clip": 0.011488,
      "auxiliary_loss_mlp": 0.01028488,
      "balance_loss_clip": 1.04643333,
      "balance_loss_mlp": 1.02028012,
      "epoch": 0.5225756027174894,
      "flos": 14826043946880.0,
      "grad_norm": 2.5589274779245352,
      "language_loss": 0.82239318,
      "learning_rate": 1.9509297107212657e-06,
      "loss": 0.84416604,
      "num_input_tokens_seen": 93877365,
      "step": 4346,
      "time_per_iteration": 2.650106430053711
    },
    {
      "auxiliary_loss_clip": 0.01182573,
      "auxiliary_loss_mlp": 0.01028503,
      "balance_loss_clip": 1.05226314,
      "balance_loss_mlp": 1.02062953,
      "epoch": 0.5226958456081284,
      "flos": 23512009029120.0,
      "grad_norm": 1.5930931813832145,
      "language_loss": 0.7935974,
      "learning_rate": 1.95015097731098e-06,
      "loss": 0.81570816,
      "num_input_tokens_seen": 93896855,
      "step": 4347,
      "time_per_iteration": 2.634014844894409
    },
    {
      "auxiliary_loss_clip": 0.01185438,
      "auxiliary_loss_mlp": 0.01028103,
      "balance_loss_clip": 1.0540669,
      "balance_loss_mlp": 1.01976991,
      "epoch": 0.5228160884987675,
      "flos": 19062928690560.0,
      "grad_norm": 2.038168117958015,
      "language_loss": 0.81896257,
      "learning_rate": 1.949372251462751e-06,
      "loss": 0.84109801,
      "num_input_tokens_seen": 93914270,
      "step": 4348,
      "time_per_iteration": 2.6558678150177
    },
    {
      "auxiliary_loss_clip": 0.01141856,
      "auxiliary_loss_mlp": 0.00762895,
      "balance_loss_clip": 1.04783893,
      "balance_loss_mlp": 1.00032616,
      "epoch": 0.5229363313894067,
      "flos": 21063224252160.0,
      "grad_norm": 1.827963527577306,
      "language_loss": 0.83261585,
      "learning_rate": 1.9485935332947124e-06,
      "loss": 0.85166335,
      "num_input_tokens_seen": 93932180,
      "step": 4349,
      "time_per_iteration": 2.773237943649292
    },
    {
      "auxiliary_loss_clip": 0.01150069,
      "auxiliary_loss_mlp": 0.01025608,
      "balance_loss_clip": 1.04896331,
      "balance_loss_mlp": 1.01812744,
      "epoch": 0.5230565742800457,
      "flos": 14830389492480.0,
      "grad_norm": 2.283413072049421,
      "language_loss": 0.83994591,
      "learning_rate": 1.947814822924993e-06,
      "loss": 0.86170262,
      "num_input_tokens_seen": 93949690,
      "step": 4350,
      "time_per_iteration": 2.6666791439056396
    },
    {
      "auxiliary_loss_clip": 0.01184874,
      "auxiliary_loss_mlp": 0.01028789,
      "balance_loss_clip": 1.05433583,
      "balance_loss_mlp": 1.02098036,
      "epoch": 0.5231768171706848,
      "flos": 25813021253760.0,
      "grad_norm": 2.0816664226167227,
      "language_loss": 0.82921231,
      "learning_rate": 1.9470361204717236e-06,
      "loss": 0.85134894,
      "num_input_tokens_seen": 93968830,
      "step": 4351,
      "time_per_iteration": 2.7068676948547363
    },
    {
      "auxiliary_loss_clip": 0.01144351,
      "auxiliary_loss_mlp": 0.00763875,
      "balance_loss_clip": 1.04693508,
      "balance_loss_mlp": 1.00033426,
      "epoch": 0.5232970600613239,
      "flos": 22743807834240.0,
      "grad_norm": 1.632024740308647,
      "language_loss": 0.80980182,
      "learning_rate": 1.9462574260530326e-06,
      "loss": 0.82888407,
      "num_input_tokens_seen": 93989110,
      "step": 4352,
      "time_per_iteration": 2.755526065826416
    },
    {
      "auxiliary_loss_clip": 0.01156773,
      "auxiliary_loss_mlp": 0.01028895,
      "balance_loss_clip": 1.04615664,
      "balance_loss_mlp": 1.02080035,
      "epoch": 0.523417302951963,
      "flos": 17310703432320.0,
      "grad_norm": 1.961091391878979,
      "language_loss": 0.80835783,
      "learning_rate": 1.9454787397870472e-06,
      "loss": 0.8302145,
      "num_input_tokens_seen": 94006430,
      "step": 4353,
      "time_per_iteration": 3.5751290321350098
    },
    {
      "auxiliary_loss_clip": 0.01106266,
      "auxiliary_loss_mlp": 0.01027946,
      "balance_loss_clip": 1.04615068,
      "balance_loss_mlp": 1.01968527,
      "epoch": 0.523537545842602,
      "flos": 18551740285440.0,
      "grad_norm": 2.193890051644502,
      "language_loss": 0.72261822,
      "learning_rate": 1.944700061791894e-06,
      "loss": 0.74396038,
      "num_input_tokens_seen": 94024825,
      "step": 4354,
      "time_per_iteration": 4.62663722038269
    },
    {
      "auxiliary_loss_clip": 0.01168091,
      "auxiliary_loss_mlp": 0.01024275,
      "balance_loss_clip": 1.05110693,
      "balance_loss_mlp": 1.01643062,
      "epoch": 0.5236577887332411,
      "flos": 19719267955200.0,
      "grad_norm": 2.220604218042637,
      "language_loss": 0.65336794,
      "learning_rate": 1.943921392185698e-06,
      "loss": 0.67529154,
      "num_input_tokens_seen": 94043450,
      "step": 4355,
      "time_per_iteration": 2.715052604675293
    },
    {
      "auxiliary_loss_clip": 0.01159456,
      "auxiliary_loss_mlp": 0.01030677,
      "balance_loss_clip": 1.04918337,
      "balance_loss_mlp": 1.02211809,
      "epoch": 0.5237780316238803,
      "flos": 23550218121600.0,
      "grad_norm": 2.024391726465528,
      "language_loss": 0.76995081,
      "learning_rate": 1.9431427310865814e-06,
      "loss": 0.79185212,
      "num_input_tokens_seen": 94063055,
      "step": 4356,
      "time_per_iteration": 2.685035467147827
    },
    {
      "auxiliary_loss_clip": 0.01120238,
      "auxiliary_loss_mlp": 0.0102618,
      "balance_loss_clip": 1.04327631,
      "balance_loss_mlp": 1.01850855,
      "epoch": 0.5238982745145193,
      "flos": 22491894775680.0,
      "grad_norm": 1.8083952270237356,
      "language_loss": 0.78515196,
      "learning_rate": 1.942364078612667e-06,
      "loss": 0.80661613,
      "num_input_tokens_seen": 94081785,
      "step": 4357,
      "time_per_iteration": 2.7654733657836914
    },
    {
      "auxiliary_loss_clip": 0.01144383,
      "auxiliary_loss_mlp": 0.01030796,
      "balance_loss_clip": 1.0453732,
      "balance_loss_mlp": 1.02257037,
      "epoch": 0.5240185174051584,
      "flos": 27088927234560.0,
      "grad_norm": 1.687236475496652,
      "language_loss": 0.75255591,
      "learning_rate": 1.9415854348820765e-06,
      "loss": 0.77430767,
      "num_input_tokens_seen": 94101635,
      "step": 4358,
      "time_per_iteration": 3.673182964324951
    },
    {
      "auxiliary_loss_clip": 0.01172674,
      "auxiliary_loss_mlp": 0.01033064,
      "balance_loss_clip": 1.04899645,
      "balance_loss_mlp": 1.02453434,
      "epoch": 0.5241387602957975,
      "flos": 22674680110080.0,
      "grad_norm": 2.688943752040632,
      "language_loss": 0.68549097,
      "learning_rate": 1.940806800012929e-06,
      "loss": 0.70754832,
      "num_input_tokens_seen": 94121705,
      "step": 4359,
      "time_per_iteration": 2.6536269187927246
    },
    {
      "auxiliary_loss_clip": 0.01124994,
      "auxiliary_loss_mlp": 0.00763609,
      "balance_loss_clip": 1.04687381,
      "balance_loss_mlp": 1.0003202,
      "epoch": 0.5242590031864366,
      "flos": 40553453134080.0,
      "grad_norm": 2.0971870061398126,
      "language_loss": 0.63679707,
      "learning_rate": 1.9400281741233432e-06,
      "loss": 0.6556831,
      "num_input_tokens_seen": 94146595,
      "step": 4360,
      "time_per_iteration": 2.947805166244507
    },
    {
      "auxiliary_loss_clip": 0.01038248,
      "auxiliary_loss_mlp": 0.01002253,
      "balance_loss_clip": 1.01176298,
      "balance_loss_mlp": 1.00069165,
      "epoch": 0.5243792460770756,
      "flos": 66676313105280.0,
      "grad_norm": 0.6512341783764806,
      "language_loss": 0.52556771,
      "learning_rate": 1.939249557331435e-06,
      "loss": 0.5459727,
      "num_input_tokens_seen": 94212410,
      "step": 4361,
      "time_per_iteration": 3.3155996799468994
    },
    {
      "auxiliary_loss_clip": 0.01150282,
      "auxiliary_loss_mlp": 0.0102616,
      "balance_loss_clip": 1.05023479,
      "balance_loss_mlp": 1.01783061,
      "epoch": 0.5244994889677148,
      "flos": 28183663992960.0,
      "grad_norm": 2.5713229585375084,
      "language_loss": 0.72902334,
      "learning_rate": 1.938470949755321e-06,
      "loss": 0.75078779,
      "num_input_tokens_seen": 94232290,
      "step": 4362,
      "time_per_iteration": 2.798473596572876
    },
    {
      "auxiliary_loss_clip": 0.01043246,
      "auxiliary_loss_mlp": 0.01004028,
      "balance_loss_clip": 1.01013207,
      "balance_loss_mlp": 1.00257921,
      "epoch": 0.5246197318583539,
      "flos": 65950379239680.0,
      "grad_norm": 0.8073214132040221,
      "language_loss": 0.55658919,
      "learning_rate": 1.937692351513115e-06,
      "loss": 0.57706195,
      "num_input_tokens_seen": 94291285,
      "step": 4363,
      "time_per_iteration": 3.254577398300171
    },
    {
      "auxiliary_loss_clip": 0.01174107,
      "auxiliary_loss_mlp": 0.01026547,
      "balance_loss_clip": 1.05096412,
      "balance_loss_mlp": 1.0184288,
      "epoch": 0.5247399747489929,
      "flos": 21033490769280.0,
      "grad_norm": 1.6269043785970971,
      "language_loss": 0.80721205,
      "learning_rate": 1.9369137627229297e-06,
      "loss": 0.82921863,
      "num_input_tokens_seen": 94309685,
      "step": 4364,
      "time_per_iteration": 2.6650190353393555
    },
    {
      "auxiliary_loss_clip": 0.01167283,
      "auxiliary_loss_mlp": 0.0102446,
      "balance_loss_clip": 1.05030978,
      "balance_loss_mlp": 1.01650858,
      "epoch": 0.5248602176396321,
      "flos": 19025940660480.0,
      "grad_norm": 2.461514790756503,
      "language_loss": 0.88343751,
      "learning_rate": 1.936135183502877e-06,
      "loss": 0.90535498,
      "num_input_tokens_seen": 94326985,
      "step": 4365,
      "time_per_iteration": 2.6811435222625732
    },
    {
      "auxiliary_loss_clip": 0.01145904,
      "auxiliary_loss_mlp": 0.01033814,
      "balance_loss_clip": 1.0488106,
      "balance_loss_mlp": 1.02499259,
      "epoch": 0.5249804605302711,
      "flos": 22200084685440.0,
      "grad_norm": 2.036694965775395,
      "language_loss": 0.8025319,
      "learning_rate": 1.935356613971066e-06,
      "loss": 0.82432902,
      "num_input_tokens_seen": 94347645,
      "step": 4366,
      "time_per_iteration": 2.7800798416137695
    },
    {
      "auxiliary_loss_clip": 0.01155993,
      "auxiliary_loss_mlp": 0.00763202,
      "balance_loss_clip": 1.05044365,
      "balance_loss_mlp": 1.00027013,
      "epoch": 0.5251007034209102,
      "flos": 23805686626560.0,
      "grad_norm": 1.7759401931312815,
      "language_loss": 0.76889539,
      "learning_rate": 1.9345780542456047e-06,
      "loss": 0.78808737,
      "num_input_tokens_seen": 94367020,
      "step": 4367,
      "time_per_iteration": 2.7514760494232178
    },
    {
      "auxiliary_loss_clip": 0.01160478,
      "auxiliary_loss_mlp": 0.01029583,
      "balance_loss_clip": 1.04903281,
      "balance_loss_mlp": 1.02107096,
      "epoch": 0.5252209463115494,
      "flos": 23294605962240.0,
      "grad_norm": 2.089035609182676,
      "language_loss": 0.71662956,
      "learning_rate": 1.9337995044446007e-06,
      "loss": 0.73853016,
      "num_input_tokens_seen": 94385860,
      "step": 4368,
      "time_per_iteration": 2.696051597595215
    },
    {
      "auxiliary_loss_clip": 0.01176647,
      "auxiliary_loss_mlp": 0.01033346,
      "balance_loss_clip": 1.05257964,
      "balance_loss_mlp": 1.0247035,
      "epoch": 0.5253411892021884,
      "flos": 19828687760640.0,
      "grad_norm": 2.0711986272542893,
      "language_loss": 0.7996586,
      "learning_rate": 1.9330209646861596e-06,
      "loss": 0.82175851,
      "num_input_tokens_seen": 94405010,
      "step": 4369,
      "time_per_iteration": 2.686502456665039
    },
    {
      "auxiliary_loss_clip": 0.01153936,
      "auxiliary_loss_mlp": 0.01024255,
      "balance_loss_clip": 1.04990518,
      "balance_loss_mlp": 1.01613665,
      "epoch": 0.5254614320928275,
      "flos": 24133730561280.0,
      "grad_norm": 1.6270279663297633,
      "language_loss": 0.78032243,
      "learning_rate": 1.9322424350883843e-06,
      "loss": 0.80210429,
      "num_input_tokens_seen": 94426845,
      "step": 4370,
      "time_per_iteration": 2.7543272972106934
    },
    {
      "auxiliary_loss_clip": 0.01155333,
      "auxiliary_loss_mlp": 0.0102767,
      "balance_loss_clip": 1.04756975,
      "balance_loss_mlp": 1.01991582,
      "epoch": 0.5255816749834666,
      "flos": 24644954880000.0,
      "grad_norm": 1.6592539420659862,
      "language_loss": 0.78915703,
      "learning_rate": 1.931463915769379e-06,
      "loss": 0.81098706,
      "num_input_tokens_seen": 94446960,
      "step": 4371,
      "time_per_iteration": 2.7543201446533203
    },
    {
      "auxiliary_loss_clip": 0.01125078,
      "auxiliary_loss_mlp": 0.01027632,
      "balance_loss_clip": 1.04586434,
      "balance_loss_mlp": 1.01904321,
      "epoch": 0.5257019178741057,
      "flos": 14136595320960.0,
      "grad_norm": 2.1673817482100968,
      "language_loss": 0.74022889,
      "learning_rate": 1.930685406847242e-06,
      "loss": 0.76175594,
      "num_input_tokens_seen": 94461535,
      "step": 4372,
      "time_per_iteration": 2.645777940750122
    },
    {
      "auxiliary_loss_clip": 0.01153001,
      "auxiliary_loss_mlp": 0.01029259,
      "balance_loss_clip": 1.05044758,
      "balance_loss_mlp": 1.02182627,
      "epoch": 0.5258221607647448,
      "flos": 23548961145600.0,
      "grad_norm": 1.4894360566837161,
      "language_loss": 0.81828511,
      "learning_rate": 1.9299069084400734e-06,
      "loss": 0.8401078,
      "num_input_tokens_seen": 94482395,
      "step": 4373,
      "time_per_iteration": 2.728212356567383
    },
    {
      "auxiliary_loss_clip": 0.011413,
      "auxiliary_loss_mlp": 0.01027199,
      "balance_loss_clip": 1.05192196,
      "balance_loss_mlp": 1.01847231,
      "epoch": 0.5259424036553839,
      "flos": 24966103403520.0,
      "grad_norm": 2.19954946984769,
      "language_loss": 0.69679385,
      "learning_rate": 1.9291284206659717e-06,
      "loss": 0.71847886,
      "num_input_tokens_seen": 94500580,
      "step": 4374,
      "time_per_iteration": 2.7593822479248047
    },
    {
      "auxiliary_loss_clip": 0.01187278,
      "auxiliary_loss_mlp": 0.01032698,
      "balance_loss_clip": 1.05484319,
      "balance_loss_mlp": 1.0237211,
      "epoch": 0.526062646546023,
      "flos": 28763908295040.0,
      "grad_norm": 2.111169143140802,
      "language_loss": 0.71761185,
      "learning_rate": 1.928349943643032e-06,
      "loss": 0.73981166,
      "num_input_tokens_seen": 94519680,
      "step": 4375,
      "time_per_iteration": 2.860269784927368
    },
    {
      "auxiliary_loss_clip": 0.01168791,
      "auxiliary_loss_mlp": 0.01029193,
      "balance_loss_clip": 1.0535152,
      "balance_loss_mlp": 1.02092624,
      "epoch": 0.526182889436662,
      "flos": 22821375254400.0,
      "grad_norm": 1.7062473921669083,
      "language_loss": 0.81933182,
      "learning_rate": 1.9275714774893493e-06,
      "loss": 0.84131169,
      "num_input_tokens_seen": 94539135,
      "step": 4376,
      "time_per_iteration": 2.7256040573120117
    },
    {
      "auxiliary_loss_clip": 0.01131616,
      "auxiliary_loss_mlp": 0.01035953,
      "balance_loss_clip": 1.04558349,
      "balance_loss_mlp": 1.0270009,
      "epoch": 0.5263031323273012,
      "flos": 22929466256640.0,
      "grad_norm": 2.3654260252469164,
      "language_loss": 0.72913843,
      "learning_rate": 1.9267930223230154e-06,
      "loss": 0.75081414,
      "num_input_tokens_seen": 94557610,
      "step": 4377,
      "time_per_iteration": 2.7389309406280518
    },
    {
      "auxiliary_loss_clip": 0.01158432,
      "auxiliary_loss_mlp": 0.01029701,
      "balance_loss_clip": 1.05097485,
      "balance_loss_mlp": 1.02135599,
      "epoch": 0.5264233752179402,
      "flos": 17748634049280.0,
      "grad_norm": 2.1659660277051693,
      "language_loss": 0.77929312,
      "learning_rate": 1.9260145782621224e-06,
      "loss": 0.80117446,
      "num_input_tokens_seen": 94575390,
      "step": 4378,
      "time_per_iteration": 2.6920933723449707
    },
    {
      "auxiliary_loss_clip": 0.01153243,
      "auxiliary_loss_mlp": 0.01033912,
      "balance_loss_clip": 1.0524745,
      "balance_loss_mlp": 1.02574027,
      "epoch": 0.5265436181085793,
      "flos": 24421626069120.0,
      "grad_norm": 2.0134292168063164,
      "language_loss": 0.88095212,
      "learning_rate": 1.925236145424758e-06,
      "loss": 0.90282369,
      "num_input_tokens_seen": 94594210,
      "step": 4379,
      "time_per_iteration": 3.6833786964416504
    },
    {
      "auxiliary_loss_clip": 0.01066381,
      "auxiliary_loss_mlp": 0.01004472,
      "balance_loss_clip": 1.01203322,
      "balance_loss_mlp": 1.00285089,
      "epoch": 0.5266638609992185,
      "flos": 69207298156800.0,
      "grad_norm": 0.6949042958182249,
      "language_loss": 0.57578075,
      "learning_rate": 1.924457723929012e-06,
      "loss": 0.59648925,
      "num_input_tokens_seen": 94665020,
      "step": 4380,
      "time_per_iteration": 4.267938852310181
    },
    {
      "auxiliary_loss_clip": 0.01170852,
      "auxiliary_loss_mlp": 0.01024177,
      "balance_loss_clip": 1.05234742,
      "balance_loss_mlp": 1.01609433,
      "epoch": 0.5267841038898575,
      "flos": 20738699850240.0,
      "grad_norm": 1.4522060553568614,
      "language_loss": 0.82875341,
      "learning_rate": 1.9236793138929685e-06,
      "loss": 0.85070366,
      "num_input_tokens_seen": 94684290,
      "step": 4381,
      "time_per_iteration": 3.551720142364502
    },
    {
      "auxiliary_loss_clip": 0.01174502,
      "auxiliary_loss_mlp": 0.01026156,
      "balance_loss_clip": 1.05168819,
      "balance_loss_mlp": 1.01774287,
      "epoch": 0.5269043467804966,
      "flos": 17234392988160.0,
      "grad_norm": 2.061035938917171,
      "language_loss": 0.81265992,
      "learning_rate": 1.9229009154347133e-06,
      "loss": 0.83466649,
      "num_input_tokens_seen": 94701880,
      "step": 4382,
      "time_per_iteration": 2.671293020248413
    },
    {
      "auxiliary_loss_clip": 0.01109695,
      "auxiliary_loss_mlp": 0.007629,
      "balance_loss_clip": 1.04307973,
      "balance_loss_mlp": 1.00027823,
      "epoch": 0.5270245896711357,
      "flos": 18223157646720.0,
      "grad_norm": 2.0914978789530445,
      "language_loss": 0.80632377,
      "learning_rate": 1.922122528672327e-06,
      "loss": 0.8250497,
      "num_input_tokens_seen": 94720545,
      "step": 4383,
      "time_per_iteration": 2.689232587814331
    },
    {
      "auxiliary_loss_clip": 0.01181097,
      "auxiliary_loss_mlp": 0.01030657,
      "balance_loss_clip": 1.05194163,
      "balance_loss_mlp": 1.02258086,
      "epoch": 0.5271448325617748,
      "flos": 21287558643840.0,
      "grad_norm": 2.2640248823425346,
      "language_loss": 0.78788418,
      "learning_rate": 1.9213441537238914e-06,
      "loss": 0.81000173,
      "num_input_tokens_seen": 94737420,
      "step": 4384,
      "time_per_iteration": 2.666966676712036
    },
    {
      "auxiliary_loss_clip": 0.01025582,
      "auxiliary_loss_mlp": 0.01001861,
      "balance_loss_clip": 1.01026821,
      "balance_loss_mlp": 1.00050187,
      "epoch": 0.5272650754524139,
      "flos": 65495497403520.0,
      "grad_norm": 0.8306524455551767,
      "language_loss": 0.57367992,
      "learning_rate": 1.920565790707485e-06,
      "loss": 0.59395432,
      "num_input_tokens_seen": 94802810,
      "step": 4385,
      "time_per_iteration": 4.380992889404297
    },
    {
      "auxiliary_loss_clip": 0.01137732,
      "auxiliary_loss_mlp": 0.0102936,
      "balance_loss_clip": 1.04746962,
      "balance_loss_mlp": 1.02027059,
      "epoch": 0.527385318343053,
      "flos": 19676426008320.0,
      "grad_norm": 1.9740825351373121,
      "language_loss": 0.65966463,
      "learning_rate": 1.9197874397411853e-06,
      "loss": 0.68133551,
      "num_input_tokens_seen": 94819440,
      "step": 4386,
      "time_per_iteration": 3.1066417694091797
    },
    {
      "auxiliary_loss_clip": 0.01142,
      "auxiliary_loss_mlp": 0.01026374,
      "balance_loss_clip": 1.04508615,
      "balance_loss_mlp": 1.0180589,
      "epoch": 0.5275055612336921,
      "flos": 12712018947840.0,
      "grad_norm": 3.043851194874605,
      "language_loss": 0.67320013,
      "learning_rate": 1.919009100943067e-06,
      "loss": 0.69488394,
      "num_input_tokens_seen": 94835130,
      "step": 4387,
      "time_per_iteration": 2.664076566696167
    },
    {
      "auxiliary_loss_clip": 0.01136227,
      "auxiliary_loss_mlp": 0.01036837,
      "balance_loss_clip": 1.04958749,
      "balance_loss_mlp": 1.02806306,
      "epoch": 0.5276258041243311,
      "flos": 17749029098880.0,
      "grad_norm": 1.9650062244272355,
      "language_loss": 0.6566239,
      "learning_rate": 1.9182307744312043e-06,
      "loss": 0.67835456,
      "num_input_tokens_seen": 94852235,
      "step": 4388,
      "time_per_iteration": 2.693253993988037
    },
    {
      "auxiliary_loss_clip": 0.01157511,
      "auxiliary_loss_mlp": 0.01029489,
      "balance_loss_clip": 1.04867744,
      "balance_loss_mlp": 1.02126324,
      "epoch": 0.5277460470149702,
      "flos": 22710447077760.0,
      "grad_norm": 1.8297587243454958,
      "language_loss": 0.76435232,
      "learning_rate": 1.9174524603236676e-06,
      "loss": 0.78622234,
      "num_input_tokens_seen": 94871185,
      "step": 4389,
      "time_per_iteration": 2.720808744430542
    },
    {
      "auxiliary_loss_clip": 0.01155141,
      "auxiliary_loss_mlp": 0.01034464,
      "balance_loss_clip": 1.04931474,
      "balance_loss_mlp": 1.02648902,
      "epoch": 0.5278662899056094,
      "flos": 19902699734400.0,
      "grad_norm": 1.9241423295290516,
      "language_loss": 0.76452714,
      "learning_rate": 1.916674158738527e-06,
      "loss": 0.78642315,
      "num_input_tokens_seen": 94890090,
      "step": 4390,
      "time_per_iteration": 2.6828348636627197
    },
    {
      "auxiliary_loss_clip": 0.01138739,
      "auxiliary_loss_mlp": 0.00764428,
      "balance_loss_clip": 1.04972649,
      "balance_loss_mlp": 1.00026357,
      "epoch": 0.5279865327962484,
      "flos": 18005215875840.0,
      "grad_norm": 1.9769914577373713,
      "language_loss": 0.60555744,
      "learning_rate": 1.9158958697938506e-06,
      "loss": 0.62458909,
      "num_input_tokens_seen": 94908470,
      "step": 4391,
      "time_per_iteration": 2.8245067596435547
    },
    {
      "auxiliary_loss_clip": 0.01149406,
      "auxiliary_loss_mlp": 0.01030181,
      "balance_loss_clip": 1.04926538,
      "balance_loss_mlp": 1.02186024,
      "epoch": 0.5281067756868875,
      "flos": 15924443892480.0,
      "grad_norm": 2.4181960079688074,
      "language_loss": 0.85856164,
      "learning_rate": 1.9151175936077032e-06,
      "loss": 0.8803575,
      "num_input_tokens_seen": 94923440,
      "step": 4392,
      "time_per_iteration": 2.66951584815979
    },
    {
      "auxiliary_loss_clip": 0.01167205,
      "auxiliary_loss_mlp": 0.01024641,
      "balance_loss_clip": 1.05156732,
      "balance_loss_mlp": 1.0164597,
      "epoch": 0.5282270185775266,
      "flos": 19426488197760.0,
      "grad_norm": 1.7120341317480272,
      "language_loss": 0.7950784,
      "learning_rate": 1.9143393302981507e-06,
      "loss": 0.81699687,
      "num_input_tokens_seen": 94941125,
      "step": 4393,
      "time_per_iteration": 2.6540300846099854
    },
    {
      "auxiliary_loss_clip": 0.01158701,
      "auxiliary_loss_mlp": 0.01030535,
      "balance_loss_clip": 1.05178297,
      "balance_loss_mlp": 1.02223849,
      "epoch": 0.5283472614681657,
      "flos": 16399613934720.0,
      "grad_norm": 2.053135989731018,
      "language_loss": 0.83608294,
      "learning_rate": 1.913561079983252e-06,
      "loss": 0.8579753,
      "num_input_tokens_seen": 94959950,
      "step": 4394,
      "time_per_iteration": 2.6335043907165527
    },
    {
      "auxiliary_loss_clip": 0.01161579,
      "auxiliary_loss_mlp": 0.01039394,
      "balance_loss_clip": 1.04936624,
      "balance_loss_mlp": 1.03010154,
      "epoch": 0.5284675043588047,
      "flos": 26760524163840.0,
      "grad_norm": 2.652303183336842,
      "language_loss": 0.75128841,
      "learning_rate": 1.9127828427810693e-06,
      "loss": 0.77329808,
      "num_input_tokens_seen": 94980515,
      "step": 4395,
      "time_per_iteration": 2.724336862564087
    },
    {
      "auxiliary_loss_clip": 0.01149058,
      "auxiliary_loss_mlp": 0.0103109,
      "balance_loss_clip": 1.04860711,
      "balance_loss_mlp": 1.02271557,
      "epoch": 0.5285877472494439,
      "flos": 19899898473600.0,
      "grad_norm": 2.1562572205101924,
      "language_loss": 0.81049806,
      "learning_rate": 1.9120046188096607e-06,
      "loss": 0.83229959,
      "num_input_tokens_seen": 94998560,
      "step": 4396,
      "time_per_iteration": 2.732729196548462
    },
    {
      "auxiliary_loss_clip": 0.01154072,
      "auxiliary_loss_mlp": 0.01028881,
      "balance_loss_clip": 1.05323601,
      "balance_loss_mlp": 1.02087867,
      "epoch": 0.528707990140083,
      "flos": 20011257613440.0,
      "grad_norm": 3.928881035084848,
      "language_loss": 0.74568033,
      "learning_rate": 1.9112264081870804e-06,
      "loss": 0.76750982,
      "num_input_tokens_seen": 95016950,
      "step": 4397,
      "time_per_iteration": 2.6809356212615967
    },
    {
      "auxiliary_loss_clip": 0.01138544,
      "auxiliary_loss_mlp": 0.01031983,
      "balance_loss_clip": 1.04953945,
      "balance_loss_mlp": 1.02332234,
      "epoch": 0.528828233030722,
      "flos": 20667956014080.0,
      "grad_norm": 2.5484886036110703,
      "language_loss": 0.75921869,
      "learning_rate": 1.9104482110313843e-06,
      "loss": 0.78092396,
      "num_input_tokens_seen": 95036540,
      "step": 4398,
      "time_per_iteration": 2.770430564880371
    },
    {
      "auxiliary_loss_clip": 0.01168589,
      "auxiliary_loss_mlp": 0.01033851,
      "balance_loss_clip": 1.05103195,
      "balance_loss_mlp": 1.02541089,
      "epoch": 0.5289484759213612,
      "flos": 25192448956800.0,
      "grad_norm": 1.9184905142822517,
      "language_loss": 0.74196064,
      "learning_rate": 1.909670027460623e-06,
      "loss": 0.76398504,
      "num_input_tokens_seen": 95053840,
      "step": 4399,
      "time_per_iteration": 2.6679933071136475
    },
    {
      "auxiliary_loss_clip": 0.01170924,
      "auxiliary_loss_mlp": 0.01029216,
      "balance_loss_clip": 1.0532136,
      "balance_loss_mlp": 1.0209192,
      "epoch": 0.5290687188120002,
      "flos": 31139255715840.0,
      "grad_norm": 1.914597062075373,
      "language_loss": 0.71729064,
      "learning_rate": 1.908891857592847e-06,
      "loss": 0.73929203,
      "num_input_tokens_seen": 95074910,
      "step": 4400,
      "time_per_iteration": 2.774463415145874
    },
    {
      "auxiliary_loss_clip": 0.01135038,
      "auxiliary_loss_mlp": 0.01021995,
      "balance_loss_clip": 1.04904532,
      "balance_loss_mlp": 1.01342392,
      "epoch": 0.5291889617026393,
      "flos": 20119851406080.0,
      "grad_norm": 2.252218094286598,
      "language_loss": 0.90295923,
      "learning_rate": 1.9081137015461034e-06,
      "loss": 0.92452955,
      "num_input_tokens_seen": 95090985,
      "step": 4401,
      "time_per_iteration": 2.7129905223846436
    },
    {
      "auxiliary_loss_clip": 0.01120178,
      "auxiliary_loss_mlp": 0.01031757,
      "balance_loss_clip": 1.04722142,
      "balance_loss_mlp": 1.02326286,
      "epoch": 0.5293092045932785,
      "flos": 19643747610240.0,
      "grad_norm": 1.858437256453434,
      "language_loss": 0.90481651,
      "learning_rate": 1.9073355594384383e-06,
      "loss": 0.92633581,
      "num_input_tokens_seen": 95109225,
      "step": 4402,
      "time_per_iteration": 2.790578603744507
    },
    {
      "auxiliary_loss_clip": 0.01133313,
      "auxiliary_loss_mlp": 0.0102652,
      "balance_loss_clip": 1.0469749,
      "balance_loss_mlp": 1.0178411,
      "epoch": 0.5294294474839175,
      "flos": 24317736958080.0,
      "grad_norm": 2.020619459800134,
      "language_loss": 0.8036145,
      "learning_rate": 1.906557431387895e-06,
      "loss": 0.82521284,
      "num_input_tokens_seen": 95128215,
      "step": 4403,
      "time_per_iteration": 2.721859931945801
    },
    {
      "auxiliary_loss_clip": 0.011381,
      "auxiliary_loss_mlp": 0.01027817,
      "balance_loss_clip": 1.05206418,
      "balance_loss_mlp": 1.01932371,
      "epoch": 0.5295496903745566,
      "flos": 18875941464960.0,
      "grad_norm": 2.0629777318904523,
      "language_loss": 0.78558904,
      "learning_rate": 1.905779317512516e-06,
      "loss": 0.80724823,
      "num_input_tokens_seen": 95145760,
      "step": 4404,
      "time_per_iteration": 2.7082860469818115
    },
    {
      "auxiliary_loss_clip": 0.01170443,
      "auxiliary_loss_mlp": 0.01029187,
      "balance_loss_clip": 1.05320728,
      "balance_loss_mlp": 1.02147388,
      "epoch": 0.5296699332651957,
      "flos": 20923101296640.0,
      "grad_norm": 1.9335089530208531,
      "language_loss": 0.80417323,
      "learning_rate": 1.9050012179303385e-06,
      "loss": 0.82616955,
      "num_input_tokens_seen": 95164270,
      "step": 4405,
      "time_per_iteration": 3.669764518737793
    },
    {
      "auxiliary_loss_clip": 0.01171603,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.05097342,
      "balance_loss_mlp": 1.0171597,
      "epoch": 0.5297901761558348,
      "flos": 22046745525120.0,
      "grad_norm": 2.096785866707623,
      "language_loss": 0.68993914,
      "learning_rate": 1.904223132759401e-06,
      "loss": 0.71191108,
      "num_input_tokens_seen": 95182870,
      "step": 4406,
      "time_per_iteration": 4.579307556152344
    },
    {
      "auxiliary_loss_clip": 0.0117079,
      "auxiliary_loss_mlp": 0.01031436,
      "balance_loss_clip": 1.05150032,
      "balance_loss_mlp": 1.02325201,
      "epoch": 0.5299104190464738,
      "flos": 21798495653760.0,
      "grad_norm": 2.572511257112462,
      "language_loss": 0.68715513,
      "learning_rate": 1.9034450621177383e-06,
      "loss": 0.70917749,
      "num_input_tokens_seen": 95201190,
      "step": 4407,
      "time_per_iteration": 2.6572837829589844
    },
    {
      "auxiliary_loss_clip": 0.01171712,
      "auxiliary_loss_mlp": 0.01030504,
      "balance_loss_clip": 1.05432844,
      "balance_loss_mlp": 1.02161121,
      "epoch": 0.530030661937113,
      "flos": 14720790119040.0,
      "grad_norm": 2.1399127389271433,
      "language_loss": 0.70489264,
      "learning_rate": 1.9026670061233824e-06,
      "loss": 0.72691476,
      "num_input_tokens_seen": 95218625,
      "step": 4408,
      "time_per_iteration": 2.70247220993042
    },
    {
      "auxiliary_loss_clip": 0.01151804,
      "auxiliary_loss_mlp": 0.01029128,
      "balance_loss_clip": 1.05013728,
      "balance_loss_mlp": 1.021052,
      "epoch": 0.5301509048277521,
      "flos": 21251504367360.0,
      "grad_norm": 1.6838791657119094,
      "language_loss": 0.80763096,
      "learning_rate": 1.901888964894365e-06,
      "loss": 0.82944036,
      "num_input_tokens_seen": 95237665,
      "step": 4409,
      "time_per_iteration": 2.7121400833129883
    },
    {
      "auxiliary_loss_clip": 0.01186212,
      "auxiliary_loss_mlp": 0.0102942,
      "balance_loss_clip": 1.05323792,
      "balance_loss_mlp": 1.02128994,
      "epoch": 0.5302711477183911,
      "flos": 25957058791680.0,
      "grad_norm": 2.3018645924248173,
      "language_loss": 0.67912662,
      "learning_rate": 1.9011109385487134e-06,
      "loss": 0.70128286,
      "num_input_tokens_seen": 95258915,
      "step": 4410,
      "time_per_iteration": 3.552081823348999
    },
    {
      "auxiliary_loss_clip": 0.01186283,
      "auxiliary_loss_mlp": 0.01033024,
      "balance_loss_clip": 1.05359268,
      "balance_loss_mlp": 1.02480459,
      "epoch": 0.5303913906090303,
      "flos": 22273126992000.0,
      "grad_norm": 2.2629586042027947,
      "language_loss": 0.66364986,
      "learning_rate": 1.900332927204454e-06,
      "loss": 0.68584293,
      "num_input_tokens_seen": 95277365,
      "step": 4411,
      "time_per_iteration": 2.594653367996216
    },
    {
      "auxiliary_loss_clip": 0.01165185,
      "auxiliary_loss_mlp": 0.01033461,
      "balance_loss_clip": 1.05163598,
      "balance_loss_mlp": 1.02434754,
      "epoch": 0.5305116334996693,
      "flos": 24936010784640.0,
      "grad_norm": 1.8027620368650092,
      "language_loss": 0.7670294,
      "learning_rate": 1.8995549309796097e-06,
      "loss": 0.78901589,
      "num_input_tokens_seen": 95296670,
      "step": 4412,
      "time_per_iteration": 2.6839632987976074
    },
    {
      "auxiliary_loss_clip": 0.01177259,
      "auxiliary_loss_mlp": 0.01030637,
      "balance_loss_clip": 1.05464244,
      "balance_loss_mlp": 1.02227426,
      "epoch": 0.5306318763903084,
      "flos": 20189338266240.0,
      "grad_norm": 2.059473482375525,
      "language_loss": 0.76695418,
      "learning_rate": 1.8987769499922028e-06,
      "loss": 0.78903311,
      "num_input_tokens_seen": 95315640,
      "step": 4413,
      "time_per_iteration": 2.69919753074646
    },
    {
      "auxiliary_loss_clip": 0.01170063,
      "auxiliary_loss_mlp": 0.00763491,
      "balance_loss_clip": 1.05282259,
      "balance_loss_mlp": 1.00030041,
      "epoch": 0.5307521192809476,
      "flos": 20266366982400.0,
      "grad_norm": 2.1881085489188203,
      "language_loss": 0.70904779,
      "learning_rate": 1.897998984360252e-06,
      "loss": 0.7283833,
      "num_input_tokens_seen": 95334610,
      "step": 4414,
      "time_per_iteration": 2.665048599243164
    },
    {
      "auxiliary_loss_clip": 0.01152908,
      "auxiliary_loss_mlp": 0.01031252,
      "balance_loss_clip": 1.04969096,
      "balance_loss_mlp": 1.02313352,
      "epoch": 0.5308723621715866,
      "flos": 28844276976000.0,
      "grad_norm": 1.482583483614544,
      "language_loss": 0.78710443,
      "learning_rate": 1.897221034201775e-06,
      "loss": 0.80894601,
      "num_input_tokens_seen": 95358350,
      "step": 4415,
      "time_per_iteration": 2.737605094909668
    },
    {
      "auxiliary_loss_clip": 0.01142391,
      "auxiliary_loss_mlp": 0.01025286,
      "balance_loss_clip": 1.04715788,
      "balance_loss_mlp": 1.01757956,
      "epoch": 0.5309926050622257,
      "flos": 27457766040960.0,
      "grad_norm": 1.4817277788063077,
      "language_loss": 0.6648441,
      "learning_rate": 1.8964430996347842e-06,
      "loss": 0.68652081,
      "num_input_tokens_seen": 95379900,
      "step": 4416,
      "time_per_iteration": 2.7136640548706055
    },
    {
      "auxiliary_loss_clip": 0.01155662,
      "auxiliary_loss_mlp": 0.0102787,
      "balance_loss_clip": 1.04981947,
      "balance_loss_mlp": 1.01901841,
      "epoch": 0.5311128479528648,
      "flos": 20514545026560.0,
      "grad_norm": 1.6911891074210676,
      "language_loss": 0.8220048,
      "learning_rate": 1.8956651807772931e-06,
      "loss": 0.84384012,
      "num_input_tokens_seen": 95397935,
      "step": 4417,
      "time_per_iteration": 2.716836452484131
    },
    {
      "auxiliary_loss_clip": 0.01166874,
      "auxiliary_loss_mlp": 0.01025656,
      "balance_loss_clip": 1.05099237,
      "balance_loss_mlp": 1.01790714,
      "epoch": 0.5312330908435039,
      "flos": 21397660807680.0,
      "grad_norm": 1.829053089005621,
      "language_loss": 0.84315687,
      "learning_rate": 1.8948872777473115e-06,
      "loss": 0.86508214,
      "num_input_tokens_seen": 95415890,
      "step": 4418,
      "time_per_iteration": 2.619044542312622
    },
    {
      "auxiliary_loss_clip": 0.01156238,
      "auxiliary_loss_mlp": 0.01022893,
      "balance_loss_clip": 1.05198812,
      "balance_loss_mlp": 1.0148406,
      "epoch": 0.531353333734143,
      "flos": 24717350741760.0,
      "grad_norm": 1.6727673281825883,
      "language_loss": 0.63380563,
      "learning_rate": 1.8941093906628458e-06,
      "loss": 0.65559697,
      "num_input_tokens_seen": 95433675,
      "step": 4419,
      "time_per_iteration": 2.76678204536438
    },
    {
      "auxiliary_loss_clip": 0.01149875,
      "auxiliary_loss_mlp": 0.0102962,
      "balance_loss_clip": 1.04923999,
      "balance_loss_mlp": 1.02162123,
      "epoch": 0.531473576624782,
      "flos": 30480689808000.0,
      "grad_norm": 1.6773893156424067,
      "language_loss": 0.70996773,
      "learning_rate": 1.893331519641902e-06,
      "loss": 0.73176265,
      "num_input_tokens_seen": 95455820,
      "step": 4420,
      "time_per_iteration": 2.8477962017059326
    },
    {
      "auxiliary_loss_clip": 0.01129648,
      "auxiliary_loss_mlp": 0.01031061,
      "balance_loss_clip": 1.04522991,
      "balance_loss_mlp": 1.02222729,
      "epoch": 0.5315938195154212,
      "flos": 23002975440000.0,
      "grad_norm": 2.18397345109428,
      "language_loss": 0.73684978,
      "learning_rate": 1.8925536648024815e-06,
      "loss": 0.75845683,
      "num_input_tokens_seen": 95473240,
      "step": 4421,
      "time_per_iteration": 2.751095771789551
    },
    {
      "auxiliary_loss_clip": 0.0118688,
      "auxiliary_loss_mlp": 0.01027172,
      "balance_loss_clip": 1.05488527,
      "balance_loss_mlp": 1.01955414,
      "epoch": 0.5317140624060602,
      "flos": 22748584343040.0,
      "grad_norm": 2.199579337671364,
      "language_loss": 0.7597177,
      "learning_rate": 1.8917758262625849e-06,
      "loss": 0.78185821,
      "num_input_tokens_seen": 95493480,
      "step": 4422,
      "time_per_iteration": 2.5881690979003906
    },
    {
      "auxiliary_loss_clip": 0.01152644,
      "auxiliary_loss_mlp": 0.01028251,
      "balance_loss_clip": 1.05159187,
      "balance_loss_mlp": 1.02058291,
      "epoch": 0.5318343052966993,
      "flos": 22821087945600.0,
      "grad_norm": 1.7620824708031575,
      "language_loss": 0.81021965,
      "learning_rate": 1.8909980041402089e-06,
      "loss": 0.83202863,
      "num_input_tokens_seen": 95512075,
      "step": 4423,
      "time_per_iteration": 2.7035598754882812
    },
    {
      "auxiliary_loss_clip": 0.01162344,
      "auxiliary_loss_mlp": 0.01028501,
      "balance_loss_clip": 1.04918754,
      "balance_loss_mlp": 1.01980495,
      "epoch": 0.5319545481873384,
      "flos": 13626089274240.0,
      "grad_norm": 2.7121525223040317,
      "language_loss": 0.65725708,
      "learning_rate": 1.8902201985533494e-06,
      "loss": 0.67916554,
      "num_input_tokens_seen": 95529340,
      "step": 4424,
      "time_per_iteration": 2.577834129333496
    },
    {
      "auxiliary_loss_clip": 0.01155704,
      "auxiliary_loss_mlp": 0.01028781,
      "balance_loss_clip": 1.05215526,
      "balance_loss_mlp": 1.02022767,
      "epoch": 0.5320747910779775,
      "flos": 22162522037760.0,
      "grad_norm": 2.467131157895143,
      "language_loss": 0.74905348,
      "learning_rate": 1.8894424096199983e-06,
      "loss": 0.77089834,
      "num_input_tokens_seen": 95548545,
      "step": 4425,
      "time_per_iteration": 2.6902053356170654
    },
    {
      "auxiliary_loss_clip": 0.01173248,
      "auxiliary_loss_mlp": 0.01031864,
      "balance_loss_clip": 1.0546999,
      "balance_loss_mlp": 1.02365029,
      "epoch": 0.5321950339686166,
      "flos": 18588081870720.0,
      "grad_norm": 1.9860907063156943,
      "language_loss": 0.85985887,
      "learning_rate": 1.8886646374581463e-06,
      "loss": 0.88190997,
      "num_input_tokens_seen": 95567770,
      "step": 4426,
      "time_per_iteration": 2.664602041244507
    },
    {
      "auxiliary_loss_clip": 0.01166385,
      "auxiliary_loss_mlp": 0.01033825,
      "balance_loss_clip": 1.04935455,
      "balance_loss_mlp": 1.02509308,
      "epoch": 0.5323152768592557,
      "flos": 22856818999680.0,
      "grad_norm": 1.7533541527709298,
      "language_loss": 0.71279037,
      "learning_rate": 1.8878868821857795e-06,
      "loss": 0.73479247,
      "num_input_tokens_seen": 95587420,
      "step": 4427,
      "time_per_iteration": 2.630034923553467
    },
    {
      "auxiliary_loss_clip": 0.01122433,
      "auxiliary_loss_mlp": 0.01027543,
      "balance_loss_clip": 1.04486454,
      "balance_loss_mlp": 1.01891279,
      "epoch": 0.5324355197498948,
      "flos": 33948690998400.0,
      "grad_norm": 2.308206861050779,
      "language_loss": 0.75463855,
      "learning_rate": 1.8871091439208838e-06,
      "loss": 0.77613831,
      "num_input_tokens_seen": 95609030,
      "step": 4428,
      "time_per_iteration": 2.8556265830993652
    },
    {
      "auxiliary_loss_clip": 0.01124604,
      "auxiliary_loss_mlp": 0.01033957,
      "balance_loss_clip": 1.04716039,
      "balance_loss_mlp": 1.02511144,
      "epoch": 0.5325557626405338,
      "flos": 23256720092160.0,
      "grad_norm": 2.4529153823725505,
      "language_loss": 0.7719605,
      "learning_rate": 1.8863314227814414e-06,
      "loss": 0.7935462,
      "num_input_tokens_seen": 95627340,
      "step": 4429,
      "time_per_iteration": 2.743635892868042
    },
    {
      "auxiliary_loss_clip": 0.01173608,
      "auxiliary_loss_mlp": 0.01030988,
      "balance_loss_clip": 1.05248094,
      "balance_loss_mlp": 1.02233911,
      "epoch": 0.532676005531173,
      "flos": 26718687797760.0,
      "grad_norm": 2.752643607520451,
      "language_loss": 0.49089339,
      "learning_rate": 1.8855537188854313e-06,
      "loss": 0.51293939,
      "num_input_tokens_seen": 95646315,
      "step": 4430,
      "time_per_iteration": 2.712009906768799
    },
    {
      "auxiliary_loss_clip": 0.01173296,
      "auxiliary_loss_mlp": 0.01029126,
      "balance_loss_clip": 1.05087829,
      "balance_loss_mlp": 1.02149642,
      "epoch": 0.5327962484218121,
      "flos": 17894610921600.0,
      "grad_norm": 2.0093079489141816,
      "language_loss": 0.78212357,
      "learning_rate": 1.8847760323508315e-06,
      "loss": 0.80414772,
      "num_input_tokens_seen": 95665220,
      "step": 4431,
      "time_per_iteration": 3.576850175857544
    },
    {
      "auxiliary_loss_clip": 0.01151817,
      "auxiliary_loss_mlp": 0.0103114,
      "balance_loss_clip": 1.05039275,
      "balance_loss_mlp": 1.02327859,
      "epoch": 0.5329164913124511,
      "flos": 17925385898880.0,
      "grad_norm": 1.710311822583488,
      "language_loss": 0.76160765,
      "learning_rate": 1.883998363295616e-06,
      "loss": 0.78343719,
      "num_input_tokens_seen": 95682700,
      "step": 4432,
      "time_per_iteration": 4.582221031188965
    },
    {
      "auxiliary_loss_clip": 0.01055579,
      "auxiliary_loss_mlp": 0.01003699,
      "balance_loss_clip": 1.01461172,
      "balance_loss_mlp": 1.00213742,
      "epoch": 0.5330367342030903,
      "flos": 57254178781440.0,
      "grad_norm": 0.8747379982611112,
      "language_loss": 0.62657863,
      "learning_rate": 1.8832207118377565e-06,
      "loss": 0.64717138,
      "num_input_tokens_seen": 95738070,
      "step": 4433,
      "time_per_iteration": 3.1660594940185547
    },
    {
      "auxiliary_loss_clip": 0.01182488,
      "auxiliary_loss_mlp": 0.01023144,
      "balance_loss_clip": 1.05256152,
      "balance_loss_mlp": 1.01553583,
      "epoch": 0.5331569770937293,
      "flos": 17420518287360.0,
      "grad_norm": 2.5414307819962403,
      "language_loss": 0.69509089,
      "learning_rate": 1.882443078095222e-06,
      "loss": 0.71714723,
      "num_input_tokens_seen": 95756950,
      "step": 4434,
      "time_per_iteration": 2.629270315170288
    },
    {
      "auxiliary_loss_clip": 0.01036895,
      "auxiliary_loss_mlp": 0.01002943,
      "balance_loss_clip": 1.01392698,
      "balance_loss_mlp": 1.0013932,
      "epoch": 0.5332772199843684,
      "flos": 56750783627520.0,
      "grad_norm": 0.853604326992352,
      "language_loss": 0.66742158,
      "learning_rate": 1.8816654621859794e-06,
      "loss": 0.68781996,
      "num_input_tokens_seen": 95816615,
      "step": 4435,
      "time_per_iteration": 3.2092788219451904
    },
    {
      "auxiliary_loss_clip": 0.01181056,
      "auxiliary_loss_mlp": 0.01028414,
      "balance_loss_clip": 1.05097497,
      "balance_loss_mlp": 1.02023017,
      "epoch": 0.5333974628750076,
      "flos": 18697753071360.0,
      "grad_norm": 2.4864945874270874,
      "language_loss": 0.72649187,
      "learning_rate": 1.8808878642279915e-06,
      "loss": 0.7485866,
      "num_input_tokens_seen": 95832020,
      "step": 4436,
      "time_per_iteration": 3.5043752193450928
    },
    {
      "auxiliary_loss_clip": 0.01139375,
      "auxiliary_loss_mlp": 0.01030927,
      "balance_loss_clip": 1.04294443,
      "balance_loss_mlp": 1.02186763,
      "epoch": 0.5335177057656466,
      "flos": 23805507058560.0,
      "grad_norm": 2.1202774298500593,
      "language_loss": 0.65186739,
      "learning_rate": 1.8801102843392209e-06,
      "loss": 0.67357039,
      "num_input_tokens_seen": 95851425,
      "step": 4437,
      "time_per_iteration": 2.700889825820923
    },
    {
      "auxiliary_loss_clip": 0.01137082,
      "auxiliary_loss_mlp": 0.01032554,
      "balance_loss_clip": 1.04507887,
      "balance_loss_mlp": 1.02423882,
      "epoch": 0.5336379486562857,
      "flos": 25078683605760.0,
      "grad_norm": 1.5167107242185394,
      "language_loss": 0.85015595,
      "learning_rate": 1.8793327226376238e-06,
      "loss": 0.87185228,
      "num_input_tokens_seen": 95870745,
      "step": 4438,
      "time_per_iteration": 2.7530484199523926
    },
    {
      "auxiliary_loss_clip": 0.01162982,
      "auxiliary_loss_mlp": 0.0102803,
      "balance_loss_clip": 1.04927075,
      "balance_loss_mlp": 1.01984084,
      "epoch": 0.5337581915469248,
      "flos": 21396691140480.0,
      "grad_norm": 1.8922888923037553,
      "language_loss": 0.80271918,
      "learning_rate": 1.8785551792411569e-06,
      "loss": 0.82462931,
      "num_input_tokens_seen": 95889755,
      "step": 4439,
      "time_per_iteration": 2.6886472702026367
    },
    {
      "auxiliary_loss_clip": 0.01152606,
      "auxiliary_loss_mlp": 0.01028424,
      "balance_loss_clip": 1.04901922,
      "balance_loss_mlp": 1.02082479,
      "epoch": 0.5338784344375639,
      "flos": 14865905064960.0,
      "grad_norm": 2.057691034298841,
      "language_loss": 0.82768238,
      "learning_rate": 1.8777776542677733e-06,
      "loss": 0.84949267,
      "num_input_tokens_seen": 95907805,
      "step": 4440,
      "time_per_iteration": 2.6588706970214844
    },
    {
      "auxiliary_loss_clip": 0.01136957,
      "auxiliary_loss_mlp": 0.01029315,
      "balance_loss_clip": 1.04205573,
      "balance_loss_mlp": 1.02130389,
      "epoch": 0.5339986773282029,
      "flos": 20813501923200.0,
      "grad_norm": 1.726980074577696,
      "language_loss": 0.73207724,
      "learning_rate": 1.8770001478354216e-06,
      "loss": 0.75373995,
      "num_input_tokens_seen": 95927480,
      "step": 4441,
      "time_per_iteration": 2.729613780975342
    },
    {
      "auxiliary_loss_clip": 0.01166271,
      "auxiliary_loss_mlp": 0.01036738,
      "balance_loss_clip": 1.05007875,
      "balance_loss_mlp": 1.02752924,
      "epoch": 0.5341189202188421,
      "flos": 17969089772160.0,
      "grad_norm": 2.3174023448291634,
      "language_loss": 0.84282082,
      "learning_rate": 1.8762226600620504e-06,
      "loss": 0.86485088,
      "num_input_tokens_seen": 95946095,
      "step": 4442,
      "time_per_iteration": 2.6815185546875
    },
    {
      "auxiliary_loss_clip": 0.01162252,
      "auxiliary_loss_mlp": 0.01034024,
      "balance_loss_clip": 1.04784441,
      "balance_loss_mlp": 1.02488089,
      "epoch": 0.5342391631094812,
      "flos": 11031866328960.0,
      "grad_norm": 2.5302510654783923,
      "language_loss": 0.59623432,
      "learning_rate": 1.8754451910656031e-06,
      "loss": 0.61819708,
      "num_input_tokens_seen": 95959995,
      "step": 4443,
      "time_per_iteration": 2.645616292953491
    },
    {
      "auxiliary_loss_clip": 0.01135036,
      "auxiliary_loss_mlp": 0.01032738,
      "balance_loss_clip": 1.04778314,
      "balance_loss_mlp": 1.0243336,
      "epoch": 0.5343594060001202,
      "flos": 15339135772800.0,
      "grad_norm": 2.078024492829133,
      "language_loss": 0.82568276,
      "learning_rate": 1.8746677409640212e-06,
      "loss": 0.84736049,
      "num_input_tokens_seen": 95977095,
      "step": 4444,
      "time_per_iteration": 2.7675089836120605
    },
    {
      "auxiliary_loss_clip": 0.01175281,
      "auxiliary_loss_mlp": 0.01026214,
      "balance_loss_clip": 1.05335438,
      "balance_loss_mlp": 1.01837027,
      "epoch": 0.5344796488907594,
      "flos": 26900898514560.0,
      "grad_norm": 1.7536272519256981,
      "language_loss": 0.84726572,
      "learning_rate": 1.8738903098752432e-06,
      "loss": 0.86928058,
      "num_input_tokens_seen": 95996225,
      "step": 4445,
      "time_per_iteration": 2.6577117443084717
    },
    {
      "auxiliary_loss_clip": 0.01155015,
      "auxiliary_loss_mlp": 0.01029861,
      "balance_loss_clip": 1.04903984,
      "balance_loss_mlp": 1.0217185,
      "epoch": 0.5345998917813984,
      "flos": 25411216740480.0,
      "grad_norm": 2.17614825041103,
      "language_loss": 0.73181868,
      "learning_rate": 1.8731128979172052e-06,
      "loss": 0.75366741,
      "num_input_tokens_seen": 96015425,
      "step": 4446,
      "time_per_iteration": 2.679762125015259
    },
    {
      "auxiliary_loss_clip": 0.01152333,
      "auxiliary_loss_mlp": 0.01025195,
      "balance_loss_clip": 1.05085635,
      "balance_loss_mlp": 1.01705301,
      "epoch": 0.5347201346720375,
      "flos": 32853379622400.0,
      "grad_norm": 2.2795699027501986,
      "language_loss": 0.67264092,
      "learning_rate": 1.8723355052078394e-06,
      "loss": 0.69441617,
      "num_input_tokens_seen": 96035460,
      "step": 4447,
      "time_per_iteration": 2.78326416015625
    },
    {
      "auxiliary_loss_clip": 0.01165356,
      "auxiliary_loss_mlp": 0.01034606,
      "balance_loss_clip": 1.04850602,
      "balance_loss_mlp": 1.02595782,
      "epoch": 0.5348403775626767,
      "flos": 17967940536960.0,
      "grad_norm": 3.3570732539407198,
      "language_loss": 0.77177429,
      "learning_rate": 1.8715581318650765e-06,
      "loss": 0.79377389,
      "num_input_tokens_seen": 96054515,
      "step": 4448,
      "time_per_iteration": 2.6226115226745605
    },
    {
      "auxiliary_loss_clip": 0.01153836,
      "auxiliary_loss_mlp": 0.01027672,
      "balance_loss_clip": 1.05246913,
      "balance_loss_mlp": 1.01833475,
      "epoch": 0.5349606204533157,
      "flos": 17603339535360.0,
      "grad_norm": 2.392776368350053,
      "language_loss": 0.81891906,
      "learning_rate": 1.8707807780068422e-06,
      "loss": 0.84073412,
      "num_input_tokens_seen": 96072330,
      "step": 4449,
      "time_per_iteration": 2.747387647628784
    },
    {
      "auxiliary_loss_clip": 0.01156134,
      "auxiliary_loss_mlp": 0.01029046,
      "balance_loss_clip": 1.04984164,
      "balance_loss_mlp": 1.02018821,
      "epoch": 0.5350808633439548,
      "flos": 29167831710720.0,
      "grad_norm": 1.8743204529868835,
      "language_loss": 0.6663183,
      "learning_rate": 1.8700034437510611e-06,
      "loss": 0.68817008,
      "num_input_tokens_seen": 96092425,
      "step": 4450,
      "time_per_iteration": 2.7108876705169678
    },
    {
      "auxiliary_loss_clip": 0.01130611,
      "auxiliary_loss_mlp": 0.01029155,
      "balance_loss_clip": 1.04527211,
      "balance_loss_mlp": 1.02063131,
      "epoch": 0.5352011062345938,
      "flos": 19499997381120.0,
      "grad_norm": 2.2559811157254073,
      "language_loss": 0.81405675,
      "learning_rate": 1.8692261292156549e-06,
      "loss": 0.83565438,
      "num_input_tokens_seen": 96111660,
      "step": 4451,
      "time_per_iteration": 2.7242164611816406
    },
    {
      "auxiliary_loss_clip": 0.01181637,
      "auxiliary_loss_mlp": 0.01029847,
      "balance_loss_clip": 1.05338991,
      "balance_loss_mlp": 1.02190137,
      "epoch": 0.535321349125233,
      "flos": 23477642691840.0,
      "grad_norm": 1.8462635073243578,
      "language_loss": 0.81027043,
      "learning_rate": 1.8684488345185401e-06,
      "loss": 0.83238524,
      "num_input_tokens_seen": 96131835,
      "step": 4452,
      "time_per_iteration": 2.6357946395874023
    },
    {
      "auxiliary_loss_clip": 0.01186983,
      "auxiliary_loss_mlp": 0.01043369,
      "balance_loss_clip": 1.05477822,
      "balance_loss_mlp": 1.03449392,
      "epoch": 0.535441592015872,
      "flos": 20478059786880.0,
      "grad_norm": 2.1390902373440928,
      "language_loss": 0.78786927,
      "learning_rate": 1.8676715597776332e-06,
      "loss": 0.8101728,
      "num_input_tokens_seen": 96150180,
      "step": 4453,
      "time_per_iteration": 2.58256459236145
    },
    {
      "auxiliary_loss_clip": 0.01117192,
      "auxiliary_loss_mlp": 0.01023191,
      "balance_loss_clip": 1.04498363,
      "balance_loss_mlp": 1.01558876,
      "epoch": 0.5355618349065111,
      "flos": 19573147428480.0,
      "grad_norm": 1.8578144157536076,
      "language_loss": 0.76066208,
      "learning_rate": 1.8668943051108455e-06,
      "loss": 0.78206587,
      "num_input_tokens_seen": 96167485,
      "step": 4454,
      "time_per_iteration": 2.835299253463745
    },
    {
      "auxiliary_loss_clip": 0.01155457,
      "auxiliary_loss_mlp": 0.0102748,
      "balance_loss_clip": 1.04912901,
      "balance_loss_mlp": 1.01893258,
      "epoch": 0.5356820777971503,
      "flos": 24024633978240.0,
      "grad_norm": 1.8050102705231947,
      "language_loss": 0.76016104,
      "learning_rate": 1.8661170706360856e-06,
      "loss": 0.78199041,
      "num_input_tokens_seen": 96186650,
      "step": 4455,
      "time_per_iteration": 2.6151602268218994
    },
    {
      "auxiliary_loss_clip": 0.01171477,
      "auxiliary_loss_mlp": 0.01025177,
      "balance_loss_clip": 1.05402911,
      "balance_loss_mlp": 1.01748216,
      "epoch": 0.5358023206877893,
      "flos": 20884676722560.0,
      "grad_norm": 1.5906833679966104,
      "language_loss": 0.81332201,
      "learning_rate": 1.8653398564712594e-06,
      "loss": 0.83528858,
      "num_input_tokens_seen": 96205595,
      "step": 4456,
      "time_per_iteration": 2.6540868282318115
    },
    {
      "auxiliary_loss_clip": 0.011679,
      "auxiliary_loss_mlp": 0.01028573,
      "balance_loss_clip": 1.05240333,
      "balance_loss_mlp": 1.02021646,
      "epoch": 0.5359225635784284,
      "flos": 22418996123520.0,
      "grad_norm": 1.6292206450168432,
      "language_loss": 0.82466662,
      "learning_rate": 1.8645626627342704e-06,
      "loss": 0.84663129,
      "num_input_tokens_seen": 96226360,
      "step": 4457,
      "time_per_iteration": 3.6488094329833984
    },
    {
      "auxiliary_loss_clip": 0.01176575,
      "auxiliary_loss_mlp": 0.01025523,
      "balance_loss_clip": 1.05338883,
      "balance_loss_mlp": 1.01760769,
      "epoch": 0.5360428064690675,
      "flos": 24097784025600.0,
      "grad_norm": 2.231464768925505,
      "language_loss": 0.80882758,
      "learning_rate": 1.8637854895430172e-06,
      "loss": 0.83084857,
      "num_input_tokens_seen": 96245625,
      "step": 4458,
      "time_per_iteration": 3.5858066082000732
    },
    {
      "auxiliary_loss_clip": 0.0113247,
      "auxiliary_loss_mlp": 0.01028864,
      "balance_loss_clip": 1.04630351,
      "balance_loss_mlp": 1.02037024,
      "epoch": 0.5361630493597066,
      "flos": 21434505183360.0,
      "grad_norm": 2.0366759274736905,
      "language_loss": 0.69370139,
      "learning_rate": 1.8630083370153978e-06,
      "loss": 0.71531469,
      "num_input_tokens_seen": 96265265,
      "step": 4459,
      "time_per_iteration": 3.659310817718506
    },
    {
      "auxiliary_loss_clip": 0.01025198,
      "auxiliary_loss_mlp": 0.01002182,
      "balance_loss_clip": 1.01195049,
      "balance_loss_mlp": 1.00054932,
      "epoch": 0.5362832922503457,
      "flos": 68888696520960.0,
      "grad_norm": 0.7421671522865998,
      "language_loss": 0.55335021,
      "learning_rate": 1.8622312052693041e-06,
      "loss": 0.57362401,
      "num_input_tokens_seen": 96326445,
      "step": 4460,
      "time_per_iteration": 3.431311845779419
    },
    {
      "auxiliary_loss_clip": 0.01158904,
      "auxiliary_loss_mlp": 0.01025967,
      "balance_loss_clip": 1.04585373,
      "balance_loss_mlp": 1.01725888,
      "epoch": 0.5364035351409848,
      "flos": 9793702563840.0,
      "grad_norm": 2.2212954571542824,
      "language_loss": 0.71355683,
      "learning_rate": 1.8614540944226267e-06,
      "loss": 0.73540556,
      "num_input_tokens_seen": 96343115,
      "step": 4461,
      "time_per_iteration": 2.6211354732513428
    },
    {
      "auxiliary_loss_clip": 0.01151657,
      "auxiliary_loss_mlp": 0.01030833,
      "balance_loss_clip": 1.0502826,
      "balance_loss_mlp": 1.02244055,
      "epoch": 0.5365237780316239,
      "flos": 23290080848640.0,
      "grad_norm": 1.6926802908418481,
      "language_loss": 0.68141991,
      "learning_rate": 1.8606770045932537e-06,
      "loss": 0.70324481,
      "num_input_tokens_seen": 96362230,
      "step": 4462,
      "time_per_iteration": 3.543541431427002
    },
    {
      "auxiliary_loss_clip": 0.01130777,
      "auxiliary_loss_mlp": 0.01027629,
      "balance_loss_clip": 1.0424931,
      "balance_loss_mlp": 1.01897466,
      "epoch": 0.5366440209222629,
      "flos": 26578133879040.0,
      "grad_norm": 1.7669476062884524,
      "language_loss": 0.8159287,
      "learning_rate": 1.859899935899068e-06,
      "loss": 0.83751273,
      "num_input_tokens_seen": 96382085,
      "step": 4463,
      "time_per_iteration": 2.77738094329834
    },
    {
      "auxiliary_loss_clip": 0.01152432,
      "auxiliary_loss_mlp": 0.0103199,
      "balance_loss_clip": 1.05019653,
      "balance_loss_mlp": 1.02334082,
      "epoch": 0.5367642638129021,
      "flos": 19608052469760.0,
      "grad_norm": 1.7725349714344847,
      "language_loss": 0.7912401,
      "learning_rate": 1.8591228884579506e-06,
      "loss": 0.81308436,
      "num_input_tokens_seen": 96400580,
      "step": 4464,
      "time_per_iteration": 2.7159528732299805
    },
    {
      "auxiliary_loss_clip": 0.01140522,
      "auxiliary_loss_mlp": 0.01029181,
      "balance_loss_clip": 1.04504371,
      "balance_loss_mlp": 1.02128923,
      "epoch": 0.5368845067035412,
      "flos": 23915214172800.0,
      "grad_norm": 2.232577384643004,
      "language_loss": 0.82329154,
      "learning_rate": 1.8583458623877795e-06,
      "loss": 0.84498858,
      "num_input_tokens_seen": 96419680,
      "step": 4465,
      "time_per_iteration": 2.714412212371826
    },
    {
      "auxiliary_loss_clip": 0.01171737,
      "auxiliary_loss_mlp": 0.01024053,
      "balance_loss_clip": 1.0512259,
      "balance_loss_mlp": 1.01622677,
      "epoch": 0.5370047495941802,
      "flos": 16873131951360.0,
      "grad_norm": 1.8477766329767404,
      "language_loss": 0.74663341,
      "learning_rate": 1.8575688578064281e-06,
      "loss": 0.76859128,
      "num_input_tokens_seen": 96437805,
      "step": 4466,
      "time_per_iteration": 2.706969738006592
    },
    {
      "auxiliary_loss_clip": 0.01176219,
      "auxiliary_loss_mlp": 0.01034693,
      "balance_loss_clip": 1.05428028,
      "balance_loss_mlp": 1.02590716,
      "epoch": 0.5371249924848194,
      "flos": 20740926493440.0,
      "grad_norm": 1.7289069928955294,
      "language_loss": 0.76498938,
      "learning_rate": 1.8567918748317674e-06,
      "loss": 0.78709859,
      "num_input_tokens_seen": 96457155,
      "step": 4467,
      "time_per_iteration": 2.657076835632324
    },
    {
      "auxiliary_loss_clip": 0.01139387,
      "auxiliary_loss_mlp": 0.01030354,
      "balance_loss_clip": 1.04538572,
      "balance_loss_mlp": 1.0220269,
      "epoch": 0.5372452353754584,
      "flos": 17968120104960.0,
      "grad_norm": 2.1440872404973654,
      "language_loss": 0.83140898,
      "learning_rate": 1.8560149135816659e-06,
      "loss": 0.85310638,
      "num_input_tokens_seen": 96473990,
      "step": 4468,
      "time_per_iteration": 2.6917827129364014
    },
    {
      "auxiliary_loss_clip": 0.01164435,
      "auxiliary_loss_mlp": 0.01026411,
      "balance_loss_clip": 1.04822636,
      "balance_loss_mlp": 1.01837587,
      "epoch": 0.5373654782660975,
      "flos": 15377021642880.0,
      "grad_norm": 2.999371324091986,
      "language_loss": 0.84279168,
      "learning_rate": 1.8552379741739873e-06,
      "loss": 0.86470014,
      "num_input_tokens_seen": 96491335,
      "step": 4469,
      "time_per_iteration": 2.6227927207946777
    },
    {
      "auxiliary_loss_clip": 0.01042187,
      "auxiliary_loss_mlp": 0.00754113,
      "balance_loss_clip": 1.01108456,
      "balance_loss_mlp": 1.00042772,
      "epoch": 0.5374857211567367,
      "flos": 69000091574400.0,
      "grad_norm": 0.8929543015985278,
      "language_loss": 0.55655497,
      "learning_rate": 1.8544610567265935e-06,
      "loss": 0.57451802,
      "num_input_tokens_seen": 96545275,
      "step": 4470,
      "time_per_iteration": 3.204318046569824
    },
    {
      "auxiliary_loss_clip": 0.01155993,
      "auxiliary_loss_mlp": 0.00763671,
      "balance_loss_clip": 1.05192578,
      "balance_loss_mlp": 1.00031447,
      "epoch": 0.5376059640473757,
      "flos": 15085355207040.0,
      "grad_norm": 1.9564426822136085,
      "language_loss": 0.83454138,
      "learning_rate": 1.853684161357341e-06,
      "loss": 0.85373801,
      "num_input_tokens_seen": 96562935,
      "step": 4471,
      "time_per_iteration": 2.670901298522949
    },
    {
      "auxiliary_loss_clip": 0.01168134,
      "auxiliary_loss_mlp": 0.00763445,
      "balance_loss_clip": 1.051525,
      "balance_loss_mlp": 1.00024271,
      "epoch": 0.5377262069380148,
      "flos": 19792597570560.0,
      "grad_norm": 4.85384481487378,
      "language_loss": 0.76960313,
      "learning_rate": 1.852907288184085e-06,
      "loss": 0.78891897,
      "num_input_tokens_seen": 96581820,
      "step": 4472,
      "time_per_iteration": 2.6365270614624023
    },
    {
      "auxiliary_loss_clip": 0.01125355,
      "auxiliary_loss_mlp": 0.01030934,
      "balance_loss_clip": 1.04405355,
      "balance_loss_mlp": 1.02290571,
      "epoch": 0.5378464498286539,
      "flos": 30003077640960.0,
      "grad_norm": 2.106816650142174,
      "language_loss": 0.70295799,
      "learning_rate": 1.8521304373246762e-06,
      "loss": 0.72452092,
      "num_input_tokens_seen": 96602865,
      "step": 4473,
      "time_per_iteration": 2.80812406539917
    },
    {
      "auxiliary_loss_clip": 0.01172122,
      "auxiliary_loss_mlp": 0.01035677,
      "balance_loss_clip": 1.05004704,
      "balance_loss_mlp": 1.02709389,
      "epoch": 0.537966692719293,
      "flos": 21251217058560.0,
      "grad_norm": 2.580670376508372,
      "language_loss": 0.88572675,
      "learning_rate": 1.8513536088969626e-06,
      "loss": 0.90780473,
      "num_input_tokens_seen": 96620530,
      "step": 4474,
      "time_per_iteration": 2.608285903930664
    },
    {
      "auxiliary_loss_clip": 0.01171709,
      "auxiliary_loss_mlp": 0.01029283,
      "balance_loss_clip": 1.05249906,
      "balance_loss_mlp": 1.02042556,
      "epoch": 0.538086935609932,
      "flos": 21543170803200.0,
      "grad_norm": 1.6053895120399482,
      "language_loss": 0.80759811,
      "learning_rate": 1.8505768030187884e-06,
      "loss": 0.82960802,
      "num_input_tokens_seen": 96640660,
      "step": 4475,
      "time_per_iteration": 2.6804568767547607
    },
    {
      "auxiliary_loss_clip": 0.01153225,
      "auxiliary_loss_mlp": 0.01029064,
      "balance_loss_clip": 1.05153048,
      "balance_loss_mlp": 1.02087474,
      "epoch": 0.5382071785005712,
      "flos": 22747219626240.0,
      "grad_norm": 1.587945484226246,
      "language_loss": 0.79952997,
      "learning_rate": 1.849800019807995e-06,
      "loss": 0.82135284,
      "num_input_tokens_seen": 96661885,
      "step": 4476,
      "time_per_iteration": 2.6981992721557617
    },
    {
      "auxiliary_loss_clip": 0.01136979,
      "auxiliary_loss_mlp": 0.01032616,
      "balance_loss_clip": 1.04707539,
      "balance_loss_mlp": 1.02457523,
      "epoch": 0.5383274213912103,
      "flos": 24934574240640.0,
      "grad_norm": 2.2554900103437125,
      "language_loss": 0.71181369,
      "learning_rate": 1.8490232593824186e-06,
      "loss": 0.7335096,
      "num_input_tokens_seen": 96678340,
      "step": 4477,
      "time_per_iteration": 2.7794244289398193
    },
    {
      "auxiliary_loss_clip": 0.01149274,
      "auxiliary_loss_mlp": 0.01027399,
      "balance_loss_clip": 1.04782403,
      "balance_loss_mlp": 1.01932824,
      "epoch": 0.5384476642818493,
      "flos": 22310186849280.0,
      "grad_norm": 2.053523952397613,
      "language_loss": 0.84648877,
      "learning_rate": 1.8482465218598935e-06,
      "loss": 0.8682555,
      "num_input_tokens_seen": 96698285,
      "step": 4478,
      "time_per_iteration": 2.647535800933838
    },
    {
      "auxiliary_loss_clip": 0.01138561,
      "auxiliary_loss_mlp": 0.01037565,
      "balance_loss_clip": 1.04709291,
      "balance_loss_mlp": 1.02876782,
      "epoch": 0.5385679071724885,
      "flos": 22711021695360.0,
      "grad_norm": 2.3669212994290723,
      "language_loss": 0.83607042,
      "learning_rate": 1.8474698073582508e-06,
      "loss": 0.85783172,
      "num_input_tokens_seen": 96719655,
      "step": 4479,
      "time_per_iteration": 2.6946322917938232
    },
    {
      "auxiliary_loss_clip": 0.01142961,
      "auxiliary_loss_mlp": 0.01029669,
      "balance_loss_clip": 1.04496455,
      "balance_loss_mlp": 1.02093709,
      "epoch": 0.5386881500631275,
      "flos": 15953746412160.0,
      "grad_norm": 3.6155790141516784,
      "language_loss": 0.87705851,
      "learning_rate": 1.8466931159953166e-06,
      "loss": 0.89878482,
      "num_input_tokens_seen": 96736290,
      "step": 4480,
      "time_per_iteration": 2.6859610080718994
    },
    {
      "auxiliary_loss_clip": 0.01160311,
      "auxiliary_loss_mlp": 0.0102927,
      "balance_loss_clip": 1.05320072,
      "balance_loss_mlp": 1.02074671,
      "epoch": 0.5388083929537666,
      "flos": 24060041809920.0,
      "grad_norm": 1.787057636112188,
      "language_loss": 0.84246463,
      "learning_rate": 1.8459164478889158e-06,
      "loss": 0.86436045,
      "num_input_tokens_seen": 96757685,
      "step": 4481,
      "time_per_iteration": 2.6710474491119385
    },
    {
      "auxiliary_loss_clip": 0.01130914,
      "auxiliary_loss_mlp": 0.01024765,
      "balance_loss_clip": 1.04416919,
      "balance_loss_mlp": 1.01749659,
      "epoch": 0.5389286358444056,
      "flos": 22236893147520.0,
      "grad_norm": 1.6773970846725486,
      "language_loss": 0.76133442,
      "learning_rate": 1.8451398031568663e-06,
      "loss": 0.78289115,
      "num_input_tokens_seen": 96777310,
      "step": 4482,
      "time_per_iteration": 2.7020978927612305
    },
    {
      "auxiliary_loss_clip": 0.01139635,
      "auxiliary_loss_mlp": 0.01033065,
      "balance_loss_clip": 1.04823101,
      "balance_loss_mlp": 1.02433932,
      "epoch": 0.5390488787350448,
      "flos": 24281718595200.0,
      "grad_norm": 1.4841246704341133,
      "language_loss": 0.74601638,
      "learning_rate": 1.844363181916986e-06,
      "loss": 0.76774341,
      "num_input_tokens_seen": 96798035,
      "step": 4483,
      "time_per_iteration": 3.7384109497070312
    },
    {
      "auxiliary_loss_clip": 0.01164815,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.04854584,
      "balance_loss_mlp": 1.0182693,
      "epoch": 0.5391691216256839,
      "flos": 16581393688320.0,
      "grad_norm": 4.07311245129802,
      "language_loss": 0.83047181,
      "learning_rate": 1.8435865842870868e-06,
      "loss": 0.85239178,
      "num_input_tokens_seen": 96815975,
      "step": 4484,
      "time_per_iteration": 2.6335999965667725
    },
    {
      "auxiliary_loss_clip": 0.01144412,
      "auxiliary_loss_mlp": 0.00763913,
      "balance_loss_clip": 1.04543519,
      "balance_loss_mlp": 1.00019121,
      "epoch": 0.5392893645163229,
      "flos": 23330049707520.0,
      "grad_norm": 1.8326906848611888,
      "language_loss": 0.71699464,
      "learning_rate": 1.8428100103849787e-06,
      "loss": 0.73607779,
      "num_input_tokens_seen": 96835770,
      "step": 4485,
      "time_per_iteration": 3.536465883255005
    },
    {
      "auxiliary_loss_clip": 0.01153204,
      "auxiliary_loss_mlp": 0.01030125,
      "balance_loss_clip": 1.05030012,
      "balance_loss_mlp": 1.02144027,
      "epoch": 0.5394096074069621,
      "flos": 15669801400320.0,
      "grad_norm": 2.242677675524116,
      "language_loss": 0.73577929,
      "learning_rate": 1.842033460328467e-06,
      "loss": 0.75761259,
      "num_input_tokens_seen": 96854490,
      "step": 4486,
      "time_per_iteration": 2.6450300216674805
    },
    {
      "auxiliary_loss_clip": 0.01159056,
      "auxiliary_loss_mlp": 0.00763429,
      "balance_loss_clip": 1.04975605,
      "balance_loss_mlp": 1.00022733,
      "epoch": 0.5395298502976011,
      "flos": 22893447893760.0,
      "grad_norm": 2.404478916102859,
      "language_loss": 0.75182509,
      "learning_rate": 1.8412569342353541e-06,
      "loss": 0.77104998,
      "num_input_tokens_seen": 96874645,
      "step": 4487,
      "time_per_iteration": 2.744840621948242
    },
    {
      "auxiliary_loss_clip": 0.01161088,
      "auxiliary_loss_mlp": 0.01032203,
      "balance_loss_clip": 1.05171371,
      "balance_loss_mlp": 1.02288055,
      "epoch": 0.5396500931882402,
      "flos": 23842135952640.0,
      "grad_norm": 1.8751407604536472,
      "language_loss": 0.84506762,
      "learning_rate": 1.840480432223438e-06,
      "loss": 0.86700046,
      "num_input_tokens_seen": 96893650,
      "step": 4488,
      "time_per_iteration": 3.5878994464874268
    },
    {
      "auxiliary_loss_clip": 0.01157337,
      "auxiliary_loss_mlp": 0.01029756,
      "balance_loss_clip": 1.04824436,
      "balance_loss_mlp": 1.02138758,
      "epoch": 0.5397703360788794,
      "flos": 26322988596480.0,
      "grad_norm": 2.0213511915156377,
      "language_loss": 0.77587557,
      "learning_rate": 1.8397039544105131e-06,
      "loss": 0.79774642,
      "num_input_tokens_seen": 96912735,
      "step": 4489,
      "time_per_iteration": 2.757640838623047
    },
    {
      "auxiliary_loss_clip": 0.01147586,
      "auxiliary_loss_mlp": 0.01028005,
      "balance_loss_clip": 1.04481459,
      "balance_loss_mlp": 1.01942205,
      "epoch": 0.5398905789695184,
      "flos": 21214588164480.0,
      "grad_norm": 1.7151451309884629,
      "language_loss": 0.69785297,
      "learning_rate": 1.8389275009143711e-06,
      "loss": 0.71960878,
      "num_input_tokens_seen": 96932475,
      "step": 4490,
      "time_per_iteration": 2.6806225776672363
    },
    {
      "auxiliary_loss_clip": 0.01178963,
      "auxiliary_loss_mlp": 0.0102775,
      "balance_loss_clip": 1.04951239,
      "balance_loss_mlp": 1.01936984,
      "epoch": 0.5400108218601575,
      "flos": 25080335631360.0,
      "grad_norm": 1.6941582493664962,
      "language_loss": 0.73475015,
      "learning_rate": 1.8381510718527988e-06,
      "loss": 0.75681722,
      "num_input_tokens_seen": 96952085,
      "step": 4491,
      "time_per_iteration": 2.6398253440856934
    },
    {
      "auxiliary_loss_clip": 0.01156413,
      "auxiliary_loss_mlp": 0.01028473,
      "balance_loss_clip": 1.04691291,
      "balance_loss_mlp": 1.02058101,
      "epoch": 0.5401310647507966,
      "flos": 26357498588160.0,
      "grad_norm": 1.8162920823656636,
      "language_loss": 0.63757181,
      "learning_rate": 1.8373746673435812e-06,
      "loss": 0.65942061,
      "num_input_tokens_seen": 96973110,
      "step": 4492,
      "time_per_iteration": 2.801379442214966
    },
    {
      "auxiliary_loss_clip": 0.0118711,
      "auxiliary_loss_mlp": 0.01028073,
      "balance_loss_clip": 1.05333447,
      "balance_loss_mlp": 1.01952577,
      "epoch": 0.5402513076414357,
      "flos": 27855332749440.0,
      "grad_norm": 1.6407675708596272,
      "language_loss": 0.79135919,
      "learning_rate": 1.8365982875044964e-06,
      "loss": 0.81351101,
      "num_input_tokens_seen": 96993420,
      "step": 4493,
      "time_per_iteration": 2.661428689956665
    },
    {
      "auxiliary_loss_clip": 0.01176772,
      "auxiliary_loss_mlp": 0.00765003,
      "balance_loss_clip": 1.05268002,
      "balance_loss_mlp": 1.00021887,
      "epoch": 0.5403715505320748,
      "flos": 22893771116160.0,
      "grad_norm": 2.2836601906519816,
      "language_loss": 0.76513553,
      "learning_rate": 1.8358219324533217e-06,
      "loss": 0.78455329,
      "num_input_tokens_seen": 97013685,
      "step": 4494,
      "time_per_iteration": 2.679168224334717
    },
    {
      "auxiliary_loss_clip": 0.01147228,
      "auxiliary_loss_mlp": 0.01031702,
      "balance_loss_clip": 1.04581559,
      "balance_loss_mlp": 1.02407289,
      "epoch": 0.5404917934227139,
      "flos": 30224143895040.0,
      "grad_norm": 2.4634098873532246,
      "language_loss": 0.70466864,
      "learning_rate": 1.8350456023078292e-06,
      "loss": 0.72645795,
      "num_input_tokens_seen": 97036060,
      "step": 4495,
      "time_per_iteration": 2.748868942260742
    },
    {
      "auxiliary_loss_clip": 0.01188486,
      "auxiliary_loss_mlp": 0.01039458,
      "balance_loss_clip": 1.05249178,
      "balance_loss_mlp": 1.03049898,
      "epoch": 0.540612036313353,
      "flos": 19938502615680.0,
      "grad_norm": 2.842309072725374,
      "language_loss": 0.78224218,
      "learning_rate": 1.8342692971857874e-06,
      "loss": 0.80452156,
      "num_input_tokens_seen": 97055260,
      "step": 4496,
      "time_per_iteration": 2.5953657627105713
    },
    {
      "auxiliary_loss_clip": 0.01151484,
      "auxiliary_loss_mlp": 0.01027802,
      "balance_loss_clip": 1.04813075,
      "balance_loss_mlp": 1.01975489,
      "epoch": 0.540732279203992,
      "flos": 24279599692800.0,
      "grad_norm": 2.558862561593949,
      "language_loss": 0.71488833,
      "learning_rate": 1.833493017204962e-06,
      "loss": 0.73668116,
      "num_input_tokens_seen": 97075365,
      "step": 4497,
      "time_per_iteration": 2.689786911010742
    },
    {
      "auxiliary_loss_clip": 0.01182492,
      "auxiliary_loss_mlp": 0.01029527,
      "balance_loss_clip": 1.05131006,
      "balance_loss_mlp": 1.02156949,
      "epoch": 0.5408525220946312,
      "flos": 20193216935040.0,
      "grad_norm": 3.6136761814672953,
      "language_loss": 0.77928126,
      "learning_rate": 1.8327167624831134e-06,
      "loss": 0.80140144,
      "num_input_tokens_seen": 97093095,
      "step": 4498,
      "time_per_iteration": 2.707052230834961
    },
    {
      "auxiliary_loss_clip": 0.0118223,
      "auxiliary_loss_mlp": 0.0102517,
      "balance_loss_clip": 1.05299509,
      "balance_loss_mlp": 1.01653385,
      "epoch": 0.5409727649852702,
      "flos": 24134448833280.0,
      "grad_norm": 1.6025674411595683,
      "language_loss": 0.70945507,
      "learning_rate": 1.831940533137999e-06,
      "loss": 0.73152906,
      "num_input_tokens_seen": 97112000,
      "step": 4499,
      "time_per_iteration": 2.617994785308838
    },
    {
      "auxiliary_loss_clip": 0.0116607,
      "auxiliary_loss_mlp": 0.01025113,
      "balance_loss_clip": 1.05221725,
      "balance_loss_mlp": 1.01744783,
      "epoch": 0.5410930078759093,
      "flos": 23912700220800.0,
      "grad_norm": 1.716419231076312,
      "language_loss": 0.72629642,
      "learning_rate": 1.8311643292873718e-06,
      "loss": 0.74820822,
      "num_input_tokens_seen": 97130820,
      "step": 4500,
      "time_per_iteration": 2.643759250640869
    },
    {
      "auxiliary_loss_clip": 0.01166173,
      "auxiliary_loss_mlp": 0.01026369,
      "balance_loss_clip": 1.04992938,
      "balance_loss_mlp": 1.01855516,
      "epoch": 0.5412132507665485,
      "flos": 21105132445440.0,
      "grad_norm": 1.7765735358090062,
      "language_loss": 0.88022947,
      "learning_rate": 1.8303881510489818e-06,
      "loss": 0.90215492,
      "num_input_tokens_seen": 97149210,
      "step": 4501,
      "time_per_iteration": 2.5760936737060547
    },
    {
      "auxiliary_loss_clip": 0.01156291,
      "auxiliary_loss_mlp": 0.01028013,
      "balance_loss_clip": 1.05046296,
      "balance_loss_mlp": 1.01994824,
      "epoch": 0.5413334936571875,
      "flos": 30227340205440.0,
      "grad_norm": 5.375738852453281,
      "language_loss": 0.69373363,
      "learning_rate": 1.829611998540574e-06,
      "loss": 0.71557671,
      "num_input_tokens_seen": 97170415,
      "step": 4502,
      "time_per_iteration": 2.7537841796875
    },
    {
      "auxiliary_loss_clip": 0.01168647,
      "auxiliary_loss_mlp": 0.00763482,
      "balance_loss_clip": 1.05000401,
      "balance_loss_mlp": 1.00022483,
      "epoch": 0.5414537365478266,
      "flos": 24279635606400.0,
      "grad_norm": 1.8223699893858867,
      "language_loss": 0.79811013,
      "learning_rate": 1.8288358718798914e-06,
      "loss": 0.81743145,
      "num_input_tokens_seen": 97189605,
      "step": 4503,
      "time_per_iteration": 2.6696269512176514
    },
    {
      "auxiliary_loss_clip": 0.01163158,
      "auxiliary_loss_mlp": 0.00763062,
      "balance_loss_clip": 1.04963231,
      "balance_loss_mlp": 1.00022793,
      "epoch": 0.5415739794384657,
      "flos": 16654543735680.0,
      "grad_norm": 1.7694664099578135,
      "language_loss": 0.72766364,
      "learning_rate": 1.8280597711846703e-06,
      "loss": 0.74692583,
      "num_input_tokens_seen": 97207845,
      "step": 4504,
      "time_per_iteration": 2.732043981552124
    },
    {
      "auxiliary_loss_clip": 0.01166492,
      "auxiliary_loss_mlp": 0.01025417,
      "balance_loss_clip": 1.05205083,
      "balance_loss_mlp": 1.01717341,
      "epoch": 0.5416942223291048,
      "flos": 23185724860800.0,
      "grad_norm": 1.9143258249297963,
      "language_loss": 0.83254319,
      "learning_rate": 1.8272836965726455e-06,
      "loss": 0.85446233,
      "num_input_tokens_seen": 97226780,
      "step": 4505,
      "time_per_iteration": 2.6476781368255615
    },
    {
      "auxiliary_loss_clip": 0.01108364,
      "auxiliary_loss_mlp": 0.01035426,
      "balance_loss_clip": 1.04173112,
      "balance_loss_mlp": 1.02718902,
      "epoch": 0.5418144652197439,
      "flos": 20303247271680.0,
      "grad_norm": 1.7067042374171384,
      "language_loss": 0.77912915,
      "learning_rate": 1.8265076481615461e-06,
      "loss": 0.80056709,
      "num_input_tokens_seen": 97246695,
      "step": 4506,
      "time_per_iteration": 2.758835554122925
    },
    {
      "auxiliary_loss_clip": 0.01156675,
      "auxiliary_loss_mlp": 0.01026097,
      "balance_loss_clip": 1.05302978,
      "balance_loss_mlp": 1.01687074,
      "epoch": 0.541934708110383,
      "flos": 12458633431680.0,
      "grad_norm": 2.1650426962876375,
      "language_loss": 0.87106156,
      "learning_rate": 1.8257316260690987e-06,
      "loss": 0.89288926,
      "num_input_tokens_seen": 97264480,
      "step": 4507,
      "time_per_iteration": 2.654953718185425
    },
    {
      "auxiliary_loss_clip": 0.01168251,
      "auxiliary_loss_mlp": 0.01024309,
      "balance_loss_clip": 1.04996002,
      "balance_loss_mlp": 1.01687658,
      "epoch": 0.5420549510010221,
      "flos": 21253802837760.0,
      "grad_norm": 1.5857920742515368,
      "language_loss": 0.76206845,
      "learning_rate": 1.8249556304130254e-06,
      "loss": 0.78399408,
      "num_input_tokens_seen": 97285760,
      "step": 4508,
      "time_per_iteration": 2.6863386631011963
    },
    {
      "auxiliary_loss_clip": 0.01144567,
      "auxiliary_loss_mlp": 0.01033703,
      "balance_loss_clip": 1.04709375,
      "balance_loss_mlp": 1.02501822,
      "epoch": 0.5421751938916611,
      "flos": 29490524519040.0,
      "grad_norm": 2.1849595519581646,
      "language_loss": 0.68974149,
      "learning_rate": 1.824179661311044e-06,
      "loss": 0.71152425,
      "num_input_tokens_seen": 97304510,
      "step": 4509,
      "time_per_iteration": 3.8503623008728027
    },
    {
      "auxiliary_loss_clip": 0.01123665,
      "auxiliary_loss_mlp": 0.01029402,
      "balance_loss_clip": 1.0425185,
      "balance_loss_mlp": 1.02105188,
      "epoch": 0.5422954367823003,
      "flos": 18734238311040.0,
      "grad_norm": 1.8529497804113326,
      "language_loss": 0.80026221,
      "learning_rate": 1.823403718880868e-06,
      "loss": 0.82179284,
      "num_input_tokens_seen": 97323270,
      "step": 4510,
      "time_per_iteration": 3.6918458938598633
    },
    {
      "auxiliary_loss_clip": 0.01153703,
      "auxiliary_loss_mlp": 0.01025441,
      "balance_loss_clip": 1.0466814,
      "balance_loss_mlp": 1.01757312,
      "epoch": 0.5424156796729394,
      "flos": 39969006940800.0,
      "grad_norm": 1.6845400144598266,
      "language_loss": 0.66504961,
      "learning_rate": 1.822627803240207e-06,
      "loss": 0.68684107,
      "num_input_tokens_seen": 97345600,
      "step": 4511,
      "time_per_iteration": 3.7475943565368652
    },
    {
      "auxiliary_loss_clip": 0.01144557,
      "auxiliary_loss_mlp": 0.01034144,
      "balance_loss_clip": 1.0473423,
      "balance_loss_mlp": 1.025895,
      "epoch": 0.5425359225635784,
      "flos": 11546538353280.0,
      "grad_norm": 2.2147829896992226,
      "language_loss": 0.85110098,
      "learning_rate": 1.8218519145067675e-06,
      "loss": 0.87288797,
      "num_input_tokens_seen": 97361220,
      "step": 4512,
      "time_per_iteration": 2.69085693359375
    },
    {
      "auxiliary_loss_clip": 0.0113553,
      "auxiliary_loss_mlp": 0.01028299,
      "balance_loss_clip": 1.04559803,
      "balance_loss_mlp": 1.01988244,
      "epoch": 0.5426561654542175,
      "flos": 20229702174720.0,
      "grad_norm": 1.894324577942147,
      "language_loss": 0.89488119,
      "learning_rate": 1.8210760527982508e-06,
      "loss": 0.91651946,
      "num_input_tokens_seen": 97381505,
      "step": 4513,
      "time_per_iteration": 2.7229795455932617
    },
    {
      "auxiliary_loss_clip": 0.01157838,
      "auxiliary_loss_mlp": 0.0076316,
      "balance_loss_clip": 1.05284333,
      "balance_loss_mlp": 1.00021458,
      "epoch": 0.5427764083448566,
      "flos": 21871681614720.0,
      "grad_norm": 4.9002028909502195,
      "language_loss": 0.75170904,
      "learning_rate": 1.8203002182323552e-06,
      "loss": 0.77091897,
      "num_input_tokens_seen": 97399060,
      "step": 4514,
      "time_per_iteration": 3.5980536937713623
    },
    {
      "auxiliary_loss_clip": 0.01156577,
      "auxiliary_loss_mlp": 0.01029276,
      "balance_loss_clip": 1.05035305,
      "balance_loss_mlp": 1.02118218,
      "epoch": 0.5428966512354957,
      "flos": 19640946349440.0,
      "grad_norm": 1.7534062060695885,
      "language_loss": 0.75671422,
      "learning_rate": 1.819524410926773e-06,
      "loss": 0.77857274,
      "num_input_tokens_seen": 97416740,
      "step": 4515,
      "time_per_iteration": 2.7132346630096436
    },
    {
      "auxiliary_loss_clip": 0.01108971,
      "auxiliary_loss_mlp": 0.01033916,
      "balance_loss_clip": 1.04381204,
      "balance_loss_mlp": 1.02539229,
      "epoch": 0.5430168941261347,
      "flos": 22382187661440.0,
      "grad_norm": 1.684777980531034,
      "language_loss": 0.7695142,
      "learning_rate": 1.8187486309991944e-06,
      "loss": 0.79094303,
      "num_input_tokens_seen": 97437620,
      "step": 4516,
      "time_per_iteration": 2.7904083728790283
    },
    {
      "auxiliary_loss_clip": 0.01176665,
      "auxiliary_loss_mlp": 0.01028257,
      "balance_loss_clip": 1.05450773,
      "balance_loss_mlp": 1.02057362,
      "epoch": 0.5431371370167739,
      "flos": 18764187275520.0,
      "grad_norm": 1.649096604608107,
      "language_loss": 0.77478206,
      "learning_rate": 1.817972878567304e-06,
      "loss": 0.79683125,
      "num_input_tokens_seen": 97456275,
      "step": 4517,
      "time_per_iteration": 2.612614154815674
    },
    {
      "auxiliary_loss_clip": 0.01158041,
      "auxiliary_loss_mlp": 0.01028161,
      "balance_loss_clip": 1.04845107,
      "balance_loss_mlp": 1.02038264,
      "epoch": 0.543257379907413,
      "flos": 18806023641600.0,
      "grad_norm": 1.6846500083166898,
      "language_loss": 0.76326406,
      "learning_rate": 1.8171971537487834e-06,
      "loss": 0.78512609,
      "num_input_tokens_seen": 97474925,
      "step": 4518,
      "time_per_iteration": 2.720061779022217
    },
    {
      "auxiliary_loss_clip": 0.01184163,
      "auxiliary_loss_mlp": 0.01029855,
      "balance_loss_clip": 1.05149031,
      "balance_loss_mlp": 1.02153432,
      "epoch": 0.543377622798052,
      "flos": 17493381025920.0,
      "grad_norm": 1.8953825179276578,
      "language_loss": 0.80521321,
      "learning_rate": 1.8164214566613093e-06,
      "loss": 0.82735336,
      "num_input_tokens_seen": 97493550,
      "step": 4519,
      "time_per_iteration": 2.6553099155426025
    },
    {
      "auxiliary_loss_clip": 0.0118318,
      "auxiliary_loss_mlp": 0.01035449,
      "balance_loss_clip": 1.05283046,
      "balance_loss_mlp": 1.02687144,
      "epoch": 0.5434978656886912,
      "flos": 18989311766400.0,
      "grad_norm": 3.416232634415561,
      "language_loss": 0.65322006,
      "learning_rate": 1.8156457874225547e-06,
      "loss": 0.67540634,
      "num_input_tokens_seen": 97512010,
      "step": 4520,
      "time_per_iteration": 2.693605422973633
    },
    {
      "auxiliary_loss_clip": 0.01147457,
      "auxiliary_loss_mlp": 0.01026347,
      "balance_loss_clip": 1.04976714,
      "balance_loss_mlp": 1.01872957,
      "epoch": 0.5436181085793302,
      "flos": 17274936464640.0,
      "grad_norm": 2.1845858881761058,
      "language_loss": 0.80606848,
      "learning_rate": 1.814870146150187e-06,
      "loss": 0.82780653,
      "num_input_tokens_seen": 97530120,
      "step": 4521,
      "time_per_iteration": 2.7053284645080566
    },
    {
      "auxiliary_loss_clip": 0.01161907,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.04850864,
      "balance_loss_mlp": 1.01892626,
      "epoch": 0.5437383514699693,
      "flos": 19098587917440.0,
      "grad_norm": 2.7115003803420086,
      "language_loss": 0.78827161,
      "learning_rate": 1.814094532961871e-06,
      "loss": 0.81016624,
      "num_input_tokens_seen": 97548695,
      "step": 4522,
      "time_per_iteration": 2.6635892391204834
    },
    {
      "auxiliary_loss_clip": 0.01127147,
      "auxiliary_loss_mlp": 0.01027749,
      "balance_loss_clip": 1.04549885,
      "balance_loss_mlp": 1.01912451,
      "epoch": 0.5438585943606085,
      "flos": 22602715211520.0,
      "grad_norm": 4.723978148252467,
      "language_loss": 0.83982038,
      "learning_rate": 1.8133189479752666e-06,
      "loss": 0.86136937,
      "num_input_tokens_seen": 97567625,
      "step": 4523,
      "time_per_iteration": 2.7019596099853516
    },
    {
      "auxiliary_loss_clip": 0.01183196,
      "auxiliary_loss_mlp": 0.01030521,
      "balance_loss_clip": 1.05239224,
      "balance_loss_mlp": 1.02258182,
      "epoch": 0.5439788372512475,
      "flos": 21798495653760.0,
      "grad_norm": 2.3310760880464567,
      "language_loss": 0.82012916,
      "learning_rate": 1.8125433913080292e-06,
      "loss": 0.84226632,
      "num_input_tokens_seen": 97585325,
      "step": 4524,
      "time_per_iteration": 2.644388437271118
    },
    {
      "auxiliary_loss_clip": 0.01062181,
      "auxiliary_loss_mlp": 0.01028621,
      "balance_loss_clip": 1.03824019,
      "balance_loss_mlp": 1.02059817,
      "epoch": 0.5440990801418866,
      "flos": 16399362539520.0,
      "grad_norm": 2.008489841221986,
      "language_loss": 0.82773662,
      "learning_rate": 1.811767863077811e-06,
      "loss": 0.84864467,
      "num_input_tokens_seen": 97604275,
      "step": 4525,
      "time_per_iteration": 3.0538370609283447
    },
    {
      "auxiliary_loss_clip": 0.01104434,
      "auxiliary_loss_mlp": 0.0102878,
      "balance_loss_clip": 1.04565489,
      "balance_loss_mlp": 1.02119207,
      "epoch": 0.5442193230325257,
      "flos": 21615638492160.0,
      "grad_norm": 1.7610557598916503,
      "language_loss": 0.78276402,
      "learning_rate": 1.8109923634022577e-06,
      "loss": 0.80409616,
      "num_input_tokens_seen": 97624300,
      "step": 4526,
      "time_per_iteration": 3.1037790775299072
    },
    {
      "auxiliary_loss_clip": 0.01185069,
      "auxiliary_loss_mlp": 0.01029181,
      "balance_loss_clip": 1.05249953,
      "balance_loss_mlp": 1.02139115,
      "epoch": 0.5443395659231648,
      "flos": 15481198062720.0,
      "grad_norm": 2.613250103838892,
      "language_loss": 0.86778247,
      "learning_rate": 1.8102168923990128e-06,
      "loss": 0.889925,
      "num_input_tokens_seen": 97637845,
      "step": 4527,
      "time_per_iteration": 2.566715717315674
    },
    {
      "auxiliary_loss_clip": 0.01171183,
      "auxiliary_loss_mlp": 0.00763176,
      "balance_loss_clip": 1.05235362,
      "balance_loss_mlp": 1.00026202,
      "epoch": 0.5444598088138038,
      "flos": 18770436241920.0,
      "grad_norm": 2.2449723933140375,
      "language_loss": 0.79943764,
      "learning_rate": 1.809441450185714e-06,
      "loss": 0.81878126,
      "num_input_tokens_seen": 97656330,
      "step": 4528,
      "time_per_iteration": 2.611246347427368
    },
    {
      "auxiliary_loss_clip": 0.0115899,
      "auxiliary_loss_mlp": 0.01027683,
      "balance_loss_clip": 1.04768562,
      "balance_loss_mlp": 1.01949286,
      "epoch": 0.544580051704443,
      "flos": 21142335957120.0,
      "grad_norm": 2.4308110276056607,
      "language_loss": 0.732391,
      "learning_rate": 1.8086660368799958e-06,
      "loss": 0.75425774,
      "num_input_tokens_seen": 97674380,
      "step": 4529,
      "time_per_iteration": 2.6160969734191895
    },
    {
      "auxiliary_loss_clip": 0.0115846,
      "auxiliary_loss_mlp": 0.01027274,
      "balance_loss_clip": 1.05133915,
      "balance_loss_mlp": 1.0189178,
      "epoch": 0.5447002945950821,
      "flos": 32491508054400.0,
      "grad_norm": 1.879335170582177,
      "language_loss": 0.77392018,
      "learning_rate": 1.807890652599488e-06,
      "loss": 0.79577756,
      "num_input_tokens_seen": 97698765,
      "step": 4530,
      "time_per_iteration": 2.787216901779175
    },
    {
      "auxiliary_loss_clip": 0.01179685,
      "auxiliary_loss_mlp": 0.01029359,
      "balance_loss_clip": 1.05200505,
      "balance_loss_mlp": 1.02203381,
      "epoch": 0.5448205374857211,
      "flos": 11798307757440.0,
      "grad_norm": 1.901934687182875,
      "language_loss": 0.82570177,
      "learning_rate": 1.8071152974618156e-06,
      "loss": 0.84779221,
      "num_input_tokens_seen": 97716565,
      "step": 4531,
      "time_per_iteration": 2.5854718685150146
    },
    {
      "auxiliary_loss_clip": 0.01139922,
      "auxiliary_loss_mlp": 0.00762895,
      "balance_loss_clip": 1.04611719,
      "balance_loss_mlp": 1.00018883,
      "epoch": 0.5449407803763603,
      "flos": 24133766474880.0,
      "grad_norm": 2.110587676923799,
      "language_loss": 0.78649706,
      "learning_rate": 1.806339971584599e-06,
      "loss": 0.80552518,
      "num_input_tokens_seen": 97733225,
      "step": 4532,
      "time_per_iteration": 2.8077473640441895
    },
    {
      "auxiliary_loss_clip": 0.01185498,
      "auxiliary_loss_mlp": 0.01026353,
      "balance_loss_clip": 1.05445945,
      "balance_loss_mlp": 1.01893187,
      "epoch": 0.5450610232669993,
      "flos": 23258551685760.0,
      "grad_norm": 6.907644634360941,
      "language_loss": 0.85150373,
      "learning_rate": 1.8055646750854546e-06,
      "loss": 0.8736223,
      "num_input_tokens_seen": 97752735,
      "step": 4533,
      "time_per_iteration": 2.64298939704895
    },
    {
      "auxiliary_loss_clip": 0.01161976,
      "auxiliary_loss_mlp": 0.01032439,
      "balance_loss_clip": 1.05165577,
      "balance_loss_mlp": 1.02460647,
      "epoch": 0.5451812661576384,
      "flos": 17785083375360.0,
      "grad_norm": 2.9588510249166644,
      "language_loss": 0.81567597,
      "learning_rate": 1.8047894080819945e-06,
      "loss": 0.83762014,
      "num_input_tokens_seen": 97769985,
      "step": 4534,
      "time_per_iteration": 2.6212339401245117
    },
    {
      "auxiliary_loss_clip": 0.0107552,
      "auxiliary_loss_mlp": 0.01003324,
      "balance_loss_clip": 1.01361513,
      "balance_loss_mlp": 1.00194156,
      "epoch": 0.5453015090482776,
      "flos": 71062586513280.0,
      "grad_norm": 0.7303625197151306,
      "language_loss": 0.63176405,
      "learning_rate": 1.8040141706918258e-06,
      "loss": 0.65255249,
      "num_input_tokens_seen": 97831225,
      "step": 4535,
      "time_per_iteration": 4.239135265350342
    },
    {
      "auxiliary_loss_clip": 0.0115744,
      "auxiliary_loss_mlp": 0.01033493,
      "balance_loss_clip": 1.05194902,
      "balance_loss_mlp": 1.02526188,
      "epoch": 0.5454217519389166,
      "flos": 25552201622400.0,
      "grad_norm": 2.0817249009636942,
      "language_loss": 0.76880682,
      "learning_rate": 1.8032389630325525e-06,
      "loss": 0.79071617,
      "num_input_tokens_seen": 97849975,
      "step": 4536,
      "time_per_iteration": 4.225529432296753
    },
    {
      "auxiliary_loss_clip": 0.01152013,
      "auxiliary_loss_mlp": 0.01026975,
      "balance_loss_clip": 1.04570961,
      "balance_loss_mlp": 1.01870775,
      "epoch": 0.5455419948295557,
      "flos": 23658345037440.0,
      "grad_norm": 1.7680439576164435,
      "language_loss": 0.75388366,
      "learning_rate": 1.8024637852217707e-06,
      "loss": 0.77567351,
      "num_input_tokens_seen": 97869700,
      "step": 4537,
      "time_per_iteration": 3.617051601409912
    },
    {
      "auxiliary_loss_clip": 0.01155004,
      "auxiliary_loss_mlp": 0.01029881,
      "balance_loss_clip": 1.05147028,
      "balance_loss_mlp": 1.02186418,
      "epoch": 0.5456622377201948,
      "flos": 23403989854080.0,
      "grad_norm": 1.6896772536276168,
      "language_loss": 0.8450278,
      "learning_rate": 1.8016886373770766e-06,
      "loss": 0.86687666,
      "num_input_tokens_seen": 97888215,
      "step": 4538,
      "time_per_iteration": 2.7369158267974854
    },
    {
      "auxiliary_loss_clip": 0.01153455,
      "auxiliary_loss_mlp": 0.01032974,
      "balance_loss_clip": 1.05047536,
      "balance_loss_mlp": 1.02522492,
      "epoch": 0.5457824806108339,
      "flos": 23988040997760.0,
      "grad_norm": 2.4476252960590394,
      "language_loss": 0.78939152,
      "learning_rate": 1.8009135196160579e-06,
      "loss": 0.81125581,
      "num_input_tokens_seen": 97907090,
      "step": 4539,
      "time_per_iteration": 2.7281315326690674
    },
    {
      "auxiliary_loss_clip": 0.01136008,
      "auxiliary_loss_mlp": 0.01030114,
      "balance_loss_clip": 1.04636788,
      "balance_loss_mlp": 1.02219236,
      "epoch": 0.545902723501473,
      "flos": 22565870835840.0,
      "grad_norm": 1.862390761852363,
      "language_loss": 0.84493172,
      "learning_rate": 1.8001384320563e-06,
      "loss": 0.86659288,
      "num_input_tokens_seen": 97927345,
      "step": 4540,
      "time_per_iteration": 3.65763521194458
    },
    {
      "auxiliary_loss_clip": 0.01075121,
      "auxiliary_loss_mlp": 0.0100203,
      "balance_loss_clip": 1.01305199,
      "balance_loss_mlp": 1.00062335,
      "epoch": 0.5460229663921121,
      "flos": 55198399685760.0,
      "grad_norm": 0.7809450905760909,
      "language_loss": 0.57756686,
      "learning_rate": 1.7993633748153833e-06,
      "loss": 0.59833837,
      "num_input_tokens_seen": 97981950,
      "step": 4541,
      "time_per_iteration": 3.1040353775024414
    },
    {
      "auxiliary_loss_clip": 0.01173766,
      "auxiliary_loss_mlp": 0.0103115,
      "balance_loss_clip": 1.05298626,
      "balance_loss_mlp": 1.02263808,
      "epoch": 0.5461432092827512,
      "flos": 15413866018560.0,
      "grad_norm": 1.729082725606598,
      "language_loss": 0.7300936,
      "learning_rate": 1.7985883480108834e-06,
      "loss": 0.75214279,
      "num_input_tokens_seen": 97999585,
      "step": 4542,
      "time_per_iteration": 2.6568305492401123
    },
    {
      "auxiliary_loss_clip": 0.01163243,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.04979587,
      "balance_loss_mlp": 1.01770532,
      "epoch": 0.5462634521733902,
      "flos": 24024921287040.0,
      "grad_norm": 1.621367202525802,
      "language_loss": 0.72004926,
      "learning_rate": 1.797813351760371e-06,
      "loss": 0.74193358,
      "num_input_tokens_seen": 98021290,
      "step": 4543,
      "time_per_iteration": 2.6556859016418457
    },
    {
      "auxiliary_loss_clip": 0.01183194,
      "auxiliary_loss_mlp": 0.01032123,
      "balance_loss_clip": 1.05114889,
      "balance_loss_mlp": 1.02406406,
      "epoch": 0.5463836950640293,
      "flos": 22820944291200.0,
      "grad_norm": 1.6633343504205937,
      "language_loss": 0.78281111,
      "learning_rate": 1.7970383861814116e-06,
      "loss": 0.8049643,
      "num_input_tokens_seen": 98041060,
      "step": 4544,
      "time_per_iteration": 2.6639328002929688
    },
    {
      "auxiliary_loss_clip": 0.01170919,
      "auxiliary_loss_mlp": 0.01032012,
      "balance_loss_clip": 1.05248427,
      "balance_loss_mlp": 1.02394795,
      "epoch": 0.5465039379546685,
      "flos": 20448290390400.0,
      "grad_norm": 1.9790088211230228,
      "language_loss": 0.73938745,
      "learning_rate": 1.7962634513915684e-06,
      "loss": 0.76141679,
      "num_input_tokens_seen": 98058410,
      "step": 4545,
      "time_per_iteration": 2.64389705657959
    },
    {
      "auxiliary_loss_clip": 0.01180899,
      "auxiliary_loss_mlp": 0.01033344,
      "balance_loss_clip": 1.05189002,
      "balance_loss_mlp": 1.02514827,
      "epoch": 0.5466241808453075,
      "flos": 17343310003200.0,
      "grad_norm": 1.832875381525187,
      "language_loss": 0.79387629,
      "learning_rate": 1.7954885475083969e-06,
      "loss": 0.8160187,
      "num_input_tokens_seen": 98076080,
      "step": 4546,
      "time_per_iteration": 2.6242525577545166
    },
    {
      "auxiliary_loss_clip": 0.01188018,
      "auxiliary_loss_mlp": 0.01030913,
      "balance_loss_clip": 1.05474842,
      "balance_loss_mlp": 1.02303898,
      "epoch": 0.5467444237359466,
      "flos": 21617039122560.0,
      "grad_norm": 2.214701115257723,
      "language_loss": 0.72015762,
      "learning_rate": 1.7947136746494513e-06,
      "loss": 0.74234688,
      "num_input_tokens_seen": 98096995,
      "step": 4547,
      "time_per_iteration": 2.6596758365631104
    },
    {
      "auxiliary_loss_clip": 0.01168771,
      "auxiliary_loss_mlp": 0.01026795,
      "balance_loss_clip": 1.0522635,
      "balance_loss_mlp": 1.01890922,
      "epoch": 0.5468646666265857,
      "flos": 24170467196160.0,
      "grad_norm": 2.1203361590030942,
      "language_loss": 0.87858027,
      "learning_rate": 1.793938832932277e-06,
      "loss": 0.90053594,
      "num_input_tokens_seen": 98115105,
      "step": 4548,
      "time_per_iteration": 2.7142553329467773
    },
    {
      "auxiliary_loss_clip": 0.01183268,
      "auxiliary_loss_mlp": 0.01028493,
      "balance_loss_clip": 1.05294681,
      "balance_loss_mlp": 1.02055943,
      "epoch": 0.5469849095172248,
      "flos": 27527001505920.0,
      "grad_norm": 2.158356809372934,
      "language_loss": 0.70749694,
      "learning_rate": 1.7931640224744185e-06,
      "loss": 0.72961456,
      "num_input_tokens_seen": 98135655,
      "step": 4549,
      "time_per_iteration": 2.6343131065368652
    },
    {
      "auxiliary_loss_clip": 0.01129578,
      "auxiliary_loss_mlp": 0.01027375,
      "balance_loss_clip": 1.0429287,
      "balance_loss_mlp": 1.01944768,
      "epoch": 0.5471051524078638,
      "flos": 27964680727680.0,
      "grad_norm": 1.7834254866817754,
      "language_loss": 0.73573101,
      "learning_rate": 1.7923892433934127e-06,
      "loss": 0.7573005,
      "num_input_tokens_seen": 98156730,
      "step": 4550,
      "time_per_iteration": 2.7674665451049805
    },
    {
      "auxiliary_loss_clip": 0.01158711,
      "auxiliary_loss_mlp": 0.0076295,
      "balance_loss_clip": 1.05155396,
      "balance_loss_mlp": 1.00030696,
      "epoch": 0.547225395298503,
      "flos": 18150510389760.0,
      "grad_norm": 1.847272015928521,
      "language_loss": 0.78900027,
      "learning_rate": 1.7916144958067939e-06,
      "loss": 0.80821693,
      "num_input_tokens_seen": 98174590,
      "step": 4551,
      "time_per_iteration": 2.6805801391601562
    },
    {
      "auxiliary_loss_clip": 0.01172015,
      "auxiliary_loss_mlp": 0.01030884,
      "balance_loss_clip": 1.05178595,
      "balance_loss_mlp": 1.02285492,
      "epoch": 0.5473456381891421,
      "flos": 21361498790400.0,
      "grad_norm": 1.7096268688851672,
      "language_loss": 0.79019624,
      "learning_rate": 1.7908397798320905e-06,
      "loss": 0.81222522,
      "num_input_tokens_seen": 98194325,
      "step": 4552,
      "time_per_iteration": 2.776017189025879
    },
    {
      "auxiliary_loss_clip": 0.01171461,
      "auxiliary_loss_mlp": 0.00763405,
      "balance_loss_clip": 1.05235422,
      "balance_loss_mlp": 1.00030422,
      "epoch": 0.5474658810797811,
      "flos": 19932145908480.0,
      "grad_norm": 2.2476663014557485,
      "language_loss": 0.74315387,
      "learning_rate": 1.7900650955868265e-06,
      "loss": 0.76250255,
      "num_input_tokens_seen": 98213970,
      "step": 4553,
      "time_per_iteration": 2.6954760551452637
    },
    {
      "auxiliary_loss_clip": 0.01171459,
      "auxiliary_loss_mlp": 0.00762899,
      "balance_loss_clip": 1.05623746,
      "balance_loss_mlp": 1.00028205,
      "epoch": 0.5475861239704203,
      "flos": 50476217264640.0,
      "grad_norm": 1.4388617124099763,
      "language_loss": 0.76473844,
      "learning_rate": 1.7892904431885202e-06,
      "loss": 0.78408194,
      "num_input_tokens_seen": 98241145,
      "step": 4554,
      "time_per_iteration": 2.9159655570983887
    },
    {
      "auxiliary_loss_clip": 0.0112515,
      "auxiliary_loss_mlp": 0.01025334,
      "balance_loss_clip": 1.0422107,
      "balance_loss_mlp": 1.01776373,
      "epoch": 0.5477063668610593,
      "flos": 20705123612160.0,
      "grad_norm": 1.949644924001728,
      "language_loss": 0.75754976,
      "learning_rate": 1.788515822754686e-06,
      "loss": 0.77905464,
      "num_input_tokens_seen": 98261565,
      "step": 4555,
      "time_per_iteration": 2.7779712677001953
    },
    {
      "auxiliary_loss_clip": 0.01141499,
      "auxiliary_loss_mlp": 0.01030776,
      "balance_loss_clip": 1.04708934,
      "balance_loss_mlp": 1.02241969,
      "epoch": 0.5478266097516984,
      "flos": 19609740408960.0,
      "grad_norm": 2.1267169766157408,
      "language_loss": 0.78020424,
      "learning_rate": 1.7877412344028335e-06,
      "loss": 0.80192697,
      "num_input_tokens_seen": 98281370,
      "step": 4556,
      "time_per_iteration": 2.7188591957092285
    },
    {
      "auxiliary_loss_clip": 0.0117127,
      "auxiliary_loss_mlp": 0.01028882,
      "balance_loss_clip": 1.05191851,
      "balance_loss_mlp": 1.02141929,
      "epoch": 0.5479468526423376,
      "flos": 12896599962240.0,
      "grad_norm": 2.3325921088102066,
      "language_loss": 0.77402323,
      "learning_rate": 1.7869666782504668e-06,
      "loss": 0.7960248,
      "num_input_tokens_seen": 98297950,
      "step": 4557,
      "time_per_iteration": 2.7008039951324463
    },
    {
      "auxiliary_loss_clip": 0.01140197,
      "auxiliary_loss_mlp": 0.01027108,
      "balance_loss_clip": 1.04468465,
      "balance_loss_mlp": 1.01934743,
      "epoch": 0.5480670955329766,
      "flos": 18588800142720.0,
      "grad_norm": 2.6828276861388702,
      "language_loss": 0.68524241,
      "learning_rate": 1.7861921544150867e-06,
      "loss": 0.7069155,
      "num_input_tokens_seen": 98316800,
      "step": 4558,
      "time_per_iteration": 2.666175127029419
    },
    {
      "auxiliary_loss_clip": 0.0109982,
      "auxiliary_loss_mlp": 0.00762887,
      "balance_loss_clip": 1.04595375,
      "balance_loss_mlp": 1.00018752,
      "epoch": 0.5481873384236157,
      "flos": 15954608338560.0,
      "grad_norm": 1.9300026080164525,
      "language_loss": 0.7678709,
      "learning_rate": 1.7854176630141856e-06,
      "loss": 0.78649795,
      "num_input_tokens_seen": 98333935,
      "step": 4559,
      "time_per_iteration": 2.7772793769836426
    },
    {
      "auxiliary_loss_clip": 0.01185222,
      "auxiliary_loss_mlp": 0.01027723,
      "balance_loss_clip": 1.05295575,
      "balance_loss_mlp": 1.01909828,
      "epoch": 0.5483075813142548,
      "flos": 22783812606720.0,
      "grad_norm": 2.3093519149136963,
      "language_loss": 0.83958107,
      "learning_rate": 1.784643204165255e-06,
      "loss": 0.86171055,
      "num_input_tokens_seen": 98353255,
      "step": 4560,
      "time_per_iteration": 2.6628353595733643
    },
    {
      "auxiliary_loss_clip": 0.01165959,
      "auxiliary_loss_mlp": 0.01029378,
      "balance_loss_clip": 1.05320811,
      "balance_loss_mlp": 1.0216887,
      "epoch": 0.5484278242048939,
      "flos": 19317212046720.0,
      "grad_norm": 1.927334921178401,
      "language_loss": 0.77230191,
      "learning_rate": 1.7838687779857783e-06,
      "loss": 0.79425532,
      "num_input_tokens_seen": 98371130,
      "step": 4561,
      "time_per_iteration": 3.650987386703491
    },
    {
      "auxiliary_loss_clip": 0.01145757,
      "auxiliary_loss_mlp": 0.010332,
      "balance_loss_clip": 1.04698718,
      "balance_loss_mlp": 1.02559733,
      "epoch": 0.5485480670955329,
      "flos": 22816024128000.0,
      "grad_norm": 1.750582957506157,
      "language_loss": 0.6399399,
      "learning_rate": 1.7830943845932366e-06,
      "loss": 0.66172951,
      "num_input_tokens_seen": 98390455,
      "step": 4562,
      "time_per_iteration": 3.948986768722534
    },
    {
      "auxiliary_loss_clip": 0.01159143,
      "auxiliary_loss_mlp": 0.01029073,
      "balance_loss_clip": 1.05190825,
      "balance_loss_mlp": 1.0213306,
      "epoch": 0.5486683099861721,
      "flos": 22671304231680.0,
      "grad_norm": 1.6967465181504904,
      "language_loss": 0.74984336,
      "learning_rate": 1.7823200241051044e-06,
      "loss": 0.77172559,
      "num_input_tokens_seen": 98409370,
      "step": 4563,
      "time_per_iteration": 3.6697232723236084
    },
    {
      "auxiliary_loss_clip": 0.011843,
      "auxiliary_loss_mlp": 0.01027382,
      "balance_loss_clip": 1.05460441,
      "balance_loss_mlp": 1.01988339,
      "epoch": 0.5487885528768112,
      "flos": 23149383275520.0,
      "grad_norm": 2.0558343119048406,
      "language_loss": 0.80022717,
      "learning_rate": 1.7815456966388513e-06,
      "loss": 0.82234401,
      "num_input_tokens_seen": 98428465,
      "step": 4564,
      "time_per_iteration": 2.6580450534820557
    },
    {
      "auxiliary_loss_clip": 0.01141622,
      "auxiliary_loss_mlp": 0.01031508,
      "balance_loss_clip": 1.04683816,
      "balance_loss_mlp": 1.02285957,
      "epoch": 0.5489087957674502,
      "flos": 22053928245120.0,
      "grad_norm": 2.1046618929031395,
      "language_loss": 0.81191516,
      "learning_rate": 1.780771402311943e-06,
      "loss": 0.83364642,
      "num_input_tokens_seen": 98447300,
      "step": 4565,
      "time_per_iteration": 2.7053802013397217
    },
    {
      "auxiliary_loss_clip": 0.01156897,
      "auxiliary_loss_mlp": 0.01026204,
      "balance_loss_clip": 1.05154777,
      "balance_loss_mlp": 1.0174365,
      "epoch": 0.5490290386580894,
      "flos": 24315977191680.0,
      "grad_norm": 1.7679812929156262,
      "language_loss": 0.78456712,
      "learning_rate": 1.7799971412418374e-06,
      "loss": 0.80639815,
      "num_input_tokens_seen": 98468695,
      "step": 4566,
      "time_per_iteration": 3.5916011333465576
    },
    {
      "auxiliary_loss_clip": 0.01139046,
      "auxiliary_loss_mlp": 0.01027038,
      "balance_loss_clip": 1.04667687,
      "balance_loss_mlp": 1.01863945,
      "epoch": 0.5491492815487284,
      "flos": 18294942977280.0,
      "grad_norm": 2.887471909176422,
      "language_loss": 0.74158472,
      "learning_rate": 1.7792229135459918e-06,
      "loss": 0.76324558,
      "num_input_tokens_seen": 98485345,
      "step": 4567,
      "time_per_iteration": 2.649627923965454
    },
    {
      "auxiliary_loss_clip": 0.01036907,
      "auxiliary_loss_mlp": 0.01001141,
      "balance_loss_clip": 1.02226853,
      "balance_loss_mlp": 0.9998123,
      "epoch": 0.5492695244393675,
      "flos": 64550257050240.0,
      "grad_norm": 0.7402766086861442,
      "language_loss": 0.61564469,
      "learning_rate": 1.7784487193418538e-06,
      "loss": 0.63602519,
      "num_input_tokens_seen": 98543195,
      "step": 4568,
      "time_per_iteration": 3.183910369873047
    },
    {
      "auxiliary_loss_clip": 0.01123163,
      "auxiliary_loss_mlp": 0.01028216,
      "balance_loss_clip": 1.04024243,
      "balance_loss_mlp": 1.01927483,
      "epoch": 0.5493897673300067,
      "flos": 17379579761280.0,
      "grad_norm": 2.3374596128474385,
      "language_loss": 0.61091942,
      "learning_rate": 1.7776745587468698e-06,
      "loss": 0.63243324,
      "num_input_tokens_seen": 98560620,
      "step": 4569,
      "time_per_iteration": 2.707425594329834
    },
    {
      "auxiliary_loss_clip": 0.01179964,
      "auxiliary_loss_mlp": 0.01028095,
      "balance_loss_clip": 1.04986048,
      "balance_loss_mlp": 1.02018547,
      "epoch": 0.5495100102206457,
      "flos": 19901765980800.0,
      "grad_norm": 2.3719672218247805,
      "language_loss": 0.8204518,
      "learning_rate": 1.7769004318784776e-06,
      "loss": 0.8425324,
      "num_input_tokens_seen": 98578265,
      "step": 4570,
      "time_per_iteration": 2.6281685829162598
    },
    {
      "auxiliary_loss_clip": 0.01168434,
      "auxiliary_loss_mlp": 0.01022627,
      "balance_loss_clip": 1.05067277,
      "balance_loss_mlp": 1.01502168,
      "epoch": 0.5496302531112848,
      "flos": 16727190992640.0,
      "grad_norm": 1.700774211398538,
      "language_loss": 0.80817425,
      "learning_rate": 1.776126338854113e-06,
      "loss": 0.83008492,
      "num_input_tokens_seen": 98596055,
      "step": 4571,
      "time_per_iteration": 2.624265432357788
    },
    {
      "auxiliary_loss_clip": 0.01164698,
      "auxiliary_loss_mlp": 0.01027553,
      "balance_loss_clip": 1.05206084,
      "balance_loss_mlp": 1.01974809,
      "epoch": 0.5497504960019239,
      "flos": 24572343536640.0,
      "grad_norm": 1.7097697227246709,
      "language_loss": 0.84600002,
      "learning_rate": 1.7753522797912044e-06,
      "loss": 0.86792254,
      "num_input_tokens_seen": 98616140,
      "step": 4572,
      "time_per_iteration": 2.717280864715576
    },
    {
      "auxiliary_loss_clip": 0.01162467,
      "auxiliary_loss_mlp": 0.01032881,
      "balance_loss_clip": 1.04970026,
      "balance_loss_mlp": 1.02411938,
      "epoch": 0.549870738892563,
      "flos": 15450494912640.0,
      "grad_norm": 2.11000163455766,
      "language_loss": 0.69871449,
      "learning_rate": 1.7745782548071765e-06,
      "loss": 0.72066796,
      "num_input_tokens_seen": 98633035,
      "step": 4573,
      "time_per_iteration": 2.6669082641601562
    },
    {
      "auxiliary_loss_clip": 0.01136988,
      "auxiliary_loss_mlp": 0.01027152,
      "balance_loss_clip": 1.05155861,
      "balance_loss_mlp": 1.01945734,
      "epoch": 0.549990981783202,
      "flos": 21069114082560.0,
      "grad_norm": 1.7683869966856318,
      "language_loss": 0.74050939,
      "learning_rate": 1.7738042640194482e-06,
      "loss": 0.76215082,
      "num_input_tokens_seen": 98652700,
      "step": 4574,
      "time_per_iteration": 2.774610757827759
    },
    {
      "auxiliary_loss_clip": 0.01180657,
      "auxiliary_loss_mlp": 0.01026154,
      "balance_loss_clip": 1.05131233,
      "balance_loss_mlp": 1.01752973,
      "epoch": 0.5501112246738411,
      "flos": 21395901041280.0,
      "grad_norm": 1.8600651152152181,
      "language_loss": 0.70856273,
      "learning_rate": 1.7730303075454335e-06,
      "loss": 0.73063082,
      "num_input_tokens_seen": 98671590,
      "step": 4575,
      "time_per_iteration": 2.655052423477173
    },
    {
      "auxiliary_loss_clip": 0.01146434,
      "auxiliary_loss_mlp": 0.01026514,
      "balance_loss_clip": 1.04912949,
      "balance_loss_mlp": 1.01846719,
      "epoch": 0.5502314675644803,
      "flos": 17456931699840.0,
      "grad_norm": 1.870820588835222,
      "language_loss": 0.84624004,
      "learning_rate": 1.7722563855025402e-06,
      "loss": 0.86796951,
      "num_input_tokens_seen": 98689620,
      "step": 4576,
      "time_per_iteration": 2.760298252105713
    },
    {
      "auxiliary_loss_clip": 0.01153204,
      "auxiliary_loss_mlp": 0.01028006,
      "balance_loss_clip": 1.04416394,
      "balance_loss_mlp": 1.02021575,
      "epoch": 0.5503517104551193,
      "flos": 24310410583680.0,
      "grad_norm": 2.4447962244975723,
      "language_loss": 0.7098543,
      "learning_rate": 1.7714824980081721e-06,
      "loss": 0.73166645,
      "num_input_tokens_seen": 98708915,
      "step": 4577,
      "time_per_iteration": 2.6460909843444824
    },
    {
      "auxiliary_loss_clip": 0.01165777,
      "auxiliary_loss_mlp": 0.01023969,
      "balance_loss_clip": 1.05292392,
      "balance_loss_mlp": 1.0160594,
      "epoch": 0.5504719533457584,
      "flos": 22419427086720.0,
      "grad_norm": 1.6554825923371679,
      "language_loss": 0.73829013,
      "learning_rate": 1.7707086451797276e-06,
      "loss": 0.76018757,
      "num_input_tokens_seen": 98729790,
      "step": 4578,
      "time_per_iteration": 2.7348670959472656
    },
    {
      "auxiliary_loss_clip": 0.01040903,
      "auxiliary_loss_mlp": 0.01001147,
      "balance_loss_clip": 1.00943279,
      "balance_loss_mlp": 0.99981195,
      "epoch": 0.5505921962363975,
      "flos": 67294155968640.0,
      "grad_norm": 0.6973790928398105,
      "language_loss": 0.52266467,
      "learning_rate": 1.7699348271345993e-06,
      "loss": 0.54308522,
      "num_input_tokens_seen": 98792415,
      "step": 4579,
      "time_per_iteration": 3.2107629776000977
    },
    {
      "auxiliary_loss_clip": 0.01036275,
      "auxiliary_loss_mlp": 0.01003384,
      "balance_loss_clip": 1.00981236,
      "balance_loss_mlp": 1.00201893,
      "epoch": 0.5507124391270366,
      "flos": 45685125578880.0,
      "grad_norm": 0.9151903958255929,
      "language_loss": 0.54422748,
      "learning_rate": 1.7691610439901753e-06,
      "loss": 0.56462407,
      "num_input_tokens_seen": 98855350,
      "step": 4580,
      "time_per_iteration": 3.3387930393218994
    },
    {
      "auxiliary_loss_clip": 0.01168604,
      "auxiliary_loss_mlp": 0.01026798,
      "balance_loss_clip": 1.05030751,
      "balance_loss_mlp": 1.0189538,
      "epoch": 0.5508326820176757,
      "flos": 22273845264000.0,
      "grad_norm": 1.9330032795150789,
      "language_loss": 0.75688183,
      "learning_rate": 1.7683872958638367e-06,
      "loss": 0.77883589,
      "num_input_tokens_seen": 98874230,
      "step": 4581,
      "time_per_iteration": 2.6635684967041016
    },
    {
      "auxiliary_loss_clip": 0.01150411,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.04720318,
      "balance_loss_mlp": 1.01896596,
      "epoch": 0.5509529249083148,
      "flos": 20012442762240.0,
      "grad_norm": 2.723327461850464,
      "language_loss": 0.84253252,
      "learning_rate": 1.7676135828729614e-06,
      "loss": 0.86430687,
      "num_input_tokens_seen": 98893940,
      "step": 4582,
      "time_per_iteration": 2.7095084190368652
    },
    {
      "auxiliary_loss_clip": 0.0116634,
      "auxiliary_loss_mlp": 0.01028743,
      "balance_loss_clip": 1.05141902,
      "balance_loss_mlp": 1.02037442,
      "epoch": 0.5510731677989539,
      "flos": 21834801325440.0,
      "grad_norm": 1.9475738249660668,
      "language_loss": 0.8302815,
      "learning_rate": 1.7668399051349205e-06,
      "loss": 0.85223234,
      "num_input_tokens_seen": 98913620,
      "step": 4583,
      "time_per_iteration": 2.713707447052002
    },
    {
      "auxiliary_loss_clip": 0.01135973,
      "auxiliary_loss_mlp": 0.0102896,
      "balance_loss_clip": 1.04672432,
      "balance_loss_mlp": 1.02047276,
      "epoch": 0.5511934106895929,
      "flos": 21467901853440.0,
      "grad_norm": 1.9670512086049303,
      "language_loss": 0.83267999,
      "learning_rate": 1.766066262767081e-06,
      "loss": 0.85432929,
      "num_input_tokens_seen": 98931460,
      "step": 4584,
      "time_per_iteration": 2.701094388961792
    },
    {
      "auxiliary_loss_clip": 0.01150457,
      "auxiliary_loss_mlp": 0.01025034,
      "balance_loss_clip": 1.05073595,
      "balance_loss_mlp": 1.01755333,
      "epoch": 0.5513136535802321,
      "flos": 21068934514560.0,
      "grad_norm": 2.0095124304268137,
      "language_loss": 0.77069181,
      "learning_rate": 1.765292655886803e-06,
      "loss": 0.79244673,
      "num_input_tokens_seen": 98950105,
      "step": 4585,
      "time_per_iteration": 2.6748225688934326
    },
    {
      "auxiliary_loss_clip": 0.01150718,
      "auxiliary_loss_mlp": 0.01027011,
      "balance_loss_clip": 1.04962254,
      "balance_loss_mlp": 1.01936972,
      "epoch": 0.5514338964708712,
      "flos": 27815004754560.0,
      "grad_norm": 2.382590096800934,
      "language_loss": 0.71125662,
      "learning_rate": 1.764519084611443e-06,
      "loss": 0.7330339,
      "num_input_tokens_seen": 98970560,
      "step": 4586,
      "time_per_iteration": 2.7616167068481445
    },
    {
      "auxiliary_loss_clip": 0.01154655,
      "auxiliary_loss_mlp": 0.01028031,
      "balance_loss_clip": 1.04899561,
      "balance_loss_mlp": 1.01897752,
      "epoch": 0.5515541393615102,
      "flos": 21908525990400.0,
      "grad_norm": 2.0740705594237583,
      "language_loss": 0.77649224,
      "learning_rate": 1.7637455490583505e-06,
      "loss": 0.7983191,
      "num_input_tokens_seen": 98989885,
      "step": 4587,
      "time_per_iteration": 3.9394783973693848
    },
    {
      "auxiliary_loss_clip": 0.01166687,
      "auxiliary_loss_mlp": 0.01025321,
      "balance_loss_clip": 1.05013049,
      "balance_loss_mlp": 1.01757228,
      "epoch": 0.5516743822521494,
      "flos": 20485422074880.0,
      "grad_norm": 2.0272676315352447,
      "language_loss": 0.77459568,
      "learning_rate": 1.7629720493448701e-06,
      "loss": 0.79651576,
      "num_input_tokens_seen": 99007180,
      "step": 4588,
      "time_per_iteration": 3.579955816268921
    },
    {
      "auxiliary_loss_clip": 0.01161721,
      "auxiliary_loss_mlp": 0.01027529,
      "balance_loss_clip": 1.04913807,
      "balance_loss_mlp": 1.01957178,
      "epoch": 0.5517946251427884,
      "flos": 14940383915520.0,
      "grad_norm": 1.7318988537329254,
      "language_loss": 0.85487777,
      "learning_rate": 1.7621985855883418e-06,
      "loss": 0.87677026,
      "num_input_tokens_seen": 99023880,
      "step": 4589,
      "time_per_iteration": 3.6680572032928467
    },
    {
      "auxiliary_loss_clip": 0.011481,
      "auxiliary_loss_mlp": 0.0102468,
      "balance_loss_clip": 1.04896379,
      "balance_loss_mlp": 1.01708615,
      "epoch": 0.5519148680334275,
      "flos": 18404865573120.0,
      "grad_norm": 1.7779379178617607,
      "language_loss": 0.72686255,
      "learning_rate": 1.7614251579060983e-06,
      "loss": 0.74859035,
      "num_input_tokens_seen": 99042475,
      "step": 4590,
      "time_per_iteration": 2.613234043121338
    },
    {
      "auxiliary_loss_clip": 0.0114206,
      "auxiliary_loss_mlp": 0.01026784,
      "balance_loss_clip": 1.04733634,
      "balance_loss_mlp": 1.01854658,
      "epoch": 0.5520351109240667,
      "flos": 25113337251840.0,
      "grad_norm": 1.6529284590743913,
      "language_loss": 0.84686124,
      "learning_rate": 1.76065176641547e-06,
      "loss": 0.86854959,
      "num_input_tokens_seen": 99065185,
      "step": 4591,
      "time_per_iteration": 2.7440178394317627
    },
    {
      "auxiliary_loss_clip": 0.01168365,
      "auxiliary_loss_mlp": 0.01026271,
      "balance_loss_clip": 1.04851234,
      "balance_loss_mlp": 1.01809883,
      "epoch": 0.5521553538147057,
      "flos": 21069545045760.0,
      "grad_norm": 1.6310467440295227,
      "language_loss": 0.77907443,
      "learning_rate": 1.759878411233777e-06,
      "loss": 0.80102074,
      "num_input_tokens_seen": 99083645,
      "step": 4592,
      "time_per_iteration": 3.5074656009674072
    },
    {
      "auxiliary_loss_clip": 0.01167568,
      "auxiliary_loss_mlp": 0.01029003,
      "balance_loss_clip": 1.05117142,
      "balance_loss_mlp": 1.02132642,
      "epoch": 0.5522755967053448,
      "flos": 18879999701760.0,
      "grad_norm": 2.162108611188334,
      "language_loss": 0.75629735,
      "learning_rate": 1.7591050924783388e-06,
      "loss": 0.77826309,
      "num_input_tokens_seen": 99100835,
      "step": 4593,
      "time_per_iteration": 2.6860623359680176
    },
    {
      "auxiliary_loss_clip": 0.01028996,
      "auxiliary_loss_mlp": 0.01001393,
      "balance_loss_clip": 1.00992334,
      "balance_loss_mlp": 0.99985522,
      "epoch": 0.5523958395959839,
      "flos": 64675622494080.0,
      "grad_norm": 0.8426295675922896,
      "language_loss": 0.57951707,
      "learning_rate": 1.7583318102664661e-06,
      "loss": 0.59982097,
      "num_input_tokens_seen": 99168400,
      "step": 4594,
      "time_per_iteration": 3.382849931716919
    },
    {
      "auxiliary_loss_clip": 0.01168183,
      "auxiliary_loss_mlp": 0.01026158,
      "balance_loss_clip": 1.04707026,
      "balance_loss_mlp": 1.01831961,
      "epoch": 0.552516082486623,
      "flos": 10889732211840.0,
      "grad_norm": 1.8936759637587908,
      "language_loss": 0.78866303,
      "learning_rate": 1.757558564715466e-06,
      "loss": 0.81060648,
      "num_input_tokens_seen": 99186475,
      "step": 4595,
      "time_per_iteration": 2.853739023208618
    },
    {
      "auxiliary_loss_clip": 0.01167928,
      "auxiliary_loss_mlp": 0.01026018,
      "balance_loss_clip": 1.04928148,
      "balance_loss_mlp": 1.01750088,
      "epoch": 0.552636325377262,
      "flos": 22199797376640.0,
      "grad_norm": 2.70231832944461,
      "language_loss": 0.74091738,
      "learning_rate": 1.7567853559426386e-06,
      "loss": 0.76285684,
      "num_input_tokens_seen": 99203525,
      "step": 4596,
      "time_per_iteration": 2.660952568054199
    },
    {
      "auxiliary_loss_clip": 0.0116651,
      "auxiliary_loss_mlp": 0.01032361,
      "balance_loss_clip": 1.04919434,
      "balance_loss_mlp": 1.02439237,
      "epoch": 0.5527565682679012,
      "flos": 23988184652160.0,
      "grad_norm": 2.1591767732668488,
      "language_loss": 0.75077784,
      "learning_rate": 1.7560121840652797e-06,
      "loss": 0.77276653,
      "num_input_tokens_seen": 99222910,
      "step": 4597,
      "time_per_iteration": 2.659105062484741
    },
    {
      "auxiliary_loss_clip": 0.01128982,
      "auxiliary_loss_mlp": 0.01030429,
      "balance_loss_clip": 1.04729497,
      "balance_loss_mlp": 1.02201283,
      "epoch": 0.5528768111585403,
      "flos": 19719267955200.0,
      "grad_norm": 1.7458067378315032,
      "language_loss": 0.69421947,
      "learning_rate": 1.7552390492006782e-06,
      "loss": 0.71581364,
      "num_input_tokens_seen": 99241230,
      "step": 4598,
      "time_per_iteration": 2.7106409072875977
    },
    {
      "auxiliary_loss_clip": 0.01132858,
      "auxiliary_loss_mlp": 0.00762722,
      "balance_loss_clip": 1.04393899,
      "balance_loss_mlp": 1.00028527,
      "epoch": 0.5529970540491793,
      "flos": 26215975002240.0,
      "grad_norm": 1.7861366817599131,
      "language_loss": 0.65220028,
      "learning_rate": 1.7544659514661184e-06,
      "loss": 0.67115611,
      "num_input_tokens_seen": 99264320,
      "step": 4599,
      "time_per_iteration": 2.7875328063964844
    },
    {
      "auxiliary_loss_clip": 0.01149911,
      "auxiliary_loss_mlp": 0.01021289,
      "balance_loss_clip": 1.04735541,
      "balance_loss_mlp": 1.01336753,
      "epoch": 0.5531172969398185,
      "flos": 24425971614720.0,
      "grad_norm": 2.044359738319887,
      "language_loss": 0.79711807,
      "learning_rate": 1.7536928909788786e-06,
      "loss": 0.81883001,
      "num_input_tokens_seen": 99283625,
      "step": 4600,
      "time_per_iteration": 2.704087495803833
    },
    {
      "auxiliary_loss_clip": 0.01034077,
      "auxiliary_loss_mlp": 0.0100157,
      "balance_loss_clip": 1.01104271,
      "balance_loss_mlp": 1.00003183,
      "epoch": 0.5532375398304575,
      "flos": 64907316195840.0,
      "grad_norm": 0.8757320857935373,
      "language_loss": 0.61975348,
      "learning_rate": 1.752919867856231e-06,
      "loss": 0.6401099,
      "num_input_tokens_seen": 99335270,
      "step": 4601,
      "time_per_iteration": 3.1664230823516846
    },
    {
      "auxiliary_loss_clip": 0.01145958,
      "auxiliary_loss_mlp": 0.01024211,
      "balance_loss_clip": 1.04607272,
      "balance_loss_mlp": 1.01703477,
      "epoch": 0.5533577827210966,
      "flos": 19683105937920.0,
      "grad_norm": 1.6885803237073533,
      "language_loss": 0.79031229,
      "learning_rate": 1.7521468822154436e-06,
      "loss": 0.81201398,
      "num_input_tokens_seen": 99354185,
      "step": 4602,
      "time_per_iteration": 2.767529010772705
    },
    {
      "auxiliary_loss_clip": 0.01147079,
      "auxiliary_loss_mlp": 0.01025675,
      "balance_loss_clip": 1.04889083,
      "balance_loss_mlp": 1.01771736,
      "epoch": 0.5534780256117358,
      "flos": 32306496076800.0,
      "grad_norm": 2.0410691400504954,
      "language_loss": 0.75485134,
      "learning_rate": 1.751373934173777e-06,
      "loss": 0.7765789,
      "num_input_tokens_seen": 99376930,
      "step": 4603,
      "time_per_iteration": 2.8398706912994385
    },
    {
      "auxiliary_loss_clip": 0.01182593,
      "auxiliary_loss_mlp": 0.01035857,
      "balance_loss_clip": 1.0510397,
      "balance_loss_mlp": 1.02772105,
      "epoch": 0.5535982685023748,
      "flos": 23222425582080.0,
      "grad_norm": 1.5974547110112078,
      "language_loss": 0.7300936,
      "learning_rate": 1.750601023848487e-06,
      "loss": 0.75227809,
      "num_input_tokens_seen": 99397655,
      "step": 4604,
      "time_per_iteration": 2.69842267036438
    },
    {
      "auxiliary_loss_clip": 0.01180322,
      "auxiliary_loss_mlp": 0.00762841,
      "balance_loss_clip": 1.05138636,
      "balance_loss_mlp": 1.00025272,
      "epoch": 0.5537185113930139,
      "flos": 24352534258560.0,
      "grad_norm": 1.9817746293966638,
      "language_loss": 0.73789722,
      "learning_rate": 1.749828151356823e-06,
      "loss": 0.75732887,
      "num_input_tokens_seen": 99417850,
      "step": 4605,
      "time_per_iteration": 2.6311209201812744
    },
    {
      "auxiliary_loss_clip": 0.01153504,
      "auxiliary_loss_mlp": 0.01024457,
      "balance_loss_clip": 1.04943669,
      "balance_loss_mlp": 1.01721251,
      "epoch": 0.553838754283653,
      "flos": 23549068886400.0,
      "grad_norm": 1.642492212969575,
      "language_loss": 0.76012105,
      "learning_rate": 1.7490553168160297e-06,
      "loss": 0.78190064,
      "num_input_tokens_seen": 99438920,
      "step": 4606,
      "time_per_iteration": 2.6645493507385254
    },
    {
      "auxiliary_loss_clip": 0.01151513,
      "auxiliary_loss_mlp": 0.01023762,
      "balance_loss_clip": 1.04758084,
      "balance_loss_mlp": 1.01636553,
      "epoch": 0.5539589971742921,
      "flos": 17275044205440.0,
      "grad_norm": 2.0078681360448463,
      "language_loss": 0.76447642,
      "learning_rate": 1.748282520343345e-06,
      "loss": 0.78622913,
      "num_input_tokens_seen": 99457950,
      "step": 4607,
      "time_per_iteration": 2.7047903537750244
    },
    {
      "auxiliary_loss_clip": 0.01173308,
      "auxiliary_loss_mlp": 0.01030424,
      "balance_loss_clip": 1.04959357,
      "balance_loss_mlp": 1.02219319,
      "epoch": 0.5540792400649311,
      "flos": 27564169104000.0,
      "grad_norm": 1.6689292462961036,
      "language_loss": 0.78341734,
      "learning_rate": 1.7475097620560023e-06,
      "loss": 0.80545461,
      "num_input_tokens_seen": 99478015,
      "step": 4608,
      "time_per_iteration": 2.703479290008545
    },
    {
      "auxiliary_loss_clip": 0.01179599,
      "auxiliary_loss_mlp": 0.01024003,
      "balance_loss_clip": 1.05101037,
      "balance_loss_mlp": 1.01624823,
      "epoch": 0.5541994829555702,
      "flos": 23878657105920.0,
      "grad_norm": 2.054659598843556,
      "language_loss": 0.71066314,
      "learning_rate": 1.746737042071228e-06,
      "loss": 0.73269916,
      "num_input_tokens_seen": 99496520,
      "step": 4609,
      "time_per_iteration": 2.6678073406219482
    },
    {
      "auxiliary_loss_clip": 0.01146303,
      "auxiliary_loss_mlp": 0.01026708,
      "balance_loss_clip": 1.04665279,
      "balance_loss_mlp": 1.01940095,
      "epoch": 0.5543197258462094,
      "flos": 20115721342080.0,
      "grad_norm": 1.7763299925563623,
      "language_loss": 0.79068112,
      "learning_rate": 1.7459643605062424e-06,
      "loss": 0.81241119,
      "num_input_tokens_seen": 99513780,
      "step": 4610,
      "time_per_iteration": 2.64394474029541
    },
    {
      "auxiliary_loss_clip": 0.01119813,
      "auxiliary_loss_mlp": 0.01022554,
      "balance_loss_clip": 1.04517019,
      "balance_loss_mlp": 1.01460826,
      "epoch": 0.5544399687368484,
      "flos": 20916565021440.0,
      "grad_norm": 1.6245136035390824,
      "language_loss": 0.80567098,
      "learning_rate": 1.745191717478262e-06,
      "loss": 0.82709467,
      "num_input_tokens_seen": 99532360,
      "step": 4611,
      "time_per_iteration": 2.860574722290039
    },
    {
      "auxiliary_loss_clip": 0.01146385,
      "auxiliary_loss_mlp": 0.01026764,
      "balance_loss_clip": 1.04743195,
      "balance_loss_mlp": 1.01930177,
      "epoch": 0.5545602116274875,
      "flos": 25518661297920.0,
      "grad_norm": 1.968169901351404,
      "language_loss": 0.79358405,
      "learning_rate": 1.7444191131044948e-06,
      "loss": 0.81531554,
      "num_input_tokens_seen": 99552635,
      "step": 4612,
      "time_per_iteration": 2.7060670852661133
    },
    {
      "auxiliary_loss_clip": 0.01151889,
      "auxiliary_loss_mlp": 0.01024438,
      "balance_loss_clip": 1.04944777,
      "balance_loss_mlp": 1.01642704,
      "epoch": 0.5546804545181266,
      "flos": 20995568985600.0,
      "grad_norm": 1.6687913517338082,
      "language_loss": 0.73212349,
      "learning_rate": 1.7436465475021456e-06,
      "loss": 0.7538867,
      "num_input_tokens_seen": 99572685,
      "step": 4613,
      "time_per_iteration": 4.84477162361145
    },
    {
      "auxiliary_loss_clip": 0.01128245,
      "auxiliary_loss_mlp": 0.01025763,
      "balance_loss_clip": 1.04560685,
      "balance_loss_mlp": 1.0181272,
      "epoch": 0.5548006974087657,
      "flos": 26833638297600.0,
      "grad_norm": 2.3304272670695743,
      "language_loss": 0.71737158,
      "learning_rate": 1.7428740207884111e-06,
      "loss": 0.73891163,
      "num_input_tokens_seen": 99593565,
      "step": 4614,
      "time_per_iteration": 2.786623001098633
    },
    {
      "auxiliary_loss_clip": 0.01127705,
      "auxiliary_loss_mlp": 0.01025183,
      "balance_loss_clip": 1.0467,
      "balance_loss_mlp": 1.01641524,
      "epoch": 0.5549209402994048,
      "flos": 33656414031360.0,
      "grad_norm": 1.8998045727407282,
      "language_loss": 0.61149991,
      "learning_rate": 1.7421015330804833e-06,
      "loss": 0.63302875,
      "num_input_tokens_seen": 99613485,
      "step": 4615,
      "time_per_iteration": 3.6982786655426025
    },
    {
      "auxiliary_loss_clip": 0.0118029,
      "auxiliary_loss_mlp": 0.01026867,
      "balance_loss_clip": 1.05153728,
      "balance_loss_mlp": 1.01865935,
      "epoch": 0.5550411831900439,
      "flos": 23769524609280.0,
      "grad_norm": 2.1113036659079647,
      "language_loss": 0.72403216,
      "learning_rate": 1.7413290844955475e-06,
      "loss": 0.74610376,
      "num_input_tokens_seen": 99633515,
      "step": 4616,
      "time_per_iteration": 2.632690906524658
    },
    {
      "auxiliary_loss_clip": 0.01158771,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 1.05047047,
      "balance_loss_mlp": 1.01608157,
      "epoch": 0.555161426080683,
      "flos": 21651189978240.0,
      "grad_norm": 1.9560967485940315,
      "language_loss": 0.77978146,
      "learning_rate": 1.7405566751507843e-06,
      "loss": 0.8016113,
      "num_input_tokens_seen": 99651560,
      "step": 4617,
      "time_per_iteration": 3.631929874420166
    },
    {
      "auxiliary_loss_clip": 0.01135808,
      "auxiliary_loss_mlp": 0.01022922,
      "balance_loss_clip": 1.04593539,
      "balance_loss_mlp": 1.01574588,
      "epoch": 0.555281668971322,
      "flos": 49563116605440.0,
      "grad_norm": 1.474548622853772,
      "language_loss": 0.67555594,
      "learning_rate": 1.7397843051633668e-06,
      "loss": 0.6971432,
      "num_input_tokens_seen": 99674255,
      "step": 4618,
      "time_per_iteration": 3.020416736602783
    },
    {
      "auxiliary_loss_clip": 0.01161683,
      "auxiliary_loss_mlp": 0.01028247,
      "balance_loss_clip": 1.04923797,
      "balance_loss_mlp": 1.0207727,
      "epoch": 0.5554019118619612,
      "flos": 20741608851840.0,
      "grad_norm": 1.5675068297054797,
      "language_loss": 0.71718544,
      "learning_rate": 1.739011974650464e-06,
      "loss": 0.73908472,
      "num_input_tokens_seen": 99693585,
      "step": 4619,
      "time_per_iteration": 2.661813259124756
    },
    {
      "auxiliary_loss_clip": 0.01126693,
      "auxiliary_loss_mlp": 0.01028744,
      "balance_loss_clip": 1.04613137,
      "balance_loss_mlp": 1.02060831,
      "epoch": 0.5555221547526003,
      "flos": 25483217552640.0,
      "grad_norm": 2.1698073261659627,
      "language_loss": 0.76298428,
      "learning_rate": 1.7382396837292365e-06,
      "loss": 0.78453863,
      "num_input_tokens_seen": 99714045,
      "step": 4620,
      "time_per_iteration": 2.8197319507598877
    },
    {
      "auxiliary_loss_clip": 0.01182287,
      "auxiliary_loss_mlp": 0.01029718,
      "balance_loss_clip": 1.05214858,
      "balance_loss_mlp": 1.02110171,
      "epoch": 0.5556423976432393,
      "flos": 21762513204480.0,
      "grad_norm": 1.7394237933946188,
      "language_loss": 0.73829216,
      "learning_rate": 1.737467432516841e-06,
      "loss": 0.76041222,
      "num_input_tokens_seen": 99734145,
      "step": 4621,
      "time_per_iteration": 2.6147940158843994
    },
    {
      "auxiliary_loss_clip": 0.01150922,
      "auxiliary_loss_mlp": 0.01026253,
      "balance_loss_clip": 1.0450927,
      "balance_loss_mlp": 1.01811147,
      "epoch": 0.5557626405338785,
      "flos": 24900171989760.0,
      "grad_norm": 2.356186458649112,
      "language_loss": 0.74292362,
      "learning_rate": 1.7366952211304274e-06,
      "loss": 0.76469535,
      "num_input_tokens_seen": 99751990,
      "step": 4622,
      "time_per_iteration": 2.7667336463928223
    },
    {
      "auxiliary_loss_clip": 0.01141705,
      "auxiliary_loss_mlp": 0.01030081,
      "balance_loss_clip": 1.04553115,
      "balance_loss_mlp": 1.02224326,
      "epoch": 0.5558828834245175,
      "flos": 18697501676160.0,
      "grad_norm": 2.446681093163606,
      "language_loss": 0.83211637,
      "learning_rate": 1.735923049687139e-06,
      "loss": 0.85383415,
      "num_input_tokens_seen": 99768565,
      "step": 4623,
      "time_per_iteration": 2.6100871562957764
    },
    {
      "auxiliary_loss_clip": 0.01144066,
      "auxiliary_loss_mlp": 0.01024467,
      "balance_loss_clip": 1.04426479,
      "balance_loss_mlp": 1.01682615,
      "epoch": 0.5560031263151566,
      "flos": 27272179445760.0,
      "grad_norm": 1.5453904138823986,
      "language_loss": 0.73939252,
      "learning_rate": 1.7351509183041144e-06,
      "loss": 0.76107788,
      "num_input_tokens_seen": 99788895,
      "step": 4624,
      "time_per_iteration": 2.7044906616210938
    },
    {
      "auxiliary_loss_clip": 0.0118221,
      "auxiliary_loss_mlp": 0.01028218,
      "balance_loss_clip": 1.05257487,
      "balance_loss_mlp": 1.0209074,
      "epoch": 0.5561233692057957,
      "flos": 23403738458880.0,
      "grad_norm": 1.9658164247775558,
      "language_loss": 0.71557283,
      "learning_rate": 1.7343788270984852e-06,
      "loss": 0.73767716,
      "num_input_tokens_seen": 99808035,
      "step": 4625,
      "time_per_iteration": 2.595393657684326
    },
    {
      "auxiliary_loss_clip": 0.01153543,
      "auxiliary_loss_mlp": 0.01033603,
      "balance_loss_clip": 1.05037212,
      "balance_loss_mlp": 1.02506757,
      "epoch": 0.5562436120964348,
      "flos": 37670867804160.0,
      "grad_norm": 1.8252894454082575,
      "language_loss": 0.74412382,
      "learning_rate": 1.7336067761873764e-06,
      "loss": 0.76599526,
      "num_input_tokens_seen": 99830460,
      "step": 4626,
      "time_per_iteration": 2.769731283187866
    },
    {
      "auxiliary_loss_clip": 0.01171512,
      "auxiliary_loss_mlp": 0.01025581,
      "balance_loss_clip": 1.04885924,
      "balance_loss_mlp": 1.01736772,
      "epoch": 0.5563638549870739,
      "flos": 25155245445120.0,
      "grad_norm": 1.9031470493220044,
      "language_loss": 0.76179671,
      "learning_rate": 1.7328347656879076e-06,
      "loss": 0.7837677,
      "num_input_tokens_seen": 99850320,
      "step": 4627,
      "time_per_iteration": 2.6360702514648438
    },
    {
      "auxiliary_loss_clip": 0.01137538,
      "auxiliary_loss_mlp": 0.01033025,
      "balance_loss_clip": 1.04624057,
      "balance_loss_mlp": 1.02444768,
      "epoch": 0.556484097877713,
      "flos": 13581810783360.0,
      "grad_norm": 2.1717535785842914,
      "language_loss": 0.6874575,
      "learning_rate": 1.7320627957171927e-06,
      "loss": 0.70916319,
      "num_input_tokens_seen": 99864980,
      "step": 4628,
      "time_per_iteration": 2.650557041168213
    },
    {
      "auxiliary_loss_clip": 0.01180616,
      "auxiliary_loss_mlp": 0.01028965,
      "balance_loss_clip": 1.05287325,
      "balance_loss_mlp": 1.02083468,
      "epoch": 0.5566043407683521,
      "flos": 24681368292480.0,
      "grad_norm": 1.7929145190283247,
      "language_loss": 0.81170511,
      "learning_rate": 1.7312908663923382e-06,
      "loss": 0.83380091,
      "num_input_tokens_seen": 99881155,
      "step": 4629,
      "time_per_iteration": 2.6393423080444336
    },
    {
      "auxiliary_loss_clip": 0.01157388,
      "auxiliary_loss_mlp": 0.01024558,
      "balance_loss_clip": 1.04563093,
      "balance_loss_mlp": 1.0164932,
      "epoch": 0.5567245836589911,
      "flos": 20588161950720.0,
      "grad_norm": 2.3410795210458817,
      "language_loss": 0.67256689,
      "learning_rate": 1.7305189778304463e-06,
      "loss": 0.69438636,
      "num_input_tokens_seen": 99899330,
      "step": 4630,
      "time_per_iteration": 2.6303060054779053
    },
    {
      "auxiliary_loss_clip": 0.01152535,
      "auxiliary_loss_mlp": 0.01027456,
      "balance_loss_clip": 1.05128503,
      "balance_loss_mlp": 1.01942778,
      "epoch": 0.5568448265496303,
      "flos": 20704189858560.0,
      "grad_norm": 1.9390946825667246,
      "language_loss": 0.80391484,
      "learning_rate": 1.729747130148611e-06,
      "loss": 0.82571471,
      "num_input_tokens_seen": 99918525,
      "step": 4631,
      "time_per_iteration": 2.662760019302368
    },
    {
      "auxiliary_loss_clip": 0.0114447,
      "auxiliary_loss_mlp": 0.01033467,
      "balance_loss_clip": 1.0472877,
      "balance_loss_mlp": 1.0247829,
      "epoch": 0.5569650694402694,
      "flos": 25302910256640.0,
      "grad_norm": 2.0913873943411043,
      "language_loss": 0.76774192,
      "learning_rate": 1.7289753234639208e-06,
      "loss": 0.78952128,
      "num_input_tokens_seen": 99937500,
      "step": 4632,
      "time_per_iteration": 2.727421760559082
    },
    {
      "auxiliary_loss_clip": 0.01169804,
      "auxiliary_loss_mlp": 0.01036254,
      "balance_loss_clip": 1.0500617,
      "balance_loss_mlp": 1.02796292,
      "epoch": 0.5570853123309084,
      "flos": 19712623939200.0,
      "grad_norm": 1.9265301119105505,
      "language_loss": 0.76819205,
      "learning_rate": 1.7282035578934592e-06,
      "loss": 0.79025257,
      "num_input_tokens_seen": 99955665,
      "step": 4633,
      "time_per_iteration": 2.646646738052368
    },
    {
      "auxiliary_loss_clip": 0.01147354,
      "auxiliary_loss_mlp": 0.01028264,
      "balance_loss_clip": 1.05185533,
      "balance_loss_mlp": 1.02021766,
      "epoch": 0.5572055552215476,
      "flos": 16108091153280.0,
      "grad_norm": 2.120107796780374,
      "language_loss": 0.78819859,
      "learning_rate": 1.727431833554301e-06,
      "loss": 0.80995476,
      "num_input_tokens_seen": 99974140,
      "step": 4634,
      "time_per_iteration": 2.6311233043670654
    },
    {
      "auxiliary_loss_clip": 0.01115032,
      "auxiliary_loss_mlp": 0.01031136,
      "balance_loss_clip": 1.04166961,
      "balance_loss_mlp": 1.02357793,
      "epoch": 0.5573257981121866,
      "flos": 17128815937920.0,
      "grad_norm": 2.0318198295269982,
      "language_loss": 0.7733174,
      "learning_rate": 1.7266601505635175e-06,
      "loss": 0.79477906,
      "num_input_tokens_seen": 99991480,
      "step": 4635,
      "time_per_iteration": 3.0732650756835938
    },
    {
      "auxiliary_loss_clip": 0.01168505,
      "auxiliary_loss_mlp": 0.01032305,
      "balance_loss_clip": 1.05225945,
      "balance_loss_mlp": 1.02434754,
      "epoch": 0.5574460410028257,
      "flos": 18807029222400.0,
      "grad_norm": 2.051734748187882,
      "language_loss": 0.75603414,
      "learning_rate": 1.7258885090381717e-06,
      "loss": 0.77804226,
      "num_input_tokens_seen": 100009520,
      "step": 4636,
      "time_per_iteration": 2.8333001136779785
    },
    {
      "auxiliary_loss_clip": 0.01153956,
      "auxiliary_loss_mlp": 0.01027695,
      "balance_loss_clip": 1.04671133,
      "balance_loss_mlp": 1.02070284,
      "epoch": 0.5575662838934649,
      "flos": 29642678530560.0,
      "grad_norm": 2.1455819847401365,
      "language_loss": 0.78887147,
      "learning_rate": 1.7251169090953213e-06,
      "loss": 0.81068802,
      "num_input_tokens_seen": 100029995,
      "step": 4637,
      "time_per_iteration": 2.7316410541534424
    },
    {
      "auxiliary_loss_clip": 0.01165919,
      "auxiliary_loss_mlp": 0.01031207,
      "balance_loss_clip": 1.04917121,
      "balance_loss_mlp": 1.02302969,
      "epoch": 0.5576865267841039,
      "flos": 22054466949120.0,
      "grad_norm": 2.5213662132560435,
      "language_loss": 0.76683038,
      "learning_rate": 1.7243453508520168e-06,
      "loss": 0.78880167,
      "num_input_tokens_seen": 100046980,
      "step": 4638,
      "time_per_iteration": 3.64043927192688
    },
    {
      "auxiliary_loss_clip": 0.01149974,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.04423034,
      "balance_loss_mlp": 1.01802146,
      "epoch": 0.557806769674743,
      "flos": 17196040241280.0,
      "grad_norm": 2.100843130555985,
      "language_loss": 0.8469733,
      "learning_rate": 1.7235738344253038e-06,
      "loss": 0.868729,
      "num_input_tokens_seen": 100060610,
      "step": 4639,
      "time_per_iteration": 2.651303768157959
    },
    {
      "auxiliary_loss_clip": 0.01164395,
      "auxiliary_loss_mlp": 0.01028526,
      "balance_loss_clip": 1.05033505,
      "balance_loss_mlp": 1.02056289,
      "epoch": 0.557927012565382,
      "flos": 24712717887360.0,
      "grad_norm": 2.3957266494216665,
      "language_loss": 0.82811922,
      "learning_rate": 1.72280235993222e-06,
      "loss": 0.85004842,
      "num_input_tokens_seen": 100078915,
      "step": 4640,
      "time_per_iteration": 3.607492446899414
    },
    {
      "auxiliary_loss_clip": 0.01163735,
      "auxiliary_loss_mlp": 0.00763398,
      "balance_loss_clip": 1.04929519,
      "balance_loss_mlp": 1.00032496,
      "epoch": 0.5580472554560212,
      "flos": 16983090460800.0,
      "grad_norm": 2.354066777053142,
      "language_loss": 0.69865334,
      "learning_rate": 1.722030927489798e-06,
      "loss": 0.71792471,
      "num_input_tokens_seen": 100096195,
      "step": 4641,
      "time_per_iteration": 3.5771536827087402
    },
    {
      "auxiliary_loss_clip": 0.01140014,
      "auxiliary_loss_mlp": 0.01027679,
      "balance_loss_clip": 1.04910231,
      "balance_loss_mlp": 1.0193758,
      "epoch": 0.5581674983466602,
      "flos": 23509100027520.0,
      "grad_norm": 1.958369524972769,
      "language_loss": 0.74102819,
      "learning_rate": 1.7212595372150634e-06,
      "loss": 0.76270509,
      "num_input_tokens_seen": 100116175,
      "step": 4642,
      "time_per_iteration": 2.793229579925537
    },
    {
      "auxiliary_loss_clip": 0.01181204,
      "auxiliary_loss_mlp": 0.01028764,
      "balance_loss_clip": 1.05236375,
      "balance_loss_mlp": 1.02100384,
      "epoch": 0.5582877412372993,
      "flos": 13480291969920.0,
      "grad_norm": 2.458259518248705,
      "language_loss": 0.7277,
      "learning_rate": 1.720488189225035e-06,
      "loss": 0.74979973,
      "num_input_tokens_seen": 100133875,
      "step": 4643,
      "time_per_iteration": 3.5307369232177734
    },
    {
      "auxiliary_loss_clip": 0.01168086,
      "auxiliary_loss_mlp": 0.01028608,
      "balance_loss_clip": 1.04888749,
      "balance_loss_mlp": 1.02081728,
      "epoch": 0.5584079841279385,
      "flos": 21903605827200.0,
      "grad_norm": 2.537920407523779,
      "language_loss": 0.79306775,
      "learning_rate": 1.7197168836367265e-06,
      "loss": 0.81503475,
      "num_input_tokens_seen": 100150685,
      "step": 4644,
      "time_per_iteration": 2.6668155193328857
    },
    {
      "auxiliary_loss_clip": 0.01162818,
      "auxiliary_loss_mlp": 0.00762955,
      "balance_loss_clip": 1.04837108,
      "balance_loss_mlp": 1.00024545,
      "epoch": 0.5585282270185775,
      "flos": 18843550375680.0,
      "grad_norm": 1.87766278363126,
      "language_loss": 0.82353479,
      "learning_rate": 1.7189456205671433e-06,
      "loss": 0.84279251,
      "num_input_tokens_seen": 100169530,
      "step": 4645,
      "time_per_iteration": 2.684387445449829
    },
    {
      "auxiliary_loss_clip": 0.01174311,
      "auxiliary_loss_mlp": 0.01027101,
      "balance_loss_clip": 1.05107927,
      "balance_loss_mlp": 1.01897693,
      "epoch": 0.5586484699092166,
      "flos": 21868449390720.0,
      "grad_norm": 2.896251298068004,
      "language_loss": 0.82434189,
      "learning_rate": 1.7181744001332866e-06,
      "loss": 0.84635592,
      "num_input_tokens_seen": 100188140,
      "step": 4646,
      "time_per_iteration": 2.802567481994629
    },
    {
      "auxiliary_loss_clip": 0.01179839,
      "auxiliary_loss_mlp": 0.0102776,
      "balance_loss_clip": 1.05185843,
      "balance_loss_mlp": 1.02044022,
      "epoch": 0.5587687127998557,
      "flos": 22893232412160.0,
      "grad_norm": 2.4784121648226165,
      "language_loss": 0.6319648,
      "learning_rate": 1.7174032224521493e-06,
      "loss": 0.65404081,
      "num_input_tokens_seen": 100206850,
      "step": 4647,
      "time_per_iteration": 2.6848559379577637
    },
    {
      "auxiliary_loss_clip": 0.01164569,
      "auxiliary_loss_mlp": 0.01024755,
      "balance_loss_clip": 1.04950297,
      "balance_loss_mlp": 1.01725984,
      "epoch": 0.5588889556904948,
      "flos": 20303067703680.0,
      "grad_norm": 1.6733308947972305,
      "language_loss": 0.69900656,
      "learning_rate": 1.7166320876407184e-06,
      "loss": 0.7208997,
      "num_input_tokens_seen": 100226270,
      "step": 4648,
      "time_per_iteration": 2.700667142868042
    },
    {
      "auxiliary_loss_clip": 0.01184823,
      "auxiliary_loss_mlp": 0.00763826,
      "balance_loss_clip": 1.05379474,
      "balance_loss_mlp": 1.00031018,
      "epoch": 0.5590091985811338,
      "flos": 16472153450880.0,
      "grad_norm": 1.8617661272089383,
      "language_loss": 0.67926246,
      "learning_rate": 1.7158609958159742e-06,
      "loss": 0.69874895,
      "num_input_tokens_seen": 100243675,
      "step": 4649,
      "time_per_iteration": 2.7556564807891846
    },
    {
      "auxiliary_loss_clip": 0.01118771,
      "auxiliary_loss_mlp": 0.01029824,
      "balance_loss_clip": 1.04342616,
      "balance_loss_mlp": 1.02144349,
      "epoch": 0.559129441471773,
      "flos": 14532186781440.0,
      "grad_norm": 2.2839195456866643,
      "language_loss": 0.78176105,
      "learning_rate": 1.7150899470948911e-06,
      "loss": 0.80324709,
      "num_input_tokens_seen": 100258940,
      "step": 4650,
      "time_per_iteration": 2.859292984008789
    },
    {
      "auxiliary_loss_clip": 0.0104734,
      "auxiliary_loss_mlp": 0.0100232,
      "balance_loss_clip": 1.00974369,
      "balance_loss_mlp": 1.00088978,
      "epoch": 0.5592496843624121,
      "flos": 60521009852160.0,
      "grad_norm": 0.8379854314193489,
      "language_loss": 0.56629443,
      "learning_rate": 1.7143189415944365e-06,
      "loss": 0.58679098,
      "num_input_tokens_seen": 100323400,
      "step": 4651,
      "time_per_iteration": 3.2557644844055176
    },
    {
      "auxiliary_loss_clip": 0.01161611,
      "auxiliary_loss_mlp": 0.01027006,
      "balance_loss_clip": 1.04864097,
      "balance_loss_mlp": 1.01972842,
      "epoch": 0.5593699272530511,
      "flos": 20886256920960.0,
      "grad_norm": 1.6281031580553942,
      "language_loss": 0.76374686,
      "learning_rate": 1.7135479794315714e-06,
      "loss": 0.78563309,
      "num_input_tokens_seen": 100340355,
      "step": 4652,
      "time_per_iteration": 2.701763868331909
    },
    {
      "auxiliary_loss_clip": 0.01134718,
      "auxiliary_loss_mlp": 0.01027109,
      "balance_loss_clip": 1.04591453,
      "balance_loss_mlp": 1.01942039,
      "epoch": 0.5594901701436903,
      "flos": 12896743616640.0,
      "grad_norm": 1.9499210337336101,
      "language_loss": 0.78937721,
      "learning_rate": 1.7127770607232502e-06,
      "loss": 0.81099546,
      "num_input_tokens_seen": 100358900,
      "step": 4653,
      "time_per_iteration": 2.674389362335205
    },
    {
      "auxiliary_loss_clip": 0.01144589,
      "auxiliary_loss_mlp": 0.01028376,
      "balance_loss_clip": 1.04791462,
      "balance_loss_mlp": 1.02079463,
      "epoch": 0.5596104130343293,
      "flos": 23112107936640.0,
      "grad_norm": 2.608515395794512,
      "language_loss": 0.8010847,
      "learning_rate": 1.7120061855864204e-06,
      "loss": 0.82281435,
      "num_input_tokens_seen": 100378910,
      "step": 4654,
      "time_per_iteration": 2.738274335861206
    },
    {
      "auxiliary_loss_clip": 0.01166806,
      "auxiliary_loss_mlp": 0.01030348,
      "balance_loss_clip": 1.05208302,
      "balance_loss_mlp": 1.02224183,
      "epoch": 0.5597306559249684,
      "flos": 25957812977280.0,
      "grad_norm": 1.9333335772754632,
      "language_loss": 0.70735943,
      "learning_rate": 1.7112353541380233e-06,
      "loss": 0.72933096,
      "num_input_tokens_seen": 100398770,
      "step": 4655,
      "time_per_iteration": 2.681320905685425
    },
    {
      "auxiliary_loss_clip": 0.01152852,
      "auxiliary_loss_mlp": 0.0103262,
      "balance_loss_clip": 1.04886746,
      "balance_loss_mlp": 1.02479959,
      "epoch": 0.5598508988156076,
      "flos": 22492289825280.0,
      "grad_norm": 1.474541127337974,
      "language_loss": 0.72149652,
      "learning_rate": 1.7104645664949931e-06,
      "loss": 0.74335122,
      "num_input_tokens_seen": 100421240,
      "step": 4656,
      "time_per_iteration": 2.6743571758270264
    },
    {
      "auxiliary_loss_clip": 0.01151169,
      "auxiliary_loss_mlp": 0.01024358,
      "balance_loss_clip": 1.0460943,
      "balance_loss_mlp": 1.01644301,
      "epoch": 0.5599711417062466,
      "flos": 23112538899840.0,
      "grad_norm": 1.758078232720619,
      "language_loss": 0.71497601,
      "learning_rate": 1.7096938227742584e-06,
      "loss": 0.73673129,
      "num_input_tokens_seen": 100442370,
      "step": 4657,
      "time_per_iteration": 2.704732894897461
    },
    {
      "auxiliary_loss_clip": 0.01182094,
      "auxiliary_loss_mlp": 0.01027535,
      "balance_loss_clip": 1.05188084,
      "balance_loss_mlp": 1.01903546,
      "epoch": 0.5600913845968857,
      "flos": 22339345714560.0,
      "grad_norm": 2.066011457892653,
      "language_loss": 0.84145939,
      "learning_rate": 1.70892312309274e-06,
      "loss": 0.86355567,
      "num_input_tokens_seen": 100460260,
      "step": 4658,
      "time_per_iteration": 2.5490872859954834
    },
    {
      "auxiliary_loss_clip": 0.0114917,
      "auxiliary_loss_mlp": 0.01030011,
      "balance_loss_clip": 1.04159856,
      "balance_loss_mlp": 1.02166033,
      "epoch": 0.5602116274875248,
      "flos": 17633791290240.0,
      "grad_norm": 2.2116496275065014,
      "language_loss": 0.68311489,
      "learning_rate": 1.7081524675673523e-06,
      "loss": 0.7049067,
      "num_input_tokens_seen": 100475750,
      "step": 4659,
      "time_per_iteration": 2.6439735889434814
    },
    {
      "auxiliary_loss_clip": 0.01050741,
      "auxiliary_loss_mlp": 0.01000847,
      "balance_loss_clip": 1.01016057,
      "balance_loss_mlp": 0.99941635,
      "epoch": 0.5603318703781639,
      "flos": 70115945529600.0,
      "grad_norm": 0.7782727895000924,
      "language_loss": 0.59588826,
      "learning_rate": 1.7073818563150026e-06,
      "loss": 0.61640406,
      "num_input_tokens_seen": 100537830,
      "step": 4660,
      "time_per_iteration": 3.3295950889587402
    },
    {
      "auxiliary_loss_clip": 0.01160054,
      "auxiliary_loss_mlp": 0.01032824,
      "balance_loss_clip": 1.04732049,
      "balance_loss_mlp": 1.02464962,
      "epoch": 0.560452113268803,
      "flos": 18545850455040.0,
      "grad_norm": 5.268178256964221,
      "language_loss": 0.86426491,
      "learning_rate": 1.7066112894525935e-06,
      "loss": 0.88619369,
      "num_input_tokens_seen": 100555910,
      "step": 4661,
      "time_per_iteration": 2.668867349624634
    },
    {
      "auxiliary_loss_clip": 0.01145843,
      "auxiliary_loss_mlp": 0.0103271,
      "balance_loss_clip": 1.04727221,
      "balance_loss_mlp": 1.02475834,
      "epoch": 0.5605723561594421,
      "flos": 25264665250560.0,
      "grad_norm": 1.7741031654416468,
      "language_loss": 0.72954839,
      "learning_rate": 1.7058407670970177e-06,
      "loss": 0.75133389,
      "num_input_tokens_seen": 100577385,
      "step": 4662,
      "time_per_iteration": 2.6764779090881348
    },
    {
      "auxiliary_loss_clip": 0.01173894,
      "auxiliary_loss_mlp": 0.01032304,
      "balance_loss_clip": 1.05087161,
      "balance_loss_mlp": 1.02390027,
      "epoch": 0.5606925990500812,
      "flos": 20594949621120.0,
      "grad_norm": 1.6488969702076917,
      "language_loss": 0.61241871,
      "learning_rate": 1.7050702893651643e-06,
      "loss": 0.63448071,
      "num_input_tokens_seen": 100596965,
      "step": 4663,
      "time_per_iteration": 2.6287930011749268
    },
    {
      "auxiliary_loss_clip": 0.0116943,
      "auxiliary_loss_mlp": 0.01033919,
      "balance_loss_clip": 1.05168545,
      "balance_loss_mlp": 1.02560973,
      "epoch": 0.5608128419407202,
      "flos": 35006044677120.0,
      "grad_norm": 2.2950334938136243,
      "language_loss": 0.75236171,
      "learning_rate": 1.7042998563739134e-06,
      "loss": 0.77439523,
      "num_input_tokens_seen": 100615315,
      "step": 4664,
      "time_per_iteration": 3.6370463371276855
    },
    {
      "auxiliary_loss_clip": 0.01160983,
      "auxiliary_loss_mlp": 0.01028869,
      "balance_loss_clip": 1.04973924,
      "balance_loss_mlp": 1.01970148,
      "epoch": 0.5609330848313594,
      "flos": 24639819235200.0,
      "grad_norm": 2.4689023634123486,
      "language_loss": 0.7220825,
      "learning_rate": 1.703529468240139e-06,
      "loss": 0.743981,
      "num_input_tokens_seen": 100634185,
      "step": 4665,
      "time_per_iteration": 2.5581390857696533
    },
    {
      "auxiliary_loss_clip": 0.0114681,
      "auxiliary_loss_mlp": 0.01021742,
      "balance_loss_clip": 1.04829717,
      "balance_loss_mlp": 1.01400208,
      "epoch": 0.5610533277219985,
      "flos": 18762894385920.0,
      "grad_norm": 2.2164941081385288,
      "language_loss": 0.73717344,
      "learning_rate": 1.7027591250807088e-06,
      "loss": 0.75885892,
      "num_input_tokens_seen": 100651360,
      "step": 4666,
      "time_per_iteration": 3.386622667312622
    },
    {
      "auxiliary_loss_clip": 0.01183849,
      "auxiliary_loss_mlp": 0.01025368,
      "balance_loss_clip": 1.05269206,
      "balance_loss_mlp": 1.01789069,
      "epoch": 0.5611735706126375,
      "flos": 15012384727680.0,
      "grad_norm": 2.437501892487151,
      "language_loss": 0.84596705,
      "learning_rate": 1.7019888270124825e-06,
      "loss": 0.86805928,
      "num_input_tokens_seen": 100668525,
      "step": 4667,
      "time_per_iteration": 3.397329568862915
    },
    {
      "auxiliary_loss_clip": 0.01173289,
      "auxiliary_loss_mlp": 0.01030441,
      "balance_loss_clip": 1.05320394,
      "balance_loss_mlp": 1.02253723,
      "epoch": 0.5612938135032767,
      "flos": 16468167041280.0,
      "grad_norm": 2.036207372689026,
      "language_loss": 0.81986332,
      "learning_rate": 1.7012185741523147e-06,
      "loss": 0.84190059,
      "num_input_tokens_seen": 100684850,
      "step": 4668,
      "time_per_iteration": 2.562749147415161
    },
    {
      "auxiliary_loss_clip": 0.01185799,
      "auxiliary_loss_mlp": 0.01033392,
      "balance_loss_clip": 1.05535316,
      "balance_loss_mlp": 1.02493978,
      "epoch": 0.5614140563939157,
      "flos": 25666433850240.0,
      "grad_norm": 3.9196329947582655,
      "language_loss": 0.62839359,
      "learning_rate": 1.7004483666170514e-06,
      "loss": 0.65058553,
      "num_input_tokens_seen": 100705345,
      "step": 4669,
      "time_per_iteration": 3.5400469303131104
    },
    {
      "auxiliary_loss_clip": 0.01164243,
      "auxiliary_loss_mlp": 0.01033787,
      "balance_loss_clip": 1.0495739,
      "balance_loss_mlp": 1.02543652,
      "epoch": 0.5615342992845548,
      "flos": 24717566223360.0,
      "grad_norm": 2.0679862629560186,
      "language_loss": 0.80247283,
      "learning_rate": 1.699678204523533e-06,
      "loss": 0.82445306,
      "num_input_tokens_seen": 100725210,
      "step": 4670,
      "time_per_iteration": 2.6882379055023193
    },
    {
      "auxiliary_loss_clip": 0.01156417,
      "auxiliary_loss_mlp": 0.01024919,
      "balance_loss_clip": 1.05084026,
      "balance_loss_mlp": 1.01667559,
      "epoch": 0.5616545421751938,
      "flos": 22015934634240.0,
      "grad_norm": 3.2723950052186495,
      "language_loss": 0.68959939,
      "learning_rate": 1.6989080879885918e-06,
      "loss": 0.71141279,
      "num_input_tokens_seen": 100743070,
      "step": 4671,
      "time_per_iteration": 2.662358283996582
    },
    {
      "auxiliary_loss_clip": 0.01043132,
      "auxiliary_loss_mlp": 0.0100246,
      "balance_loss_clip": 1.01328444,
      "balance_loss_mlp": 1.0011071,
      "epoch": 0.561774785065833,
      "flos": 53760358690560.0,
      "grad_norm": 0.891251952382261,
      "language_loss": 0.61067647,
      "learning_rate": 1.6981380171290544e-06,
      "loss": 0.63113236,
      "num_input_tokens_seen": 100804095,
      "step": 4672,
      "time_per_iteration": 3.283496141433716
    },
    {
      "auxiliary_loss_clip": 0.01147254,
      "auxiliary_loss_mlp": 0.01023748,
      "balance_loss_clip": 1.04557002,
      "balance_loss_mlp": 1.01554012,
      "epoch": 0.5618950279564721,
      "flos": 19750007018880.0,
      "grad_norm": 5.604544801833084,
      "language_loss": 0.74217898,
      "learning_rate": 1.6973679920617396e-06,
      "loss": 0.76388901,
      "num_input_tokens_seen": 100821630,
      "step": 4673,
      "time_per_iteration": 2.671304225921631
    },
    {
      "auxiliary_loss_clip": 0.01151647,
      "auxiliary_loss_mlp": 0.01023778,
      "balance_loss_clip": 1.05052567,
      "balance_loss_mlp": 1.01596999,
      "epoch": 0.5620152708471111,
      "flos": 16800592435200.0,
      "grad_norm": 2.8883879246362905,
      "language_loss": 0.84930885,
      "learning_rate": 1.6965980129034603e-06,
      "loss": 0.87106311,
      "num_input_tokens_seen": 100839015,
      "step": 4674,
      "time_per_iteration": 2.6611170768737793
    },
    {
      "auxiliary_loss_clip": 0.01154282,
      "auxiliary_loss_mlp": 0.01021975,
      "balance_loss_clip": 1.05035889,
      "balance_loss_mlp": 1.01380324,
      "epoch": 0.5621355137377503,
      "flos": 26797799502720.0,
      "grad_norm": 1.6686521712911004,
      "language_loss": 0.76802582,
      "learning_rate": 1.6958280797710209e-06,
      "loss": 0.78978837,
      "num_input_tokens_seen": 100860940,
      "step": 4675,
      "time_per_iteration": 2.8043601512908936
    },
    {
      "auxiliary_loss_clip": 0.01051341,
      "auxiliary_loss_mlp": 0.01004174,
      "balance_loss_clip": 1.01206183,
      "balance_loss_mlp": 1.00275505,
      "epoch": 0.5622557566283893,
      "flos": 61207046686080.0,
      "grad_norm": 0.7132453224407957,
      "language_loss": 0.54724866,
      "learning_rate": 1.6950581927812198e-06,
      "loss": 0.5678038,
      "num_input_tokens_seen": 100920510,
      "step": 4676,
      "time_per_iteration": 3.1277053356170654
    },
    {
      "auxiliary_loss_clip": 0.01167739,
      "auxiliary_loss_mlp": 0.01028551,
      "balance_loss_clip": 1.04948354,
      "balance_loss_mlp": 1.02059937,
      "epoch": 0.5623759995190284,
      "flos": 26468534505600.0,
      "grad_norm": 1.9940085196562543,
      "language_loss": 0.79634237,
      "learning_rate": 1.6942883520508486e-06,
      "loss": 0.81830531,
      "num_input_tokens_seen": 100939245,
      "step": 4677,
      "time_per_iteration": 2.6593196392059326
    },
    {
      "auxiliary_loss_clip": 0.01168193,
      "auxiliary_loss_mlp": 0.01027272,
      "balance_loss_clip": 1.05100834,
      "balance_loss_mlp": 1.01968479,
      "epoch": 0.5624962424096676,
      "flos": 19390900798080.0,
      "grad_norm": 1.9835901875616895,
      "language_loss": 0.76859868,
      "learning_rate": 1.693518557696691e-06,
      "loss": 0.79055333,
      "num_input_tokens_seen": 100958385,
      "step": 4678,
      "time_per_iteration": 2.6692769527435303
    },
    {
      "auxiliary_loss_clip": 0.01160561,
      "auxiliary_loss_mlp": 0.01029598,
      "balance_loss_clip": 1.04530823,
      "balance_loss_mlp": 1.02157521,
      "epoch": 0.5626164853003066,
      "flos": 20667345482880.0,
      "grad_norm": 2.1420473693537687,
      "language_loss": 0.88855827,
      "learning_rate": 1.6927488098355252e-06,
      "loss": 0.91045988,
      "num_input_tokens_seen": 100976015,
      "step": 4679,
      "time_per_iteration": 2.7635915279388428
    },
    {
      "auxiliary_loss_clip": 0.01036669,
      "auxiliary_loss_mlp": 0.0100294,
      "balance_loss_clip": 1.01089644,
      "balance_loss_mlp": 1.00146186,
      "epoch": 0.5627367281909457,
      "flos": 62766071665920.0,
      "grad_norm": 0.8960021728672598,
      "language_loss": 0.63151956,
      "learning_rate": 1.6919791085841201e-06,
      "loss": 0.65191555,
      "num_input_tokens_seen": 101033425,
      "step": 4680,
      "time_per_iteration": 3.240887403488159
    },
    {
      "auxiliary_loss_clip": 0.01159483,
      "auxiliary_loss_mlp": 0.01029354,
      "balance_loss_clip": 1.04633522,
      "balance_loss_mlp": 1.02097344,
      "epoch": 0.5628569710815848,
      "flos": 12787144243200.0,
      "grad_norm": 2.4119958018594754,
      "language_loss": 0.79237992,
      "learning_rate": 1.6912094540592396e-06,
      "loss": 0.81426823,
      "num_input_tokens_seen": 101048945,
      "step": 4681,
      "time_per_iteration": 2.721222162246704
    },
    {
      "auxiliary_loss_clip": 0.01163215,
      "auxiliary_loss_mlp": 0.01028208,
      "balance_loss_clip": 1.0475812,
      "balance_loss_mlp": 1.02011371,
      "epoch": 0.5629772139722239,
      "flos": 13762082165760.0,
      "grad_norm": 2.692766796325648,
      "language_loss": 0.81304324,
      "learning_rate": 1.6904398463776393e-06,
      "loss": 0.83495742,
      "num_input_tokens_seen": 101062745,
      "step": 4682,
      "time_per_iteration": 2.585064649581909
    },
    {
      "auxiliary_loss_clip": 0.01165716,
      "auxiliary_loss_mlp": 0.01027238,
      "balance_loss_clip": 1.04733729,
      "balance_loss_mlp": 1.01951039,
      "epoch": 0.5630974568628629,
      "flos": 21467830026240.0,
      "grad_norm": 1.986637955573854,
      "language_loss": 0.72595185,
      "learning_rate": 1.6896702856560683e-06,
      "loss": 0.74788141,
      "num_input_tokens_seen": 101081840,
      "step": 4683,
      "time_per_iteration": 2.6382105350494385
    },
    {
      "auxiliary_loss_clip": 0.01132419,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 1.04207706,
      "balance_loss_mlp": 1.01798153,
      "epoch": 0.5632176997535021,
      "flos": 14245907385600.0,
      "grad_norm": 2.969966718035667,
      "language_loss": 0.69508636,
      "learning_rate": 1.6889007720112677e-06,
      "loss": 0.71667409,
      "num_input_tokens_seen": 101099585,
      "step": 4684,
      "time_per_iteration": 2.649993419647217
    },
    {
      "auxiliary_loss_clip": 0.01168635,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.05064785,
      "balance_loss_mlp": 1.02104497,
      "epoch": 0.5633379426441412,
      "flos": 20812244947200.0,
      "grad_norm": 1.6983158852140172,
      "language_loss": 0.7726289,
      "learning_rate": 1.6881313055599734e-06,
      "loss": 0.79460239,
      "num_input_tokens_seen": 101119515,
      "step": 4685,
      "time_per_iteration": 2.6453113555908203
    },
    {
      "auxiliary_loss_clip": 0.01136038,
      "auxiliary_loss_mlp": 0.01028974,
      "balance_loss_clip": 1.04270315,
      "balance_loss_mlp": 1.02059317,
      "epoch": 0.5634581855347802,
      "flos": 22600883617920.0,
      "grad_norm": 2.6256302771073945,
      "language_loss": 0.82396293,
      "learning_rate": 1.6873618864189117e-06,
      "loss": 0.845613,
      "num_input_tokens_seen": 101135285,
      "step": 4686,
      "time_per_iteration": 2.67458438873291
    },
    {
      "auxiliary_loss_clip": 0.01165091,
      "auxiliary_loss_mlp": 0.01028278,
      "balance_loss_clip": 1.04937327,
      "balance_loss_mlp": 1.02001119,
      "epoch": 0.5635784284254194,
      "flos": 21506972872320.0,
      "grad_norm": 2.1970487918841926,
      "language_loss": 0.78234249,
      "learning_rate": 1.686592514704803e-06,
      "loss": 0.80427617,
      "num_input_tokens_seen": 101152680,
      "step": 4687,
      "time_per_iteration": 2.6899025440216064
    },
    {
      "auxiliary_loss_clip": 0.01153591,
      "auxiliary_loss_mlp": 0.01024617,
      "balance_loss_clip": 1.05143762,
      "balance_loss_mlp": 1.01756549,
      "epoch": 0.5636986713160584,
      "flos": 19827466698240.0,
      "grad_norm": 2.528030433191082,
      "language_loss": 0.71102542,
      "learning_rate": 1.685823190534361e-06,
      "loss": 0.73280746,
      "num_input_tokens_seen": 101170920,
      "step": 4688,
      "time_per_iteration": 2.645359516143799
    },
    {
      "auxiliary_loss_clip": 0.01184426,
      "auxiliary_loss_mlp": 0.01030629,
      "balance_loss_clip": 1.05192685,
      "balance_loss_mlp": 1.02236223,
      "epoch": 0.5638189142066975,
      "flos": 19792453916160.0,
      "grad_norm": 2.2168338892957564,
      "language_loss": 0.83887362,
      "learning_rate": 1.6850539140242907e-06,
      "loss": 0.8610242,
      "num_input_tokens_seen": 101190180,
      "step": 4689,
      "time_per_iteration": 2.644681692123413
    },
    {
      "auxiliary_loss_clip": 0.01168624,
      "auxiliary_loss_mlp": 0.01027291,
      "balance_loss_clip": 1.04818666,
      "balance_loss_mlp": 1.01949453,
      "epoch": 0.5639391570973367,
      "flos": 22893771116160.0,
      "grad_norm": 2.1056113559202476,
      "language_loss": 0.82170987,
      "learning_rate": 1.684284685291292e-06,
      "loss": 0.84366906,
      "num_input_tokens_seen": 101211825,
      "step": 4690,
      "time_per_iteration": 2.6609227657318115
    },
    {
      "auxiliary_loss_clip": 0.01179598,
      "auxiliary_loss_mlp": 0.01031688,
      "balance_loss_clip": 1.05088603,
      "balance_loss_mlp": 1.02232432,
      "epoch": 0.5640593999879757,
      "flos": 23727077712000.0,
      "grad_norm": 4.096138194442154,
      "language_loss": 0.81575257,
      "learning_rate": 1.683515504452055e-06,
      "loss": 0.83786547,
      "num_input_tokens_seen": 101229200,
      "step": 4691,
      "time_per_iteration": 3.6334943771362305
    },
    {
      "auxiliary_loss_clip": 0.01128929,
      "auxiliary_loss_mlp": 0.0103062,
      "balance_loss_clip": 1.04468203,
      "balance_loss_mlp": 1.02231741,
      "epoch": 0.5641796428786148,
      "flos": 22710123855360.0,
      "grad_norm": 1.6069116922170086,
      "language_loss": 0.66375524,
      "learning_rate": 1.6827463716232648e-06,
      "loss": 0.68535072,
      "num_input_tokens_seen": 101249860,
      "step": 4692,
      "time_per_iteration": 3.822901725769043
    },
    {
      "auxiliary_loss_clip": 0.01166748,
      "auxiliary_loss_mlp": 0.00762802,
      "balance_loss_clip": 1.04976726,
      "balance_loss_mlp": 1.0003252,
      "epoch": 0.5642998857692539,
      "flos": 19791987039360.0,
      "grad_norm": 1.9293441985654194,
      "language_loss": 0.75805914,
      "learning_rate": 1.6819772869215972e-06,
      "loss": 0.7773546,
      "num_input_tokens_seen": 101268940,
      "step": 4693,
      "time_per_iteration": 3.490962266921997
    },
    {
      "auxiliary_loss_clip": 0.01158255,
      "auxiliary_loss_mlp": 0.01028615,
      "balance_loss_clip": 1.04995346,
      "balance_loss_mlp": 1.02074718,
      "epoch": 0.564420128659893,
      "flos": 23185904428800.0,
      "grad_norm": 1.7077748959889714,
      "language_loss": 0.81998897,
      "learning_rate": 1.6812082504637228e-06,
      "loss": 0.84185767,
      "num_input_tokens_seen": 101290260,
      "step": 4694,
      "time_per_iteration": 2.6713359355926514
    },
    {
      "auxiliary_loss_clip": 0.01164338,
      "auxiliary_loss_mlp": 0.01029894,
      "balance_loss_clip": 1.05193126,
      "balance_loss_mlp": 1.02176952,
      "epoch": 0.564540371550532,
      "flos": 23258264376960.0,
      "grad_norm": 1.5772503225008376,
      "language_loss": 0.74576783,
      "learning_rate": 1.6804392623663025e-06,
      "loss": 0.76771009,
      "num_input_tokens_seen": 101311465,
      "step": 4695,
      "time_per_iteration": 3.593247413635254
    },
    {
      "auxiliary_loss_clip": 0.0116098,
      "auxiliary_loss_mlp": 0.01026545,
      "balance_loss_clip": 1.04912269,
      "balance_loss_mlp": 1.01896024,
      "epoch": 0.5646606144411712,
      "flos": 25010058672000.0,
      "grad_norm": 1.965626749814175,
      "language_loss": 0.78352535,
      "learning_rate": 1.6796703227459935e-06,
      "loss": 0.80540067,
      "num_input_tokens_seen": 101329420,
      "step": 4696,
      "time_per_iteration": 2.690711498260498
    },
    {
      "auxiliary_loss_clip": 0.01111703,
      "auxiliary_loss_mlp": 0.01027193,
      "balance_loss_clip": 1.04140484,
      "balance_loss_mlp": 1.02009428,
      "epoch": 0.5647808573318103,
      "flos": 36539645806080.0,
      "grad_norm": 2.0474080920370437,
      "language_loss": 0.76008427,
      "learning_rate": 1.6789014317194407e-06,
      "loss": 0.78147316,
      "num_input_tokens_seen": 101350900,
      "step": 4697,
      "time_per_iteration": 2.8789632320404053
    },
    {
      "auxiliary_loss_clip": 0.01158149,
      "auxiliary_loss_mlp": 0.01030813,
      "balance_loss_clip": 1.04880834,
      "balance_loss_mlp": 1.02274847,
      "epoch": 0.5649011002224493,
      "flos": 22528451842560.0,
      "grad_norm": 2.4050864590594156,
      "language_loss": 0.7240926,
      "learning_rate": 1.6781325894032853e-06,
      "loss": 0.74598217,
      "num_input_tokens_seen": 101369860,
      "step": 4698,
      "time_per_iteration": 2.6996679306030273
    },
    {
      "auxiliary_loss_clip": 0.01147108,
      "auxiliary_loss_mlp": 0.01029682,
      "balance_loss_clip": 1.04969847,
      "balance_loss_mlp": 1.02244353,
      "epoch": 0.5650213431130885,
      "flos": 18515147304960.0,
      "grad_norm": 1.8970440021676518,
      "language_loss": 0.91751039,
      "learning_rate": 1.6773637959141608e-06,
      "loss": 0.9392783,
      "num_input_tokens_seen": 101386835,
      "step": 4699,
      "time_per_iteration": 2.67732572555542
    },
    {
      "auxiliary_loss_clip": 0.01141069,
      "auxiliary_loss_mlp": 0.01024321,
      "balance_loss_clip": 1.04484487,
      "balance_loss_mlp": 1.01675081,
      "epoch": 0.5651415860037275,
      "flos": 17526310819200.0,
      "grad_norm": 2.002864459168587,
      "language_loss": 0.66388977,
      "learning_rate": 1.6765950513686915e-06,
      "loss": 0.68554366,
      "num_input_tokens_seen": 101404945,
      "step": 4700,
      "time_per_iteration": 2.657261848449707
    },
    {
      "auxiliary_loss_clip": 0.01123596,
      "auxiliary_loss_mlp": 0.01028952,
      "balance_loss_clip": 1.04238117,
      "balance_loss_mlp": 1.02044058,
      "epoch": 0.5652618288943666,
      "flos": 25520026014720.0,
      "grad_norm": 2.2199005516160755,
      "language_loss": 0.76225734,
      "learning_rate": 1.675826355883496e-06,
      "loss": 0.78378284,
      "num_input_tokens_seen": 101424160,
      "step": 4701,
      "time_per_iteration": 2.753084182739258
    },
    {
      "auxiliary_loss_clip": 0.0114633,
      "auxiliary_loss_mlp": 0.01028191,
      "balance_loss_clip": 1.04971194,
      "balance_loss_mlp": 1.02059126,
      "epoch": 0.5653820717850057,
      "flos": 19683105937920.0,
      "grad_norm": 2.0392810157581414,
      "language_loss": 0.792328,
      "learning_rate": 1.6750577095751848e-06,
      "loss": 0.8140732,
      "num_input_tokens_seen": 101443270,
      "step": 4702,
      "time_per_iteration": 2.7329561710357666
    },
    {
      "auxiliary_loss_clip": 0.01178658,
      "auxiliary_loss_mlp": 0.01025083,
      "balance_loss_clip": 1.05198693,
      "balance_loss_mlp": 1.01773429,
      "epoch": 0.5655023146756448,
      "flos": 26979722910720.0,
      "grad_norm": 1.8156720423021762,
      "language_loss": 0.72827089,
      "learning_rate": 1.6742891125603605e-06,
      "loss": 0.75030828,
      "num_input_tokens_seen": 101464175,
      "step": 4703,
      "time_per_iteration": 2.6545629501342773
    },
    {
      "auxiliary_loss_clip": 0.01165895,
      "auxiliary_loss_mlp": 0.01025287,
      "balance_loss_clip": 1.05004811,
      "balance_loss_mlp": 1.01667428,
      "epoch": 0.5656225575662839,
      "flos": 27669351104640.0,
      "grad_norm": 2.2336567512951695,
      "language_loss": 0.71963716,
      "learning_rate": 1.6735205649556185e-06,
      "loss": 0.74154902,
      "num_input_tokens_seen": 101484045,
      "step": 4704,
      "time_per_iteration": 2.7313594818115234
    },
    {
      "auxiliary_loss_clip": 0.01145256,
      "auxiliary_loss_mlp": 0.01024171,
      "balance_loss_clip": 1.05019951,
      "balance_loss_mlp": 1.01679826,
      "epoch": 0.5657428004569229,
      "flos": 24349732997760.0,
      "grad_norm": 1.5637498540681032,
      "language_loss": 0.85015428,
      "learning_rate": 1.6727520668775476e-06,
      "loss": 0.87184858,
      "num_input_tokens_seen": 101504330,
      "step": 4705,
      "time_per_iteration": 2.6997270584106445
    },
    {
      "auxiliary_loss_clip": 0.01181357,
      "auxiliary_loss_mlp": 0.01032041,
      "balance_loss_clip": 1.05035233,
      "balance_loss_mlp": 1.02394664,
      "epoch": 0.5658630433475621,
      "flos": 21944041562880.0,
      "grad_norm": 1.6281431111509017,
      "language_loss": 0.74962127,
      "learning_rate": 1.6719836184427275e-06,
      "loss": 0.77175528,
      "num_input_tokens_seen": 101524635,
      "step": 4706,
      "time_per_iteration": 2.674062967300415
    },
    {
      "auxiliary_loss_clip": 0.0114598,
      "auxiliary_loss_mlp": 0.01032561,
      "balance_loss_clip": 1.04451466,
      "balance_loss_mlp": 1.02433562,
      "epoch": 0.5659832862382012,
      "flos": 30409012218240.0,
      "grad_norm": 1.944619954142119,
      "language_loss": 0.64288545,
      "learning_rate": 1.671215219767733e-06,
      "loss": 0.66467083,
      "num_input_tokens_seen": 101544095,
      "step": 4707,
      "time_per_iteration": 2.6963918209075928
    },
    {
      "auxiliary_loss_clip": 0.01123936,
      "auxiliary_loss_mlp": 0.01027104,
      "balance_loss_clip": 1.04375541,
      "balance_loss_mlp": 1.01921797,
      "epoch": 0.5661035291288402,
      "flos": 13188194570880.0,
      "grad_norm": 2.6564815143361473,
      "language_loss": 0.76511264,
      "learning_rate": 1.670446870969127e-06,
      "loss": 0.786623,
      "num_input_tokens_seen": 101561760,
      "step": 4708,
      "time_per_iteration": 2.776554584503174
    },
    {
      "auxiliary_loss_clip": 0.0115417,
      "auxiliary_loss_mlp": 0.01026271,
      "balance_loss_clip": 1.04836178,
      "balance_loss_mlp": 1.01842988,
      "epoch": 0.5662237720194794,
      "flos": 16143032108160.0,
      "grad_norm": 2.021039342406061,
      "language_loss": 0.7978307,
      "learning_rate": 1.6696785721634685e-06,
      "loss": 0.81963515,
      "num_input_tokens_seen": 101576245,
      "step": 4709,
      "time_per_iteration": 2.6210153102874756
    },
    {
      "auxiliary_loss_clip": 0.01169614,
      "auxiliary_loss_mlp": 0.01029343,
      "balance_loss_clip": 1.05140567,
      "balance_loss_mlp": 1.02106953,
      "epoch": 0.5663440149101184,
      "flos": 17676848718720.0,
      "grad_norm": 7.392332701143294,
      "language_loss": 0.7351824,
      "learning_rate": 1.6689103234673086e-06,
      "loss": 0.75717199,
      "num_input_tokens_seen": 101594565,
      "step": 4710,
      "time_per_iteration": 2.618055820465088
    },
    {
      "auxiliary_loss_clip": 0.01151504,
      "auxiliary_loss_mlp": 0.01028069,
      "balance_loss_clip": 1.05026877,
      "balance_loss_mlp": 1.02013588,
      "epoch": 0.5664642578007575,
      "flos": 23368330627200.0,
      "grad_norm": 2.4630765834242996,
      "language_loss": 0.77050215,
      "learning_rate": 1.668142124997189e-06,
      "loss": 0.7922979,
      "num_input_tokens_seen": 101614225,
      "step": 4711,
      "time_per_iteration": 2.709811210632324
    },
    {
      "auxiliary_loss_clip": 0.01045409,
      "auxiliary_loss_mlp": 0.01005647,
      "balance_loss_clip": 1.01370883,
      "balance_loss_mlp": 1.00416303,
      "epoch": 0.5665845006913967,
      "flos": 65516470945920.0,
      "grad_norm": 0.7493948026478168,
      "language_loss": 0.59820884,
      "learning_rate": 1.6673739768696453e-06,
      "loss": 0.6187194,
      "num_input_tokens_seen": 101680795,
      "step": 4712,
      "time_per_iteration": 3.253933906555176
    },
    {
      "auxiliary_loss_clip": 0.01157574,
      "auxiliary_loss_mlp": 0.01027588,
      "balance_loss_clip": 1.04814649,
      "balance_loss_mlp": 1.01924896,
      "epoch": 0.5667047435820357,
      "flos": 26140885620480.0,
      "grad_norm": 1.780029599441437,
      "language_loss": 0.77186775,
      "learning_rate": 1.6666058792012052e-06,
      "loss": 0.79371935,
      "num_input_tokens_seen": 101701680,
      "step": 4713,
      "time_per_iteration": 2.714496612548828
    },
    {
      "auxiliary_loss_clip": 0.01062585,
      "auxiliary_loss_mlp": 0.01002736,
      "balance_loss_clip": 1.01085687,
      "balance_loss_mlp": 1.00138903,
      "epoch": 0.5668249864726748,
      "flos": 71866949725440.0,
      "grad_norm": 0.8779696345571035,
      "language_loss": 0.68687618,
      "learning_rate": 1.6658378321083878e-06,
      "loss": 0.70752943,
      "num_input_tokens_seen": 101766010,
      "step": 4714,
      "time_per_iteration": 3.24955677986145
    },
    {
      "auxiliary_loss_clip": 0.01112575,
      "auxiliary_loss_mlp": 0.0102566,
      "balance_loss_clip": 1.0449326,
      "balance_loss_mlp": 1.0174706,
      "epoch": 0.5669452293633139,
      "flos": 22195667312640.0,
      "grad_norm": 1.6704845474246885,
      "language_loss": 0.82267237,
      "learning_rate": 1.6650698357077055e-06,
      "loss": 0.8440547,
      "num_input_tokens_seen": 101783055,
      "step": 4715,
      "time_per_iteration": 2.9100146293640137
    },
    {
      "auxiliary_loss_clip": 0.01156,
      "auxiliary_loss_mlp": 0.01029063,
      "balance_loss_clip": 1.04740822,
      "balance_loss_mlp": 1.02086687,
      "epoch": 0.567065472253953,
      "flos": 18223193560320.0,
      "grad_norm": 3.245905905082006,
      "language_loss": 0.80695146,
      "learning_rate": 1.6643018901156632e-06,
      "loss": 0.82880211,
      "num_input_tokens_seen": 101802150,
      "step": 4716,
      "time_per_iteration": 2.96390962600708
    },
    {
      "auxiliary_loss_clip": 0.01154132,
      "auxiliary_loss_mlp": 0.01023372,
      "balance_loss_clip": 1.04751921,
      "balance_loss_mlp": 1.01546848,
      "epoch": 0.567185715144592,
      "flos": 20371548983040.0,
      "grad_norm": 2.8571552292920397,
      "language_loss": 0.79456997,
      "learning_rate": 1.6635339954487566e-06,
      "loss": 0.81634498,
      "num_input_tokens_seen": 101818025,
      "step": 4717,
      "time_per_iteration": 3.651630163192749
    },
    {
      "auxiliary_loss_clip": 0.01156929,
      "auxiliary_loss_mlp": 0.01030936,
      "balance_loss_clip": 1.04945445,
      "balance_loss_mlp": 1.0224725,
      "epoch": 0.5673059580352312,
      "flos": 23221348174080.0,
      "grad_norm": 1.95336007079537,
      "language_loss": 0.82057697,
      "learning_rate": 1.6627661518234765e-06,
      "loss": 0.84245563,
      "num_input_tokens_seen": 101837280,
      "step": 4718,
      "time_per_iteration": 3.7306814193725586
    },
    {
      "auxiliary_loss_clip": 0.01126826,
      "auxiliary_loss_mlp": 0.01026736,
      "balance_loss_clip": 1.04767931,
      "balance_loss_mlp": 1.01854038,
      "epoch": 0.5674262009258703,
      "flos": 21719599430400.0,
      "grad_norm": 1.6202535056846814,
      "language_loss": 0.85536981,
      "learning_rate": 1.661998359356302e-06,
      "loss": 0.87690544,
      "num_input_tokens_seen": 101856310,
      "step": 4719,
      "time_per_iteration": 3.6449577808380127
    },
    {
      "auxiliary_loss_clip": 0.01073219,
      "auxiliary_loss_mlp": 0.01001277,
      "balance_loss_clip": 1.01264477,
      "balance_loss_mlp": 0.99991763,
      "epoch": 0.5675464438165093,
      "flos": 67470369114240.0,
      "grad_norm": 0.7518157801377123,
      "language_loss": 0.55733144,
      "learning_rate": 1.6612306181637077e-06,
      "loss": 0.57807636,
      "num_input_tokens_seen": 101915635,
      "step": 4720,
      "time_per_iteration": 3.1676645278930664
    },
    {
      "auxiliary_loss_clip": 0.01137225,
      "auxiliary_loss_mlp": 0.01029907,
      "balance_loss_clip": 1.04608965,
      "balance_loss_mlp": 1.02264094,
      "epoch": 0.5676666867071485,
      "flos": 18879173688960.0,
      "grad_norm": 2.9805401531988096,
      "language_loss": 0.65901518,
      "learning_rate": 1.6604629283621598e-06,
      "loss": 0.68068647,
      "num_input_tokens_seen": 101933565,
      "step": 4721,
      "time_per_iteration": 3.547450065612793
    },
    {
      "auxiliary_loss_clip": 0.01184287,
      "auxiliary_loss_mlp": 0.01028436,
      "balance_loss_clip": 1.0530616,
      "balance_loss_mlp": 1.01973343,
      "epoch": 0.5677869295977875,
      "flos": 33546778744320.0,
      "grad_norm": 1.9998568794351688,
      "language_loss": 0.74333549,
      "learning_rate": 1.6596952900681152e-06,
      "loss": 0.7654627,
      "num_input_tokens_seen": 101954325,
      "step": 4722,
      "time_per_iteration": 2.729280948638916
    },
    {
      "auxiliary_loss_clip": 0.01115595,
      "auxiliary_loss_mlp": 0.01032343,
      "balance_loss_clip": 1.04742289,
      "balance_loss_mlp": 1.02336669,
      "epoch": 0.5679071724884266,
      "flos": 28037256157440.0,
      "grad_norm": 2.2360990370454807,
      "language_loss": 0.82481933,
      "learning_rate": 1.658927703398025e-06,
      "loss": 0.84629869,
      "num_input_tokens_seen": 101974390,
      "step": 4723,
      "time_per_iteration": 2.762680768966675
    },
    {
      "auxiliary_loss_clip": 0.01120654,
      "auxiliary_loss_mlp": 0.01025621,
      "balance_loss_clip": 1.03929734,
      "balance_loss_mlp": 1.01755118,
      "epoch": 0.5680274153790658,
      "flos": 23550110380800.0,
      "grad_norm": 2.86813818670903,
      "language_loss": 0.77606392,
      "learning_rate": 1.6581601684683309e-06,
      "loss": 0.7975266,
      "num_input_tokens_seen": 101994815,
      "step": 4724,
      "time_per_iteration": 2.7953555583953857
    },
    {
      "auxiliary_loss_clip": 0.01167497,
      "auxiliary_loss_mlp": 0.01029347,
      "balance_loss_clip": 1.05139279,
      "balance_loss_mlp": 1.02108586,
      "epoch": 0.5681476582697048,
      "flos": 22455158140800.0,
      "grad_norm": 3.094482368370397,
      "language_loss": 0.68461072,
      "learning_rate": 1.6573926853954674e-06,
      "loss": 0.70657915,
      "num_input_tokens_seen": 102012400,
      "step": 4725,
      "time_per_iteration": 2.6643662452697754
    },
    {
      "auxiliary_loss_clip": 0.01146093,
      "auxiliary_loss_mlp": 0.01030442,
      "balance_loss_clip": 1.04643941,
      "balance_loss_mlp": 1.02320051,
      "epoch": 0.5682679011603439,
      "flos": 19536913584000.0,
      "grad_norm": 2.129182663745878,
      "language_loss": 0.83072746,
      "learning_rate": 1.6566252542958608e-06,
      "loss": 0.85249281,
      "num_input_tokens_seen": 102031900,
      "step": 4726,
      "time_per_iteration": 2.685662269592285
    },
    {
      "auxiliary_loss_clip": 0.01128263,
      "auxiliary_loss_mlp": 0.01032073,
      "balance_loss_clip": 1.0448916,
      "balance_loss_mlp": 1.02338552,
      "epoch": 0.568388144050983,
      "flos": 28765488493440.0,
      "grad_norm": 2.2135270746356164,
      "language_loss": 0.78581095,
      "learning_rate": 1.6558578752859305e-06,
      "loss": 0.80741423,
      "num_input_tokens_seen": 102050860,
      "step": 4727,
      "time_per_iteration": 2.758790969848633
    },
    {
      "auxiliary_loss_clip": 0.01135694,
      "auxiliary_loss_mlp": 0.01029122,
      "balance_loss_clip": 1.04406118,
      "balance_loss_mlp": 1.02028227,
      "epoch": 0.5685083869416221,
      "flos": 21209452519680.0,
      "grad_norm": 1.7668884400671239,
      "language_loss": 0.79134822,
      "learning_rate": 1.6550905484820865e-06,
      "loss": 0.81299645,
      "num_input_tokens_seen": 102069320,
      "step": 4728,
      "time_per_iteration": 2.741384744644165
    },
    {
      "auxiliary_loss_clip": 0.01182285,
      "auxiliary_loss_mlp": 0.01037926,
      "balance_loss_clip": 1.05184817,
      "balance_loss_mlp": 1.02904439,
      "epoch": 0.5686286298322611,
      "flos": 24827021942400.0,
      "grad_norm": 2.262928547137723,
      "language_loss": 0.78927761,
      "learning_rate": 1.6543232740007328e-06,
      "loss": 0.81147981,
      "num_input_tokens_seen": 102086435,
      "step": 4729,
      "time_per_iteration": 2.6308155059814453
    },
    {
      "auxiliary_loss_clip": 0.01169339,
      "auxiliary_loss_mlp": 0.01032598,
      "balance_loss_clip": 1.04993403,
      "balance_loss_mlp": 1.02462316,
      "epoch": 0.5687488727229003,
      "flos": 26615121909120.0,
      "grad_norm": 2.494375921299094,
      "language_loss": 0.66834134,
      "learning_rate": 1.653556051958263e-06,
      "loss": 0.69036072,
      "num_input_tokens_seen": 102106115,
      "step": 4730,
      "time_per_iteration": 2.70350980758667
    },
    {
      "auxiliary_loss_clip": 0.01092268,
      "auxiliary_loss_mlp": 0.01028582,
      "balance_loss_clip": 1.04137075,
      "balance_loss_mlp": 1.01976013,
      "epoch": 0.5688691156135394,
      "flos": 20808725414400.0,
      "grad_norm": 1.7114310408710192,
      "language_loss": 0.73470879,
      "learning_rate": 1.6527888824710642e-06,
      "loss": 0.75591731,
      "num_input_tokens_seen": 102125715,
      "step": 4731,
      "time_per_iteration": 2.721170425415039
    },
    {
      "auxiliary_loss_clip": 0.01126342,
      "auxiliary_loss_mlp": 0.01031462,
      "balance_loss_clip": 1.04130578,
      "balance_loss_mlp": 1.02306914,
      "epoch": 0.5689893585041784,
      "flos": 25880963829120.0,
      "grad_norm": 2.5488976194060267,
      "language_loss": 0.76722813,
      "learning_rate": 1.6520217656555166e-06,
      "loss": 0.7888062,
      "num_input_tokens_seen": 102145005,
      "step": 4732,
      "time_per_iteration": 2.8119709491729736
    },
    {
      "auxiliary_loss_clip": 0.01137748,
      "auxiliary_loss_mlp": 0.01024202,
      "balance_loss_clip": 1.04397845,
      "balance_loss_mlp": 1.0160538,
      "epoch": 0.5691096013948175,
      "flos": 23477463123840.0,
      "grad_norm": 1.7618986310024225,
      "language_loss": 0.70529413,
      "learning_rate": 1.65125470162799e-06,
      "loss": 0.72691369,
      "num_input_tokens_seen": 102165360,
      "step": 4733,
      "time_per_iteration": 2.7242531776428223
    },
    {
      "auxiliary_loss_clip": 0.01140142,
      "auxiliary_loss_mlp": 0.01034352,
      "balance_loss_clip": 1.04557419,
      "balance_loss_mlp": 1.02618647,
      "epoch": 0.5692298442854566,
      "flos": 18075600576000.0,
      "grad_norm": 2.2049772232491325,
      "language_loss": 0.69626254,
      "learning_rate": 1.6504876905048485e-06,
      "loss": 0.7180075,
      "num_input_tokens_seen": 102182320,
      "step": 4734,
      "time_per_iteration": 2.664064645767212
    },
    {
      "auxiliary_loss_clip": 0.01179993,
      "auxiliary_loss_mlp": 0.01028844,
      "balance_loss_clip": 1.05296361,
      "balance_loss_mlp": 1.02129853,
      "epoch": 0.5693500871760957,
      "flos": 23039317025280.0,
      "grad_norm": 1.6142362854811159,
      "language_loss": 0.7223109,
      "learning_rate": 1.6497207324024464e-06,
      "loss": 0.74439925,
      "num_input_tokens_seen": 102201220,
      "step": 4735,
      "time_per_iteration": 2.7004737854003906
    },
    {
      "auxiliary_loss_clip": 0.01158031,
      "auxiliary_loss_mlp": 0.01028243,
      "balance_loss_clip": 1.04732978,
      "balance_loss_mlp": 1.02052426,
      "epoch": 0.5694703300667348,
      "flos": 18989670902400.0,
      "grad_norm": 1.8028209091249696,
      "language_loss": 0.82902896,
      "learning_rate": 1.6489538274371305e-06,
      "loss": 0.85089177,
      "num_input_tokens_seen": 102219825,
      "step": 4736,
      "time_per_iteration": 2.6663053035736084
    },
    {
      "auxiliary_loss_clip": 0.01160869,
      "auxiliary_loss_mlp": 0.01030054,
      "balance_loss_clip": 1.05115271,
      "balance_loss_mlp": 1.02215064,
      "epoch": 0.5695905729573739,
      "flos": 21908705558400.0,
      "grad_norm": 1.874104928532494,
      "language_loss": 0.83129978,
      "learning_rate": 1.6481869757252396e-06,
      "loss": 0.85320902,
      "num_input_tokens_seen": 102238160,
      "step": 4737,
      "time_per_iteration": 2.688802719116211
    },
    {
      "auxiliary_loss_clip": 0.01165494,
      "auxiliary_loss_mlp": 0.01030232,
      "balance_loss_clip": 1.04988134,
      "balance_loss_mlp": 1.0221374,
      "epoch": 0.569710815848013,
      "flos": 28476659232000.0,
      "grad_norm": 1.4912227449595794,
      "language_loss": 0.71969545,
      "learning_rate": 1.647420177383105e-06,
      "loss": 0.74165267,
      "num_input_tokens_seen": 102261030,
      "step": 4738,
      "time_per_iteration": 2.6924633979797363
    },
    {
      "auxiliary_loss_clip": 0.01164544,
      "auxiliary_loss_mlp": 0.01028642,
      "balance_loss_clip": 1.05369902,
      "balance_loss_mlp": 1.0210309,
      "epoch": 0.569831058738652,
      "flos": 28366162018560.0,
      "grad_norm": 1.9963237409571895,
      "language_loss": 0.72659218,
      "learning_rate": 1.646653432527049e-06,
      "loss": 0.74852395,
      "num_input_tokens_seen": 102281670,
      "step": 4739,
      "time_per_iteration": 2.8014814853668213
    },
    {
      "auxiliary_loss_clip": 0.01138414,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.04756534,
      "balance_loss_mlp": 1.01833904,
      "epoch": 0.5699513016292912,
      "flos": 25849973370240.0,
      "grad_norm": 1.4916160629304287,
      "language_loss": 0.74416494,
      "learning_rate": 1.645886741273387e-06,
      "loss": 0.76580769,
      "num_input_tokens_seen": 102303485,
      "step": 4740,
      "time_per_iteration": 2.767712116241455
    },
    {
      "auxiliary_loss_clip": 0.01132245,
      "auxiliary_loss_mlp": 0.01024948,
      "balance_loss_clip": 1.04987788,
      "balance_loss_mlp": 1.01679444,
      "epoch": 0.5700715445199303,
      "flos": 18037858360320.0,
      "grad_norm": 1.9387046903319267,
      "language_loss": 0.73816633,
      "learning_rate": 1.645120103738424e-06,
      "loss": 0.75973827,
      "num_input_tokens_seen": 102320995,
      "step": 4741,
      "time_per_iteration": 2.8393783569335938
    },
    {
      "auxiliary_loss_clip": 0.01152744,
      "auxiliary_loss_mlp": 0.00762209,
      "balance_loss_clip": 1.04685116,
      "balance_loss_mlp": 1.00022197,
      "epoch": 0.5701917874105693,
      "flos": 11473352392320.0,
      "grad_norm": 2.292097819540539,
      "language_loss": 0.83757287,
      "learning_rate": 1.6443535200384591e-06,
      "loss": 0.85672241,
      "num_input_tokens_seen": 102339170,
      "step": 4742,
      "time_per_iteration": 2.6524109840393066
    },
    {
      "auxiliary_loss_clip": 0.01182585,
      "auxiliary_loss_mlp": 0.01027454,
      "balance_loss_clip": 1.05522537,
      "balance_loss_mlp": 1.01943731,
      "epoch": 0.5703120303012085,
      "flos": 21761759018880.0,
      "grad_norm": 1.893381218909576,
      "language_loss": 0.70646393,
      "learning_rate": 1.6435869902897827e-06,
      "loss": 0.72856438,
      "num_input_tokens_seen": 102357750,
      "step": 4743,
      "time_per_iteration": 3.5527892112731934
    },
    {
      "auxiliary_loss_clip": 0.01040169,
      "auxiliary_loss_mlp": 0.01001466,
      "balance_loss_clip": 1.01211977,
      "balance_loss_mlp": 1.00003552,
      "epoch": 0.5704322731918475,
      "flos": 56746258513920.0,
      "grad_norm": 1.034349178747778,
      "language_loss": 0.61949646,
      "learning_rate": 1.6428205146086764e-06,
      "loss": 0.63991284,
      "num_input_tokens_seen": 102419730,
      "step": 4744,
      "time_per_iteration": 5.237643718719482
    },
    {
      "auxiliary_loss_clip": 0.01155123,
      "auxiliary_loss_mlp": 0.01025617,
      "balance_loss_clip": 1.0472393,
      "balance_loss_mlp": 1.01790428,
      "epoch": 0.5705525160824866,
      "flos": 20741141975040.0,
      "grad_norm": 1.4586842549074204,
      "language_loss": 0.70616704,
      "learning_rate": 1.6420540931114142e-06,
      "loss": 0.72797447,
      "num_input_tokens_seen": 102440320,
      "step": 4745,
      "time_per_iteration": 2.735504388809204
    },
    {
      "auxiliary_loss_clip": 0.01151085,
      "auxiliary_loss_mlp": 0.01029578,
      "balance_loss_clip": 1.04777765,
      "balance_loss_mlp": 1.02171052,
      "epoch": 0.5706727589731257,
      "flos": 18771262254720.0,
      "grad_norm": 1.811355173018474,
      "language_loss": 0.79025048,
      "learning_rate": 1.6412877259142616e-06,
      "loss": 0.81205714,
      "num_input_tokens_seen": 102460240,
      "step": 4746,
      "time_per_iteration": 2.655766725540161
    },
    {
      "auxiliary_loss_clip": 0.0114691,
      "auxiliary_loss_mlp": 0.01032386,
      "balance_loss_clip": 1.04662859,
      "balance_loss_mlp": 1.0243516,
      "epoch": 0.5707930018637648,
      "flos": 27634733372160.0,
      "grad_norm": 2.0610969293286874,
      "language_loss": 0.73927337,
      "learning_rate": 1.6405214131334757e-06,
      "loss": 0.76106632,
      "num_input_tokens_seen": 102478765,
      "step": 4747,
      "time_per_iteration": 3.6808650493621826
    },
    {
      "auxiliary_loss_clip": 0.01120672,
      "auxiliary_loss_mlp": 0.01026471,
      "balance_loss_clip": 1.04742873,
      "balance_loss_mlp": 1.01894021,
      "epoch": 0.5709132447544039,
      "flos": 27597673514880.0,
      "grad_norm": 2.407921030969635,
      "language_loss": 0.79225737,
      "learning_rate": 1.6397551548853052e-06,
      "loss": 0.81372881,
      "num_input_tokens_seen": 102496930,
      "step": 4748,
      "time_per_iteration": 2.8126418590545654
    },
    {
      "auxiliary_loss_clip": 0.0114901,
      "auxiliary_loss_mlp": 0.01023376,
      "balance_loss_clip": 1.04649997,
      "balance_loss_mlp": 1.01516211,
      "epoch": 0.571033487645043,
      "flos": 21686095019520.0,
      "grad_norm": 1.958676945865462,
      "language_loss": 0.7090112,
      "learning_rate": 1.6389889512859917e-06,
      "loss": 0.73073506,
      "num_input_tokens_seen": 102516590,
      "step": 4749,
      "time_per_iteration": 2.731142520904541
    },
    {
      "auxiliary_loss_clip": 0.01050565,
      "auxiliary_loss_mlp": 0.01001315,
      "balance_loss_clip": 1.01193082,
      "balance_loss_mlp": 0.99999231,
      "epoch": 0.5711537305356821,
      "flos": 70181445980160.0,
      "grad_norm": 0.8110282473493834,
      "language_loss": 0.60302418,
      "learning_rate": 1.638222802451767e-06,
      "loss": 0.62354302,
      "num_input_tokens_seen": 102578070,
      "step": 4750,
      "time_per_iteration": 3.2817327976226807
    },
    {
      "auxiliary_loss_clip": 0.01159368,
      "auxiliary_loss_mlp": 0.01027421,
      "balance_loss_clip": 1.05030334,
      "balance_loss_mlp": 1.01977324,
      "epoch": 0.5712739734263211,
      "flos": 24717494396160.0,
      "grad_norm": 1.9329624921447397,
      "language_loss": 0.75452483,
      "learning_rate": 1.6374567084988561e-06,
      "loss": 0.77639276,
      "num_input_tokens_seen": 102599255,
      "step": 4751,
      "time_per_iteration": 2.7464630603790283
    },
    {
      "auxiliary_loss_clip": 0.0115799,
      "auxiliary_loss_mlp": 0.01028624,
      "balance_loss_clip": 1.05236816,
      "balance_loss_mlp": 1.02029753,
      "epoch": 0.5713942163169603,
      "flos": 26578169792640.0,
      "grad_norm": 1.8112156398654655,
      "language_loss": 0.76546997,
      "learning_rate": 1.6366906695434738e-06,
      "loss": 0.78733611,
      "num_input_tokens_seen": 102621775,
      "step": 4752,
      "time_per_iteration": 2.7545156478881836
    },
    {
      "auxiliary_loss_clip": 0.01163674,
      "auxiliary_loss_mlp": 0.0103046,
      "balance_loss_clip": 1.04930389,
      "balance_loss_mlp": 1.02249694,
      "epoch": 0.5715144592075994,
      "flos": 21142443697920.0,
      "grad_norm": 2.099414425886561,
      "language_loss": 0.85860556,
      "learning_rate": 1.6359246857018275e-06,
      "loss": 0.88054693,
      "num_input_tokens_seen": 102639305,
      "step": 4753,
      "time_per_iteration": 2.59777569770813
    },
    {
      "auxiliary_loss_clip": 0.01122212,
      "auxiliary_loss_mlp": 0.01026896,
      "balance_loss_clip": 1.04305625,
      "balance_loss_mlp": 1.01936245,
      "epoch": 0.5716347020982384,
      "flos": 23330265189120.0,
      "grad_norm": 2.28662688857944,
      "language_loss": 0.78246939,
      "learning_rate": 1.6351587570901178e-06,
      "loss": 0.8039605,
      "num_input_tokens_seen": 102659430,
      "step": 4754,
      "time_per_iteration": 2.753528118133545
    },
    {
      "auxiliary_loss_clip": 0.01138322,
      "auxiliary_loss_mlp": 0.01030386,
      "balance_loss_clip": 1.04862046,
      "balance_loss_mlp": 1.02186894,
      "epoch": 0.5717549449888776,
      "flos": 17009555806080.0,
      "grad_norm": 2.6946333470368207,
      "language_loss": 0.75951552,
      "learning_rate": 1.634392883824534e-06,
      "loss": 0.78120255,
      "num_input_tokens_seen": 102671430,
      "step": 4755,
      "time_per_iteration": 2.6603033542633057
    },
    {
      "auxiliary_loss_clip": 0.01127162,
      "auxiliary_loss_mlp": 0.01028153,
      "balance_loss_clip": 1.04544652,
      "balance_loss_mlp": 1.0203743,
      "epoch": 0.5718751878795166,
      "flos": 35518130922240.0,
      "grad_norm": 1.8117781874392749,
      "language_loss": 0.6793381,
      "learning_rate": 1.6336270660212595e-06,
      "loss": 0.70089126,
      "num_input_tokens_seen": 102693025,
      "step": 4756,
      "time_per_iteration": 2.896963357925415
    },
    {
      "auxiliary_loss_clip": 0.01152108,
      "auxiliary_loss_mlp": 0.01034287,
      "balance_loss_clip": 1.05393362,
      "balance_loss_mlp": 1.02626467,
      "epoch": 0.5719954307701557,
      "flos": 38613989255040.0,
      "grad_norm": 2.83015982457641,
      "language_loss": 0.66014016,
      "learning_rate": 1.6328613037964676e-06,
      "loss": 0.68200409,
      "num_input_tokens_seen": 102716090,
      "step": 4757,
      "time_per_iteration": 2.857815742492676
    },
    {
      "auxiliary_loss_clip": 0.0116443,
      "auxiliary_loss_mlp": 0.01024632,
      "balance_loss_clip": 1.04841745,
      "balance_loss_mlp": 1.01702058,
      "epoch": 0.5721156736607949,
      "flos": 20631111638400.0,
      "grad_norm": 1.8347617046797027,
      "language_loss": 0.68173903,
      "learning_rate": 1.6320955972663241e-06,
      "loss": 0.70362961,
      "num_input_tokens_seen": 102735685,
      "step": 4758,
      "time_per_iteration": 2.626655101776123
    },
    {
      "auxiliary_loss_clip": 0.01162618,
      "auxiliary_loss_mlp": 0.01026185,
      "balance_loss_clip": 1.0480268,
      "balance_loss_mlp": 1.01837707,
      "epoch": 0.5722359165514339,
      "flos": 37415076076800.0,
      "grad_norm": 2.4449492373539727,
      "language_loss": 0.65299791,
      "learning_rate": 1.6313299465469857e-06,
      "loss": 0.67488593,
      "num_input_tokens_seen": 102758415,
      "step": 4759,
      "time_per_iteration": 2.8303604125976562
    },
    {
      "auxiliary_loss_clip": 0.01160888,
      "auxiliary_loss_mlp": 0.01029202,
      "balance_loss_clip": 1.04828072,
      "balance_loss_mlp": 1.02062488,
      "epoch": 0.572356159442073,
      "flos": 21972877205760.0,
      "grad_norm": 2.6416889958032757,
      "language_loss": 0.79556584,
      "learning_rate": 1.6305643517546014e-06,
      "loss": 0.81746674,
      "num_input_tokens_seen": 102773795,
      "step": 4760,
      "time_per_iteration": 2.6117725372314453
    },
    {
      "auxiliary_loss_clip": 0.01179032,
      "auxiliary_loss_mlp": 0.01028901,
      "balance_loss_clip": 1.05169535,
      "balance_loss_mlp": 1.0212183,
      "epoch": 0.5724764023327121,
      "flos": 19135540033920.0,
      "grad_norm": 1.826157504749068,
      "language_loss": 0.84550393,
      "learning_rate": 1.629798813005311e-06,
      "loss": 0.86758327,
      "num_input_tokens_seen": 102793515,
      "step": 4761,
      "time_per_iteration": 2.6745831966400146
    },
    {
      "auxiliary_loss_clip": 0.01120616,
      "auxiliary_loss_mlp": 0.01025577,
      "balance_loss_clip": 1.04600906,
      "balance_loss_mlp": 1.0177536,
      "epoch": 0.5725966452233512,
      "flos": 22819759142400.0,
      "grad_norm": 1.9917075510248097,
      "language_loss": 0.70698607,
      "learning_rate": 1.6290333304152473e-06,
      "loss": 0.72844803,
      "num_input_tokens_seen": 102813390,
      "step": 4762,
      "time_per_iteration": 2.7662999629974365
    },
    {
      "auxiliary_loss_clip": 0.01147307,
      "auxiliary_loss_mlp": 0.01030145,
      "balance_loss_clip": 1.05144298,
      "balance_loss_mlp": 1.02205706,
      "epoch": 0.5727168881139902,
      "flos": 41496610498560.0,
      "grad_norm": 1.784370124125701,
      "language_loss": 0.57096016,
      "learning_rate": 1.6282679041005314e-06,
      "loss": 0.59273475,
      "num_input_tokens_seen": 102838980,
      "step": 4763,
      "time_per_iteration": 2.90012264251709
    },
    {
      "auxiliary_loss_clip": 0.01139696,
      "auxiliary_loss_mlp": 0.01027915,
      "balance_loss_clip": 1.04210019,
      "balance_loss_mlp": 1.01998758,
      "epoch": 0.5728371310046293,
      "flos": 14647675985280.0,
      "grad_norm": 2.26006617253784,
      "language_loss": 0.87213522,
      "learning_rate": 1.6275025341772789e-06,
      "loss": 0.89381135,
      "num_input_tokens_seen": 102855285,
      "step": 4764,
      "time_per_iteration": 2.6794021129608154
    },
    {
      "auxiliary_loss_clip": 0.01151278,
      "auxiliary_loss_mlp": 0.01036406,
      "balance_loss_clip": 1.04751873,
      "balance_loss_mlp": 1.02834797,
      "epoch": 0.5729573738952685,
      "flos": 21506613736320.0,
      "grad_norm": 2.4060288070597085,
      "language_loss": 0.81941235,
      "learning_rate": 1.626737220761596e-06,
      "loss": 0.84128922,
      "num_input_tokens_seen": 102872750,
      "step": 4765,
      "time_per_iteration": 2.6775851249694824
    },
    {
      "auxiliary_loss_clip": 0.01164314,
      "auxiliary_loss_mlp": 0.01026416,
      "balance_loss_clip": 1.05143964,
      "balance_loss_mlp": 1.01870871,
      "epoch": 0.5730776167859075,
      "flos": 23621680229760.0,
      "grad_norm": 3.092610066699434,
      "language_loss": 0.7868886,
      "learning_rate": 1.62597196396958e-06,
      "loss": 0.80879593,
      "num_input_tokens_seen": 102890920,
      "step": 4766,
      "time_per_iteration": 2.658482789993286
    },
    {
      "auxiliary_loss_clip": 0.01165042,
      "auxiliary_loss_mlp": 0.01028582,
      "balance_loss_clip": 1.05115998,
      "balance_loss_mlp": 1.02062476,
      "epoch": 0.5731978596765466,
      "flos": 25739224761600.0,
      "grad_norm": 1.9985190408509372,
      "language_loss": 0.8589375,
      "learning_rate": 1.6252067639173197e-06,
      "loss": 0.8808738,
      "num_input_tokens_seen": 102912830,
      "step": 4767,
      "time_per_iteration": 2.6694159507751465
    },
    {
      "auxiliary_loss_clip": 0.01168266,
      "auxiliary_loss_mlp": 0.01029013,
      "balance_loss_clip": 1.05068624,
      "balance_loss_mlp": 1.02142262,
      "epoch": 0.5733181025671857,
      "flos": 26359509749760.0,
      "grad_norm": 1.7799905302689805,
      "language_loss": 0.69406056,
      "learning_rate": 1.6244416207208956e-06,
      "loss": 0.71603334,
      "num_input_tokens_seen": 102933765,
      "step": 4768,
      "time_per_iteration": 2.755201816558838
    },
    {
      "auxiliary_loss_clip": 0.01137898,
      "auxiliary_loss_mlp": 0.01030892,
      "balance_loss_clip": 1.04786551,
      "balance_loss_mlp": 1.02242827,
      "epoch": 0.5734383454578248,
      "flos": 29423874833280.0,
      "grad_norm": 1.6698256192247165,
      "language_loss": 0.73371649,
      "learning_rate": 1.6236765344963787e-06,
      "loss": 0.75540447,
      "num_input_tokens_seen": 102955025,
      "step": 4769,
      "time_per_iteration": 3.7249560356140137
    },
    {
      "auxiliary_loss_clip": 0.01151921,
      "auxiliary_loss_mlp": 0.01032209,
      "balance_loss_clip": 1.04821837,
      "balance_loss_mlp": 1.02436543,
      "epoch": 0.5735585883484638,
      "flos": 34969954487040.0,
      "grad_norm": 2.177647406318903,
      "language_loss": 0.69201505,
      "learning_rate": 1.6229115053598322e-06,
      "loss": 0.71385646,
      "num_input_tokens_seen": 102976780,
      "step": 4770,
      "time_per_iteration": 3.709921360015869
    },
    {
      "auxiliary_loss_clip": 0.01165618,
      "auxiliary_loss_mlp": 0.01025208,
      "balance_loss_clip": 1.05149174,
      "balance_loss_mlp": 1.01732254,
      "epoch": 0.573678831239103,
      "flos": 18770759464320.0,
      "grad_norm": 2.221817620577284,
      "language_loss": 0.72385871,
      "learning_rate": 1.6221465334273108e-06,
      "loss": 0.745767,
      "num_input_tokens_seen": 102995990,
      "step": 4771,
      "time_per_iteration": 3.536027431488037
    },
    {
      "auxiliary_loss_clip": 0.01141759,
      "auxiliary_loss_mlp": 0.0102992,
      "balance_loss_clip": 1.04757071,
      "balance_loss_mlp": 1.02129507,
      "epoch": 0.5737990741297421,
      "flos": 25702883176320.0,
      "grad_norm": 1.9283236277087668,
      "language_loss": 0.61365873,
      "learning_rate": 1.6213816188148593e-06,
      "loss": 0.63537556,
      "num_input_tokens_seen": 103014695,
      "step": 4772,
      "time_per_iteration": 2.6991376876831055
    },
    {
      "auxiliary_loss_clip": 0.0114112,
      "auxiliary_loss_mlp": 0.01023175,
      "balance_loss_clip": 1.04793286,
      "balance_loss_mlp": 1.01563537,
      "epoch": 0.5739193170203811,
      "flos": 27269234530560.0,
      "grad_norm": 1.6764218683516148,
      "language_loss": 0.77354246,
      "learning_rate": 1.6206167616385162e-06,
      "loss": 0.79518539,
      "num_input_tokens_seen": 103035760,
      "step": 4773,
      "time_per_iteration": 3.6526262760162354
    },
    {
      "auxiliary_loss_clip": 0.01157894,
      "auxiliary_loss_mlp": 0.01028996,
      "balance_loss_clip": 1.05059326,
      "balance_loss_mlp": 1.02038932,
      "epoch": 0.5740395599110203,
      "flos": 12239721993600.0,
      "grad_norm": 2.095504114987213,
      "language_loss": 0.73755515,
      "learning_rate": 1.6198519620143078e-06,
      "loss": 0.75942397,
      "num_input_tokens_seen": 103052915,
      "step": 4774,
      "time_per_iteration": 2.7699227333068848
    },
    {
      "auxiliary_loss_clip": 0.01140627,
      "auxiliary_loss_mlp": 0.01030166,
      "balance_loss_clip": 1.04859126,
      "balance_loss_mlp": 1.02193499,
      "epoch": 0.5741598028016593,
      "flos": 25921399564800.0,
      "grad_norm": 1.5537592664489628,
      "language_loss": 0.78041452,
      "learning_rate": 1.6190872200582546e-06,
      "loss": 0.80212247,
      "num_input_tokens_seen": 103074655,
      "step": 4775,
      "time_per_iteration": 2.7176008224487305
    },
    {
      "auxiliary_loss_clip": 0.01144064,
      "auxiliary_loss_mlp": 0.00762969,
      "balance_loss_clip": 1.0453341,
      "balance_loss_mlp": 1.00039148,
      "epoch": 0.5742800456922984,
      "flos": 19244133826560.0,
      "grad_norm": 3.417168084251413,
      "language_loss": 0.78439486,
      "learning_rate": 1.6183225358863676e-06,
      "loss": 0.80346519,
      "num_input_tokens_seen": 103091550,
      "step": 4776,
      "time_per_iteration": 2.697296619415283
    },
    {
      "auxiliary_loss_clip": 0.01140934,
      "auxiliary_loss_mlp": 0.01028771,
      "balance_loss_clip": 1.04415965,
      "balance_loss_mlp": 1.02068865,
      "epoch": 0.5744002885829376,
      "flos": 30920487932160.0,
      "grad_norm": 2.095863713194777,
      "language_loss": 0.71682191,
      "learning_rate": 1.617557909614648e-06,
      "loss": 0.73851895,
      "num_input_tokens_seen": 103110985,
      "step": 4777,
      "time_per_iteration": 2.7670600414276123
    },
    {
      "auxiliary_loss_clip": 0.01134273,
      "auxiliary_loss_mlp": 0.01034729,
      "balance_loss_clip": 1.04433179,
      "balance_loss_mlp": 1.02717113,
      "epoch": 0.5745205314735766,
      "flos": 23840017050240.0,
      "grad_norm": 1.9315257643681496,
      "language_loss": 0.86138296,
      "learning_rate": 1.6167933413590899e-06,
      "loss": 0.88307303,
      "num_input_tokens_seen": 103129890,
      "step": 4778,
      "time_per_iteration": 2.758100986480713
    },
    {
      "auxiliary_loss_clip": 0.01164307,
      "auxiliary_loss_mlp": 0.01025369,
      "balance_loss_clip": 1.04871023,
      "balance_loss_mlp": 1.01729262,
      "epoch": 0.5746407743642157,
      "flos": 12311902373760.0,
      "grad_norm": 2.3543268769487913,
      "language_loss": 0.90495336,
      "learning_rate": 1.6160288312356773e-06,
      "loss": 0.92685014,
      "num_input_tokens_seen": 103147020,
      "step": 4779,
      "time_per_iteration": 2.584770679473877
    },
    {
      "auxiliary_loss_clip": 0.01167986,
      "auxiliary_loss_mlp": 0.01028317,
      "balance_loss_clip": 1.04870319,
      "balance_loss_mlp": 1.0207119,
      "epoch": 0.5747610172548548,
      "flos": 24133658734080.0,
      "grad_norm": 1.7234452507953941,
      "language_loss": 0.81762356,
      "learning_rate": 1.6152643793603857e-06,
      "loss": 0.83958656,
      "num_input_tokens_seen": 103167370,
      "step": 4780,
      "time_per_iteration": 2.7057905197143555
    },
    {
      "auxiliary_loss_clip": 0.01178458,
      "auxiliary_loss_mlp": 0.01031713,
      "balance_loss_clip": 1.05167627,
      "balance_loss_mlp": 1.02368414,
      "epoch": 0.5748812601454939,
      "flos": 25408451393280.0,
      "grad_norm": 1.7070949083966485,
      "language_loss": 0.87604463,
      "learning_rate": 1.6144999858491815e-06,
      "loss": 0.89814633,
      "num_input_tokens_seen": 103186000,
      "step": 4781,
      "time_per_iteration": 2.617354393005371
    },
    {
      "auxiliary_loss_clip": 0.01153906,
      "auxiliary_loss_mlp": 0.01021744,
      "balance_loss_clip": 1.04643893,
      "balance_loss_mlp": 1.01344109,
      "epoch": 0.575001503036133,
      "flos": 30624942827520.0,
      "grad_norm": 1.642124066481995,
      "language_loss": 0.85716397,
      "learning_rate": 1.6137356508180232e-06,
      "loss": 0.87892044,
      "num_input_tokens_seen": 103207710,
      "step": 4782,
      "time_per_iteration": 2.7888691425323486
    },
    {
      "auxiliary_loss_clip": 0.01176687,
      "auxiliary_loss_mlp": 0.00763158,
      "balance_loss_clip": 1.04846954,
      "balance_loss_mlp": 1.00030005,
      "epoch": 0.5751217459267721,
      "flos": 21726566668800.0,
      "grad_norm": 1.9522228511530828,
      "language_loss": 0.81645799,
      "learning_rate": 1.6129713743828593e-06,
      "loss": 0.83585644,
      "num_input_tokens_seen": 103226720,
      "step": 4783,
      "time_per_iteration": 2.636209726333618
    },
    {
      "auxiliary_loss_clip": 0.01149869,
      "auxiliary_loss_mlp": 0.01024267,
      "balance_loss_clip": 1.04473197,
      "balance_loss_mlp": 1.01631534,
      "epoch": 0.5752419888174112,
      "flos": 21651620941440.0,
      "grad_norm": 1.4487592488173378,
      "language_loss": 0.75577283,
      "learning_rate": 1.6122071566596306e-06,
      "loss": 0.77751422,
      "num_input_tokens_seen": 103246995,
      "step": 4784,
      "time_per_iteration": 2.7845652103424072
    },
    {
      "auxiliary_loss_clip": 0.01168928,
      "auxiliary_loss_mlp": 0.01026422,
      "balance_loss_clip": 1.05127776,
      "balance_loss_mlp": 1.01827431,
      "epoch": 0.5753622317080502,
      "flos": 17775997234560.0,
      "grad_norm": 2.2516838698757895,
      "language_loss": 0.83491778,
      "learning_rate": 1.6114429977642674e-06,
      "loss": 0.85687125,
      "num_input_tokens_seen": 103261500,
      "step": 4785,
      "time_per_iteration": 2.6625168323516846
    },
    {
      "auxiliary_loss_clip": 0.01165802,
      "auxiliary_loss_mlp": 0.01029312,
      "balance_loss_clip": 1.05305219,
      "balance_loss_mlp": 1.02105653,
      "epoch": 0.5754824745986894,
      "flos": 19789616741760.0,
      "grad_norm": 1.7282142493476607,
      "language_loss": 0.73689878,
      "learning_rate": 1.6106788978126926e-06,
      "loss": 0.75884992,
      "num_input_tokens_seen": 103280475,
      "step": 4786,
      "time_per_iteration": 2.656543493270874
    },
    {
      "auxiliary_loss_clip": 0.01116733,
      "auxiliary_loss_mlp": 0.01026896,
      "balance_loss_clip": 1.04239845,
      "balance_loss_mlp": 1.01893866,
      "epoch": 0.5756027174893285,
      "flos": 30985665160320.0,
      "grad_norm": 2.3024469624386543,
      "language_loss": 0.79120374,
      "learning_rate": 1.6099148569208196e-06,
      "loss": 0.81264007,
      "num_input_tokens_seen": 103297695,
      "step": 4787,
      "time_per_iteration": 2.88704252243042
    },
    {
      "auxiliary_loss_clip": 0.0115041,
      "auxiliary_loss_mlp": 0.01030075,
      "balance_loss_clip": 1.04919481,
      "balance_loss_mlp": 1.02194452,
      "epoch": 0.5757229603799675,
      "flos": 28546864364160.0,
      "grad_norm": 2.7075001635951264,
      "language_loss": 0.625754,
      "learning_rate": 1.6091508752045523e-06,
      "loss": 0.64755881,
      "num_input_tokens_seen": 103318575,
      "step": 4788,
      "time_per_iteration": 2.7798895835876465
    },
    {
      "auxiliary_loss_clip": 0.01126065,
      "auxiliary_loss_mlp": 0.01032161,
      "balance_loss_clip": 1.04167163,
      "balance_loss_mlp": 1.02381599,
      "epoch": 0.5758432032706067,
      "flos": 22999024944000.0,
      "grad_norm": 1.648430164990783,
      "language_loss": 0.86553705,
      "learning_rate": 1.608386952779787e-06,
      "loss": 0.88711929,
      "num_input_tokens_seen": 103337945,
      "step": 4789,
      "time_per_iteration": 2.839604616165161
    },
    {
      "auxiliary_loss_clip": 0.0115481,
      "auxiliary_loss_mlp": 0.01029522,
      "balance_loss_clip": 1.04789758,
      "balance_loss_mlp": 1.02136767,
      "epoch": 0.5759634461612457,
      "flos": 25739727552000.0,
      "grad_norm": 1.6721436864373276,
      "language_loss": 0.75079656,
      "learning_rate": 1.6076230897624098e-06,
      "loss": 0.77263981,
      "num_input_tokens_seen": 103360150,
      "step": 4790,
      "time_per_iteration": 2.770280361175537
    },
    {
      "auxiliary_loss_clip": 0.01165567,
      "auxiliary_loss_mlp": 0.01027046,
      "balance_loss_clip": 1.04670358,
      "balance_loss_mlp": 1.01929176,
      "epoch": 0.5760836890518848,
      "flos": 30591761639040.0,
      "grad_norm": 2.2616303685058328,
      "language_loss": 0.77408433,
      "learning_rate": 1.6068592862682974e-06,
      "loss": 0.79601049,
      "num_input_tokens_seen": 103378305,
      "step": 4791,
      "time_per_iteration": 2.7082903385162354
    },
    {
      "auxiliary_loss_clip": 0.01151264,
      "auxiliary_loss_mlp": 0.01030546,
      "balance_loss_clip": 1.04694104,
      "balance_loss_mlp": 1.02296388,
      "epoch": 0.576203931942524,
      "flos": 36538963447680.0,
      "grad_norm": 1.7701982789470414,
      "language_loss": 0.73684037,
      "learning_rate": 1.6060955424133187e-06,
      "loss": 0.75865847,
      "num_input_tokens_seen": 103399230,
      "step": 4792,
      "time_per_iteration": 2.9551002979278564
    },
    {
      "auxiliary_loss_clip": 0.01164441,
      "auxiliary_loss_mlp": 0.01027504,
      "balance_loss_clip": 1.0506537,
      "balance_loss_mlp": 1.01914167,
      "epoch": 0.576324174833163,
      "flos": 25516937445120.0,
      "grad_norm": 1.6864632674856155,
      "language_loss": 0.89452451,
      "learning_rate": 1.6053318583133332e-06,
      "loss": 0.91644394,
      "num_input_tokens_seen": 103420100,
      "step": 4793,
      "time_per_iteration": 2.668419599533081
    },
    {
      "auxiliary_loss_clip": 0.01162224,
      "auxiliary_loss_mlp": 0.01026347,
      "balance_loss_clip": 1.0489701,
      "balance_loss_mlp": 1.01864028,
      "epoch": 0.5764444177238021,
      "flos": 25119262995840.0,
      "grad_norm": 2.2683321821463682,
      "language_loss": 0.74831522,
      "learning_rate": 1.6045682340841907e-06,
      "loss": 0.77020091,
      "num_input_tokens_seen": 103439025,
      "step": 4794,
      "time_per_iteration": 2.7149057388305664
    },
    {
      "auxiliary_loss_clip": 0.01041976,
      "auxiliary_loss_mlp": 0.00754358,
      "balance_loss_clip": 1.0117383,
      "balance_loss_mlp": 1.00044274,
      "epoch": 0.5765646606144411,
      "flos": 62212687758720.0,
      "grad_norm": 0.7519051442805301,
      "language_loss": 0.57912672,
      "learning_rate": 1.6038046698417336e-06,
      "loss": 0.59709007,
      "num_input_tokens_seen": 103499920,
      "step": 4795,
      "time_per_iteration": 4.166090726852417
    },
    {
      "auxiliary_loss_clip": 0.01164881,
      "auxiliary_loss_mlp": 0.01025988,
      "balance_loss_clip": 1.04835653,
      "balance_loss_mlp": 1.01820421,
      "epoch": 0.5766849035050803,
      "flos": 25118760205440.0,
      "grad_norm": 2.6579629314622926,
      "language_loss": 0.69013107,
      "learning_rate": 1.6030411657017919e-06,
      "loss": 0.71203971,
      "num_input_tokens_seen": 103519575,
      "step": 4796,
      "time_per_iteration": 4.642420291900635
    },
    {
      "auxiliary_loss_clip": 0.01155057,
      "auxiliary_loss_mlp": 0.01027465,
      "balance_loss_clip": 1.04710126,
      "balance_loss_mlp": 1.01986504,
      "epoch": 0.5768051463957193,
      "flos": 15991093578240.0,
      "grad_norm": 1.781648522493346,
      "language_loss": 0.84323764,
      "learning_rate": 1.6022777217801903e-06,
      "loss": 0.86506283,
      "num_input_tokens_seen": 103536530,
      "step": 4797,
      "time_per_iteration": 2.662346363067627
    },
    {
      "auxiliary_loss_clip": 0.0113691,
      "auxiliary_loss_mlp": 0.01030625,
      "balance_loss_clip": 1.0493927,
      "balance_loss_mlp": 1.02257288,
      "epoch": 0.5769253892863584,
      "flos": 22163635359360.0,
      "grad_norm": 1.9763608687826577,
      "language_loss": 0.73670685,
      "learning_rate": 1.601514338192742e-06,
      "loss": 0.7583822,
      "num_input_tokens_seen": 103556460,
      "step": 4798,
      "time_per_iteration": 2.889357805252075
    },
    {
      "auxiliary_loss_clip": 0.01175207,
      "auxiliary_loss_mlp": 0.01023944,
      "balance_loss_clip": 1.04895258,
      "balance_loss_mlp": 1.01691628,
      "epoch": 0.5770456321769976,
      "flos": 22856388036480.0,
      "grad_norm": 2.0093607511826237,
      "language_loss": 0.71475172,
      "learning_rate": 1.6007510150552514e-06,
      "loss": 0.73674327,
      "num_input_tokens_seen": 103574520,
      "step": 4799,
      "time_per_iteration": 3.549248695373535
    },
    {
      "auxiliary_loss_clip": 0.01169663,
      "auxiliary_loss_mlp": 0.01032923,
      "balance_loss_clip": 1.04862952,
      "balance_loss_mlp": 1.02482247,
      "epoch": 0.5771658750676366,
      "flos": 46353672489600.0,
      "grad_norm": 2.357252598134425,
      "language_loss": 0.62375629,
      "learning_rate": 1.599987752483515e-06,
      "loss": 0.64578223,
      "num_input_tokens_seen": 103598965,
      "step": 4800,
      "time_per_iteration": 2.9047908782958984
    },
    {
      "auxiliary_loss_clip": 0.01129376,
      "auxiliary_loss_mlp": 0.01024747,
      "balance_loss_clip": 1.04389632,
      "balance_loss_mlp": 1.01680768,
      "epoch": 0.5772861179582757,
      "flos": 22159972172160.0,
      "grad_norm": 1.606018328135046,
      "language_loss": 0.68362385,
      "learning_rate": 1.5992245505933184e-06,
      "loss": 0.70516515,
      "num_input_tokens_seen": 103618665,
      "step": 4801,
      "time_per_iteration": 2.7833058834075928
    },
    {
      "auxiliary_loss_clip": 0.01179653,
      "auxiliary_loss_mlp": 0.01028443,
      "balance_loss_clip": 1.05195022,
      "balance_loss_mlp": 1.02071786,
      "epoch": 0.5774063608489148,
      "flos": 31248926916480.0,
      "grad_norm": 1.9964211535834067,
      "language_loss": 0.71462095,
      "learning_rate": 1.5984614095004388e-06,
      "loss": 0.73670197,
      "num_input_tokens_seen": 103639800,
      "step": 4802,
      "time_per_iteration": 2.7912344932556152
    },
    {
      "auxiliary_loss_clip": 0.01158333,
      "auxiliary_loss_mlp": 0.01030387,
      "balance_loss_clip": 1.04756069,
      "balance_loss_mlp": 1.02209592,
      "epoch": 0.5775266037395539,
      "flos": 22527123039360.0,
      "grad_norm": 2.2694279212731385,
      "language_loss": 0.8096993,
      "learning_rate": 1.5976983293206438e-06,
      "loss": 0.83158648,
      "num_input_tokens_seen": 103655605,
      "step": 4803,
      "time_per_iteration": 2.6690542697906494
    },
    {
      "auxiliary_loss_clip": 0.01143855,
      "auxiliary_loss_mlp": 0.01025673,
      "balance_loss_clip": 1.0429554,
      "balance_loss_mlp": 1.01790023,
      "epoch": 0.577646846630193,
      "flos": 21068790860160.0,
      "grad_norm": 1.837381740565245,
      "language_loss": 0.71404636,
      "learning_rate": 1.5969353101696928e-06,
      "loss": 0.73574162,
      "num_input_tokens_seen": 103674045,
      "step": 4804,
      "time_per_iteration": 2.7176706790924072
    },
    {
      "auxiliary_loss_clip": 0.01163037,
      "auxiliary_loss_mlp": 0.01025682,
      "balance_loss_clip": 1.04773378,
      "balance_loss_mlp": 1.01810324,
      "epoch": 0.5777670895208321,
      "flos": 29714284293120.0,
      "grad_norm": 3.610257481967702,
      "language_loss": 0.79829144,
      "learning_rate": 1.5961723521633341e-06,
      "loss": 0.82017857,
      "num_input_tokens_seen": 103695285,
      "step": 4805,
      "time_per_iteration": 2.842996835708618
    },
    {
      "auxiliary_loss_clip": 0.0114601,
      "auxiliary_loss_mlp": 0.01021914,
      "balance_loss_clip": 1.04530549,
      "balance_loss_mlp": 1.01463056,
      "epoch": 0.5778873324114712,
      "flos": 19500428344320.0,
      "grad_norm": 2.3407323841831826,
      "language_loss": 0.9093622,
      "learning_rate": 1.5954094554173097e-06,
      "loss": 0.93104148,
      "num_input_tokens_seen": 103713275,
      "step": 4806,
      "time_per_iteration": 2.729781150817871
    },
    {
      "auxiliary_loss_clip": 0.01155824,
      "auxiliary_loss_mlp": 0.01028364,
      "balance_loss_clip": 1.04847491,
      "balance_loss_mlp": 1.0211246,
      "epoch": 0.5780075753021102,
      "flos": 14136846716160.0,
      "grad_norm": 2.098557728275986,
      "language_loss": 0.79145223,
      "learning_rate": 1.5946466200473482e-06,
      "loss": 0.81329411,
      "num_input_tokens_seen": 103731185,
      "step": 4807,
      "time_per_iteration": 2.7304961681365967
    },
    {
      "auxiliary_loss_clip": 0.01154822,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.04680276,
      "balance_loss_mlp": 1.01835942,
      "epoch": 0.5781278181927494,
      "flos": 15262178883840.0,
      "grad_norm": 1.7911139202453887,
      "language_loss": 0.83283859,
      "learning_rate": 1.5938838461691723e-06,
      "loss": 0.85464102,
      "num_input_tokens_seen": 103748095,
      "step": 4808,
      "time_per_iteration": 2.645071506500244
    },
    {
      "auxiliary_loss_clip": 0.01178625,
      "auxiliary_loss_mlp": 0.0102844,
      "balance_loss_clip": 1.05026555,
      "balance_loss_mlp": 1.02057242,
      "epoch": 0.5782480610833884,
      "flos": 16726831856640.0,
      "grad_norm": 2.9820301336252872,
      "language_loss": 0.82881004,
      "learning_rate": 1.593121133898494e-06,
      "loss": 0.85088074,
      "num_input_tokens_seen": 103765300,
      "step": 4809,
      "time_per_iteration": 2.6404664516448975
    },
    {
      "auxiliary_loss_clip": 0.0117144,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.05091047,
      "balance_loss_mlp": 1.0213598,
      "epoch": 0.5783683039740275,
      "flos": 25482140144640.0,
      "grad_norm": 3.915205444579454,
      "language_loss": 0.7931546,
      "learning_rate": 1.592358483351016e-06,
      "loss": 0.8151632,
      "num_input_tokens_seen": 103785475,
      "step": 4810,
      "time_per_iteration": 2.7308287620544434
    },
    {
      "auxiliary_loss_clip": 0.01161339,
      "auxiliary_loss_mlp": 0.0102376,
      "balance_loss_clip": 1.04904461,
      "balance_loss_mlp": 1.01611841,
      "epoch": 0.5784885468646667,
      "flos": 18405835240320.0,
      "grad_norm": 1.8673580119218698,
      "language_loss": 0.72515088,
      "learning_rate": 1.5915958946424326e-06,
      "loss": 0.74700189,
      "num_input_tokens_seen": 103804160,
      "step": 4811,
      "time_per_iteration": 2.646707773208618
    },
    {
      "auxiliary_loss_clip": 0.01137858,
      "auxiliary_loss_mlp": 0.00763696,
      "balance_loss_clip": 1.04701853,
      "balance_loss_mlp": 1.00031853,
      "epoch": 0.5786087897553057,
      "flos": 46100717936640.0,
      "grad_norm": 1.7628471907730814,
      "language_loss": 0.74400818,
      "learning_rate": 1.5908333678884271e-06,
      "loss": 0.76302373,
      "num_input_tokens_seen": 103830580,
      "step": 4812,
      "time_per_iteration": 2.9720335006713867
    },
    {
      "auxiliary_loss_clip": 0.01165115,
      "auxiliary_loss_mlp": 0.01031334,
      "balance_loss_clip": 1.05095243,
      "balance_loss_mlp": 1.02342486,
      "epoch": 0.5787290326459448,
      "flos": 12385950261120.0,
      "grad_norm": 1.8009008410000176,
      "language_loss": 0.74025285,
      "learning_rate": 1.5900709032046743e-06,
      "loss": 0.7622174,
      "num_input_tokens_seen": 103848655,
      "step": 4813,
      "time_per_iteration": 2.715083122253418
    },
    {
      "auxiliary_loss_clip": 0.01147256,
      "auxiliary_loss_mlp": 0.01028773,
      "balance_loss_clip": 1.05154037,
      "balance_loss_mlp": 1.0210247,
      "epoch": 0.5788492755365839,
      "flos": 23290332243840.0,
      "grad_norm": 2.399356385328847,
      "language_loss": 0.7821824,
      "learning_rate": 1.5893085007068391e-06,
      "loss": 0.80394268,
      "num_input_tokens_seen": 103866215,
      "step": 4814,
      "time_per_iteration": 2.7025232315063477
    },
    {
      "auxiliary_loss_clip": 0.01136455,
      "auxiliary_loss_mlp": 0.01028521,
      "balance_loss_clip": 1.04385185,
      "balance_loss_mlp": 1.02038479,
      "epoch": 0.578969518427223,
      "flos": 24061047390720.0,
      "grad_norm": 1.9928238413938981,
      "language_loss": 0.71054566,
      "learning_rate": 1.5885461605105786e-06,
      "loss": 0.7321955,
      "num_input_tokens_seen": 103887815,
      "step": 4815,
      "time_per_iteration": 2.8755176067352295
    },
    {
      "auxiliary_loss_clip": 0.01152273,
      "auxiliary_loss_mlp": 0.01024204,
      "balance_loss_clip": 1.04823208,
      "balance_loss_mlp": 1.0157994,
      "epoch": 0.579089761317862,
      "flos": 21871825269120.0,
      "grad_norm": 1.8462444993773925,
      "language_loss": 0.76932299,
      "learning_rate": 1.5877838827315375e-06,
      "loss": 0.79108781,
      "num_input_tokens_seen": 103906360,
      "step": 4816,
      "time_per_iteration": 2.744380474090576
    },
    {
      "auxiliary_loss_clip": 0.0117663,
      "auxiliary_loss_mlp": 0.01027976,
      "balance_loss_clip": 1.05016804,
      "balance_loss_mlp": 1.01953578,
      "epoch": 0.5792100042085012,
      "flos": 22929681738240.0,
      "grad_norm": 2.745596647260364,
      "language_loss": 0.70155478,
      "learning_rate": 1.587021667485355e-06,
      "loss": 0.72360086,
      "num_input_tokens_seen": 103925730,
      "step": 4817,
      "time_per_iteration": 2.7295403480529785
    },
    {
      "auxiliary_loss_clip": 0.01150966,
      "auxiliary_loss_mlp": 0.01029794,
      "balance_loss_clip": 1.04439664,
      "balance_loss_mlp": 1.02252245,
      "epoch": 0.5793302470991403,
      "flos": 21470056669440.0,
      "grad_norm": 1.7999477665836339,
      "language_loss": 0.7851429,
      "learning_rate": 1.5862595148876559e-06,
      "loss": 0.80695045,
      "num_input_tokens_seen": 103945835,
      "step": 4818,
      "time_per_iteration": 2.769815444946289
    },
    {
      "auxiliary_loss_clip": 0.01120777,
      "auxiliary_loss_mlp": 0.01026441,
      "balance_loss_clip": 1.04301167,
      "balance_loss_mlp": 1.01849258,
      "epoch": 0.5794504899897793,
      "flos": 12711013367040.0,
      "grad_norm": 2.1494615963444113,
      "language_loss": 0.76276273,
      "learning_rate": 1.58549742505406e-06,
      "loss": 0.78423488,
      "num_input_tokens_seen": 103960580,
      "step": 4819,
      "time_per_iteration": 2.751699924468994
    },
    {
      "auxiliary_loss_clip": 0.01178647,
      "auxiliary_loss_mlp": 0.01027829,
      "balance_loss_clip": 1.05050313,
      "balance_loss_mlp": 1.01969278,
      "epoch": 0.5795707328804185,
      "flos": 14867054300160.0,
      "grad_norm": 2.1074461119388865,
      "language_loss": 0.75822258,
      "learning_rate": 1.5847353981001747e-06,
      "loss": 0.78028733,
      "num_input_tokens_seen": 103977760,
      "step": 4820,
      "time_per_iteration": 2.610646963119507
    },
    {
      "auxiliary_loss_clip": 0.01143103,
      "auxiliary_loss_mlp": 0.01024807,
      "balance_loss_clip": 1.04395545,
      "balance_loss_mlp": 1.01694536,
      "epoch": 0.5796909757710575,
      "flos": 36430046432640.0,
      "grad_norm": 1.8129513760307927,
      "language_loss": 0.69826722,
      "learning_rate": 1.5839734341415993e-06,
      "loss": 0.71994638,
      "num_input_tokens_seen": 103999960,
      "step": 4821,
      "time_per_iteration": 3.7006351947784424
    },
    {
      "auxiliary_loss_clip": 0.01160454,
      "auxiliary_loss_mlp": 0.01032292,
      "balance_loss_clip": 1.05262184,
      "balance_loss_mlp": 1.02488327,
      "epoch": 0.5798112186616966,
      "flos": 23039891642880.0,
      "grad_norm": 1.8559977177210367,
      "language_loss": 0.76569676,
      "learning_rate": 1.5832115332939238e-06,
      "loss": 0.78762424,
      "num_input_tokens_seen": 104018400,
      "step": 4822,
      "time_per_iteration": 4.611956596374512
    },
    {
      "auxiliary_loss_clip": 0.01167165,
      "auxiliary_loss_mlp": 0.01029806,
      "balance_loss_clip": 1.05064332,
      "balance_loss_mlp": 1.02168822,
      "epoch": 0.5799314615523358,
      "flos": 16652604401280.0,
      "grad_norm": 2.245662271165546,
      "language_loss": 0.74971801,
      "learning_rate": 1.5824496956727272e-06,
      "loss": 0.77168775,
      "num_input_tokens_seen": 104035605,
      "step": 4823,
      "time_per_iteration": 2.710479736328125
    },
    {
      "auxiliary_loss_clip": 0.01149858,
      "auxiliary_loss_mlp": 0.01020901,
      "balance_loss_clip": 1.04832518,
      "balance_loss_mlp": 1.0136385,
      "epoch": 0.5800517044429748,
      "flos": 20485673470080.0,
      "grad_norm": 2.1425306122115217,
      "language_loss": 0.73133302,
      "learning_rate": 1.5816879213935797e-06,
      "loss": 0.75304061,
      "num_input_tokens_seen": 104054415,
      "step": 4824,
      "time_per_iteration": 2.694781541824341
    },
    {
      "auxiliary_loss_clip": 0.01160843,
      "auxiliary_loss_mlp": 0.01024574,
      "balance_loss_clip": 1.05021083,
      "balance_loss_mlp": 1.01734662,
      "epoch": 0.5801719473336139,
      "flos": 31538258968320.0,
      "grad_norm": 1.510273454970695,
      "language_loss": 0.79766512,
      "learning_rate": 1.5809262105720416e-06,
      "loss": 0.81951928,
      "num_input_tokens_seen": 104075455,
      "step": 4825,
      "time_per_iteration": 3.609889030456543
    },
    {
      "auxiliary_loss_clip": 0.01174474,
      "auxiliary_loss_mlp": 0.01024978,
      "balance_loss_clip": 1.04920864,
      "balance_loss_mlp": 1.0177542,
      "epoch": 0.580292190224253,
      "flos": 20375966355840.0,
      "grad_norm": 1.562397784236009,
      "language_loss": 0.79597008,
      "learning_rate": 1.5801645633236644e-06,
      "loss": 0.81796455,
      "num_input_tokens_seen": 104096440,
      "step": 4826,
      "time_per_iteration": 2.6447839736938477
    },
    {
      "auxiliary_loss_clip": 0.01141881,
      "auxiliary_loss_mlp": 0.01027988,
      "balance_loss_clip": 1.04517055,
      "balance_loss_mlp": 1.02096033,
      "epoch": 0.5804124331148921,
      "flos": 26615373304320.0,
      "grad_norm": 3.177937797559018,
      "language_loss": 0.77271628,
      "learning_rate": 1.579402979763989e-06,
      "loss": 0.794415,
      "num_input_tokens_seen": 104116775,
      "step": 4827,
      "time_per_iteration": 2.6867330074310303
    },
    {
      "auxiliary_loss_clip": 0.0112008,
      "auxiliary_loss_mlp": 0.01027708,
      "balance_loss_clip": 1.04607582,
      "balance_loss_mlp": 1.01976299,
      "epoch": 0.5805326760055312,
      "flos": 13478496289920.0,
      "grad_norm": 2.1589701146353546,
      "language_loss": 0.80955815,
      "learning_rate": 1.578641460008548e-06,
      "loss": 0.83103609,
      "num_input_tokens_seen": 104134510,
      "step": 4828,
      "time_per_iteration": 2.8279311656951904
    },
    {
      "auxiliary_loss_clip": 0.01161144,
      "auxiliary_loss_mlp": 0.01023436,
      "balance_loss_clip": 1.05047703,
      "balance_loss_mlp": 1.01594353,
      "epoch": 0.5806529188961702,
      "flos": 12091374823680.0,
      "grad_norm": 1.9776188956979768,
      "language_loss": 0.68005401,
      "learning_rate": 1.5778800041728613e-06,
      "loss": 0.70189977,
      "num_input_tokens_seen": 104150800,
      "step": 4829,
      "time_per_iteration": 2.6512575149536133
    },
    {
      "auxiliary_loss_clip": 0.01156771,
      "auxiliary_loss_mlp": 0.01028171,
      "balance_loss_clip": 1.04892397,
      "balance_loss_mlp": 1.02069616,
      "epoch": 0.5807731617868094,
      "flos": 26214107495040.0,
      "grad_norm": 1.5685985644723586,
      "language_loss": 0.66465056,
      "learning_rate": 1.577118612372443e-06,
      "loss": 0.68649995,
      "num_input_tokens_seen": 104172640,
      "step": 4830,
      "time_per_iteration": 2.7201194763183594
    },
    {
      "auxiliary_loss_clip": 0.01142599,
      "auxiliary_loss_mlp": 0.00763781,
      "balance_loss_clip": 1.04394054,
      "balance_loss_mlp": 1.00036252,
      "epoch": 0.5808934046774484,
      "flos": 37962139190400.0,
      "grad_norm": 1.950952780857524,
      "language_loss": 0.70375872,
      "learning_rate": 1.5763572847227943e-06,
      "loss": 0.72282243,
      "num_input_tokens_seen": 104193525,
      "step": 4831,
      "time_per_iteration": 2.8177454471588135
    },
    {
      "auxiliary_loss_clip": 0.01161378,
      "auxiliary_loss_mlp": 0.01026498,
      "balance_loss_clip": 1.04649568,
      "balance_loss_mlp": 1.01909494,
      "epoch": 0.5810136475680875,
      "flos": 20485853038080.0,
      "grad_norm": 1.7001054571471879,
      "language_loss": 0.81632513,
      "learning_rate": 1.5755960213394091e-06,
      "loss": 0.83820391,
      "num_input_tokens_seen": 104210625,
      "step": 4832,
      "time_per_iteration": 2.6602303981781006
    },
    {
      "auxiliary_loss_clip": 0.01140388,
      "auxiliary_loss_mlp": 0.01027765,
      "balance_loss_clip": 1.04698825,
      "balance_loss_mlp": 1.01990306,
      "epoch": 0.5811338904587267,
      "flos": 17530153574400.0,
      "grad_norm": 1.857175395675574,
      "language_loss": 0.7815336,
      "learning_rate": 1.5748348223377703e-06,
      "loss": 0.80321515,
      "num_input_tokens_seen": 104228180,
      "step": 4833,
      "time_per_iteration": 2.7245676517486572
    },
    {
      "auxiliary_loss_clip": 0.01150114,
      "auxiliary_loss_mlp": 0.01029238,
      "balance_loss_clip": 1.05032301,
      "balance_loss_mlp": 1.02169204,
      "epoch": 0.5812541333493657,
      "flos": 19458017360640.0,
      "grad_norm": 1.831640634637841,
      "language_loss": 0.77856165,
      "learning_rate": 1.5740736878333507e-06,
      "loss": 0.80035514,
      "num_input_tokens_seen": 104246020,
      "step": 4834,
      "time_per_iteration": 2.700939178466797
    },
    {
      "auxiliary_loss_clip": 0.01153504,
      "auxiliary_loss_mlp": 0.01024923,
      "balance_loss_clip": 1.0467515,
      "balance_loss_mlp": 1.01710916,
      "epoch": 0.5813743762400048,
      "flos": 20594949621120.0,
      "grad_norm": 4.046233416423454,
      "language_loss": 0.78068811,
      "learning_rate": 1.5733126179416143e-06,
      "loss": 0.80247241,
      "num_input_tokens_seen": 104260505,
      "step": 4835,
      "time_per_iteration": 2.6953349113464355
    },
    {
      "auxiliary_loss_clip": 0.01161623,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 1.04966497,
      "balance_loss_mlp": 1.01857138,
      "epoch": 0.5814946191306439,
      "flos": 33178227246720.0,
      "grad_norm": 2.305735061835053,
      "language_loss": 0.72547507,
      "learning_rate": 1.5725516127780137e-06,
      "loss": 0.74735475,
      "num_input_tokens_seen": 104282640,
      "step": 4836,
      "time_per_iteration": 2.7221062183380127
    },
    {
      "auxiliary_loss_clip": 0.01171033,
      "auxiliary_loss_mlp": 0.01028576,
      "balance_loss_clip": 1.04948449,
      "balance_loss_mlp": 1.02045822,
      "epoch": 0.581614862021283,
      "flos": 16143283503360.0,
      "grad_norm": 2.2501332324962293,
      "language_loss": 0.88714397,
      "learning_rate": 1.5717906724579943e-06,
      "loss": 0.90913999,
      "num_input_tokens_seen": 104299700,
      "step": 4837,
      "time_per_iteration": 2.6497535705566406
    },
    {
      "auxiliary_loss_clip": 0.01142625,
      "auxiliary_loss_mlp": 0.0103001,
      "balance_loss_clip": 1.04559803,
      "balance_loss_mlp": 1.02189779,
      "epoch": 0.581735104911922,
      "flos": 33802642298880.0,
      "grad_norm": 2.721232335176114,
      "language_loss": 0.67619133,
      "learning_rate": 1.571029797096989e-06,
      "loss": 0.6979177,
      "num_input_tokens_seen": 104320805,
      "step": 4838,
      "time_per_iteration": 2.866564989089966
    },
    {
      "auxiliary_loss_clip": 0.01175611,
      "auxiliary_loss_mlp": 0.01023148,
      "balance_loss_clip": 1.05003917,
      "balance_loss_mlp": 1.01585293,
      "epoch": 0.5818553478025612,
      "flos": 23331163029120.0,
      "grad_norm": 1.755628989243359,
      "language_loss": 0.78863204,
      "learning_rate": 1.570268986810423e-06,
      "loss": 0.81061971,
      "num_input_tokens_seen": 104340700,
      "step": 4839,
      "time_per_iteration": 2.800551176071167
    },
    {
      "auxiliary_loss_clip": 0.01146608,
      "auxiliary_loss_mlp": 0.01025244,
      "balance_loss_clip": 1.04734325,
      "balance_loss_mlp": 1.01801407,
      "epoch": 0.5819755906932003,
      "flos": 20996143603200.0,
      "grad_norm": 1.941383028692232,
      "language_loss": 0.74885219,
      "learning_rate": 1.5695082417137096e-06,
      "loss": 0.7705707,
      "num_input_tokens_seen": 104358575,
      "step": 4840,
      "time_per_iteration": 2.6957130432128906
    },
    {
      "auxiliary_loss_clip": 0.01144875,
      "auxiliary_loss_mlp": 0.01022881,
      "balance_loss_clip": 1.04437149,
      "balance_loss_mlp": 1.01602316,
      "epoch": 0.5820958335838393,
      "flos": 21431668008960.0,
      "grad_norm": 2.0925431665375447,
      "language_loss": 0.75186765,
      "learning_rate": 1.5687475619222539e-06,
      "loss": 0.77354521,
      "num_input_tokens_seen": 104378530,
      "step": 4841,
      "time_per_iteration": 2.75512957572937
    },
    {
      "auxiliary_loss_clip": 0.01143064,
      "auxiliary_loss_mlp": 0.01028599,
      "balance_loss_clip": 1.04422164,
      "balance_loss_mlp": 1.02014995,
      "epoch": 0.5822160764744785,
      "flos": 17967473660160.0,
      "grad_norm": 2.292013531357256,
      "language_loss": 0.73779178,
      "learning_rate": 1.5679869475514496e-06,
      "loss": 0.75950837,
      "num_input_tokens_seen": 104395465,
      "step": 4842,
      "time_per_iteration": 2.64642596244812
    },
    {
      "auxiliary_loss_clip": 0.01162252,
      "auxiliary_loss_mlp": 0.01020357,
      "balance_loss_clip": 1.04755747,
      "balance_loss_mlp": 1.01215529,
      "epoch": 0.5823363193651175,
      "flos": 23033858158080.0,
      "grad_norm": 2.1018675092715027,
      "language_loss": 0.81454873,
      "learning_rate": 1.567226398716682e-06,
      "loss": 0.83637482,
      "num_input_tokens_seen": 104415380,
      "step": 4843,
      "time_per_iteration": 2.6630876064300537
    },
    {
      "auxiliary_loss_clip": 0.0115735,
      "auxiliary_loss_mlp": 0.01022976,
      "balance_loss_clip": 1.04851866,
      "balance_loss_mlp": 1.01489961,
      "epoch": 0.5824565622557566,
      "flos": 32891840110080.0,
      "grad_norm": 1.5817796973552598,
      "language_loss": 0.61647075,
      "learning_rate": 1.566465915533326e-06,
      "loss": 0.63827395,
      "num_input_tokens_seen": 104437410,
      "step": 4844,
      "time_per_iteration": 2.7603211402893066
    },
    {
      "auxiliary_loss_clip": 0.01162473,
      "auxiliary_loss_mlp": 0.01026621,
      "balance_loss_clip": 1.04945803,
      "balance_loss_mlp": 1.01884818,
      "epoch": 0.5825768051463958,
      "flos": 22229674513920.0,
      "grad_norm": 2.308391537125453,
      "language_loss": 0.88281685,
      "learning_rate": 1.5657054981167458e-06,
      "loss": 0.90470773,
      "num_input_tokens_seen": 104456305,
      "step": 4845,
      "time_per_iteration": 2.638922929763794
    },
    {
      "auxiliary_loss_clip": 0.01159996,
      "auxiliary_loss_mlp": 0.01026409,
      "balance_loss_clip": 1.04828835,
      "balance_loss_mlp": 1.01936722,
      "epoch": 0.5826970480370348,
      "flos": 28001561016960.0,
      "grad_norm": 2.3806884686418597,
      "language_loss": 0.67396128,
      "learning_rate": 1.5649451465822965e-06,
      "loss": 0.69582534,
      "num_input_tokens_seen": 104477695,
      "step": 4846,
      "time_per_iteration": 3.6124820709228516
    },
    {
      "auxiliary_loss_clip": 0.01120036,
      "auxiliary_loss_mlp": 0.01027694,
      "balance_loss_clip": 1.04741037,
      "balance_loss_mlp": 1.02040982,
      "epoch": 0.5828172909276739,
      "flos": 17858053854720.0,
      "grad_norm": 1.67778973283529,
      "language_loss": 0.83771163,
      "learning_rate": 1.5641848610453218e-06,
      "loss": 0.85918891,
      "num_input_tokens_seen": 104496355,
      "step": 4847,
      "time_per_iteration": 2.720186948776245
    },
    {
      "auxiliary_loss_clip": 0.01162565,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.05101657,
      "balance_loss_mlp": 1.0220685,
      "epoch": 0.582937533818313,
      "flos": 19865244827520.0,
      "grad_norm": 2.0890814243998106,
      "language_loss": 0.86056793,
      "learning_rate": 1.563424641621158e-06,
      "loss": 0.88249439,
      "num_input_tokens_seen": 104515535,
      "step": 4848,
      "time_per_iteration": 3.6739563941955566
    },
    {
      "auxiliary_loss_clip": 0.01151093,
      "auxiliary_loss_mlp": 0.01022186,
      "balance_loss_clip": 1.0477705,
      "balance_loss_mlp": 1.01428819,
      "epoch": 0.5830577767089521,
      "flos": 26870734068480.0,
      "grad_norm": 2.0946185914631186,
      "language_loss": 0.6994496,
      "learning_rate": 1.5626644884251282e-06,
      "loss": 0.72118235,
      "num_input_tokens_seen": 104535055,
      "step": 4849,
      "time_per_iteration": 2.759951114654541
    },
    {
      "auxiliary_loss_clip": 0.01174509,
      "auxiliary_loss_mlp": 0.01029545,
      "balance_loss_clip": 1.04913974,
      "balance_loss_mlp": 1.02189779,
      "epoch": 0.5831780195995911,
      "flos": 25298205575040.0,
      "grad_norm": 1.6862776510407782,
      "language_loss": 0.88074148,
      "learning_rate": 1.5619044015725488e-06,
      "loss": 0.90278208,
      "num_input_tokens_seen": 104554745,
      "step": 4850,
      "time_per_iteration": 2.6001181602478027
    },
    {
      "auxiliary_loss_clip": 0.0118552,
      "auxiliary_loss_mlp": 0.01029903,
      "balance_loss_clip": 1.05467308,
      "balance_loss_mlp": 1.02113557,
      "epoch": 0.5832982624902303,
      "flos": 14756988049920.0,
      "grad_norm": 2.20730254942022,
      "language_loss": 0.86969006,
      "learning_rate": 1.5611443811787224e-06,
      "loss": 0.89184439,
      "num_input_tokens_seen": 104568870,
      "step": 4851,
      "time_per_iteration": 3.550734758377075
    },
    {
      "auxiliary_loss_clip": 0.01159454,
      "auxiliary_loss_mlp": 0.01023568,
      "balance_loss_clip": 1.05040145,
      "balance_loss_mlp": 1.01651645,
      "epoch": 0.5834185053808694,
      "flos": 20444555376000.0,
      "grad_norm": 2.265616943615762,
      "language_loss": 0.69282818,
      "learning_rate": 1.560384427358945e-06,
      "loss": 0.71465838,
      "num_input_tokens_seen": 104588415,
      "step": 4852,
      "time_per_iteration": 2.6184380054473877
    },
    {
      "auxiliary_loss_clip": 0.01142898,
      "auxiliary_loss_mlp": 0.01024222,
      "balance_loss_clip": 1.04310727,
      "balance_loss_mlp": 1.01637232,
      "epoch": 0.5835387482715084,
      "flos": 27200394115200.0,
      "grad_norm": 1.622122838907256,
      "language_loss": 0.73052275,
      "learning_rate": 1.5596245402284998e-06,
      "loss": 0.75219399,
      "num_input_tokens_seen": 104611940,
      "step": 4853,
      "time_per_iteration": 2.73319935798645
    },
    {
      "auxiliary_loss_clip": 0.01166556,
      "auxiliary_loss_mlp": 0.01030183,
      "balance_loss_clip": 1.05300987,
      "balance_loss_mlp": 1.02269113,
      "epoch": 0.5836589911621476,
      "flos": 16654615562880.0,
      "grad_norm": 1.7878507873432576,
      "language_loss": 0.82146072,
      "learning_rate": 1.5588647199026619e-06,
      "loss": 0.84342813,
      "num_input_tokens_seen": 104629675,
      "step": 4854,
      "time_per_iteration": 2.618955612182617
    },
    {
      "auxiliary_loss_clip": 0.01182003,
      "auxiliary_loss_mlp": 0.01027153,
      "balance_loss_clip": 1.05317163,
      "balance_loss_mlp": 1.01904082,
      "epoch": 0.5837792340527866,
      "flos": 20446817932800.0,
      "grad_norm": 2.002784173198503,
      "language_loss": 0.87576073,
      "learning_rate": 1.5581049664966956e-06,
      "loss": 0.8978523,
      "num_input_tokens_seen": 104647435,
      "step": 4855,
      "time_per_iteration": 2.633305311203003
    },
    {
      "auxiliary_loss_clip": 0.01022941,
      "auxiliary_loss_mlp": 0.01001231,
      "balance_loss_clip": 1.01093793,
      "balance_loss_mlp": 0.99984771,
      "epoch": 0.5838994769434257,
      "flos": 65995480765440.0,
      "grad_norm": 2.189559679819557,
      "language_loss": 0.65154046,
      "learning_rate": 1.5573452801258545e-06,
      "loss": 0.6717822,
      "num_input_tokens_seen": 104694605,
      "step": 4856,
      "time_per_iteration": 3.1412153244018555
    },
    {
      "auxiliary_loss_clip": 0.01170113,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 1.05144763,
      "balance_loss_mlp": 1.02102602,
      "epoch": 0.5840197198340649,
      "flos": 21470523546240.0,
      "grad_norm": 1.933447614807768,
      "language_loss": 0.63370502,
      "learning_rate": 1.5565856609053824e-06,
      "loss": 0.65569317,
      "num_input_tokens_seen": 104713400,
      "step": 4857,
      "time_per_iteration": 2.6318531036376953
    },
    {
      "auxiliary_loss_clip": 0.01180361,
      "auxiliary_loss_mlp": 0.01028871,
      "balance_loss_clip": 1.05309498,
      "balance_loss_mlp": 1.02071738,
      "epoch": 0.5841399627247039,
      "flos": 19135144984320.0,
      "grad_norm": 1.8286063410427205,
      "language_loss": 0.80105531,
      "learning_rate": 1.5558261089505127e-06,
      "loss": 0.82314765,
      "num_input_tokens_seen": 104732130,
      "step": 4858,
      "time_per_iteration": 2.612337112426758
    },
    {
      "auxiliary_loss_clip": 0.01164647,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 1.05064762,
      "balance_loss_mlp": 1.02037287,
      "epoch": 0.584260205615343,
      "flos": 26425692558720.0,
      "grad_norm": 2.1609520629328665,
      "language_loss": 0.80105972,
      "learning_rate": 1.5550666243764697e-06,
      "loss": 0.82299078,
      "num_input_tokens_seen": 104750290,
      "step": 4859,
      "time_per_iteration": 2.6635963916778564
    },
    {
      "auxiliary_loss_clip": 0.01163559,
      "auxiliary_loss_mlp": 0.01026781,
      "balance_loss_clip": 1.04901218,
      "balance_loss_mlp": 1.01965857,
      "epoch": 0.584380448505982,
      "flos": 13881809174400.0,
      "grad_norm": 1.927599509881764,
      "language_loss": 0.76820534,
      "learning_rate": 1.554307207298465e-06,
      "loss": 0.79010874,
      "num_input_tokens_seen": 104768550,
      "step": 4860,
      "time_per_iteration": 2.7117021083831787
    },
    {
      "auxiliary_loss_clip": 0.01183015,
      "auxiliary_loss_mlp": 0.01028546,
      "balance_loss_clip": 1.05350721,
      "balance_loss_mlp": 1.02046072,
      "epoch": 0.5845006913966212,
      "flos": 21543709507200.0,
      "grad_norm": 1.8958881826774852,
      "language_loss": 0.78429496,
      "learning_rate": 1.553547857831704e-06,
      "loss": 0.80641055,
      "num_input_tokens_seen": 104785060,
      "step": 4861,
      "time_per_iteration": 2.5912201404571533
    },
    {
      "auxiliary_loss_clip": 0.01073935,
      "auxiliary_loss_mlp": 0.01002639,
      "balance_loss_clip": 1.01305985,
      "balance_loss_mlp": 1.00107753,
      "epoch": 0.5846209342872603,
      "flos": 58375452712320.0,
      "grad_norm": 0.8801084609693695,
      "language_loss": 0.64158285,
      "learning_rate": 1.5527885760913771e-06,
      "loss": 0.66234857,
      "num_input_tokens_seen": 104834950,
      "step": 4862,
      "time_per_iteration": 3.0212576389312744
    },
    {
      "auxiliary_loss_clip": 0.01147365,
      "auxiliary_loss_mlp": 0.01023602,
      "balance_loss_clip": 1.04914606,
      "balance_loss_mlp": 1.01701832,
      "epoch": 0.5847411771778993,
      "flos": 18588045957120.0,
      "grad_norm": 1.7950834296259093,
      "language_loss": 0.76532221,
      "learning_rate": 1.552029362192668e-06,
      "loss": 0.78703189,
      "num_input_tokens_seen": 104854210,
      "step": 4863,
      "time_per_iteration": 2.644284248352051
    },
    {
      "auxiliary_loss_clip": 0.01130239,
      "auxiliary_loss_mlp": 0.0102612,
      "balance_loss_clip": 1.04676723,
      "balance_loss_mlp": 1.01803434,
      "epoch": 0.5848614200685385,
      "flos": 24240780069120.0,
      "grad_norm": 8.084568101077373,
      "language_loss": 0.72532213,
      "learning_rate": 1.5512702162507478e-06,
      "loss": 0.74688578,
      "num_input_tokens_seen": 104874525,
      "step": 4864,
      "time_per_iteration": 2.798851251602173
    },
    {
      "auxiliary_loss_clip": 0.01052198,
      "auxiliary_loss_mlp": 0.01001656,
      "balance_loss_clip": 1.01240718,
      "balance_loss_mlp": 1.00008285,
      "epoch": 0.5849816629591775,
      "flos": 71660245933440.0,
      "grad_norm": 1.1001091077984224,
      "language_loss": 0.55748677,
      "learning_rate": 1.5505111383807792e-06,
      "loss": 0.57802528,
      "num_input_tokens_seen": 104937195,
      "step": 4865,
      "time_per_iteration": 3.295311689376831
    },
    {
      "auxiliary_loss_clip": 0.01123318,
      "auxiliary_loss_mlp": 0.01027844,
      "balance_loss_clip": 1.04301214,
      "balance_loss_mlp": 1.01995802,
      "epoch": 0.5851019058498166,
      "flos": 23802095266560.0,
      "grad_norm": 1.8713735851802469,
      "language_loss": 0.80758423,
      "learning_rate": 1.5497521286979138e-06,
      "loss": 0.82909584,
      "num_input_tokens_seen": 104957435,
      "step": 4866,
      "time_per_iteration": 2.743245840072632
    },
    {
      "auxiliary_loss_clip": 0.01138018,
      "auxiliary_loss_mlp": 0.01031754,
      "balance_loss_clip": 1.04480863,
      "balance_loss_mlp": 1.023368,
      "epoch": 0.5852221487404557,
      "flos": 24388516707840.0,
      "grad_norm": 2.268170688300956,
      "language_loss": 0.74260116,
      "learning_rate": 1.5489931873172927e-06,
      "loss": 0.76429886,
      "num_input_tokens_seen": 104978755,
      "step": 4867,
      "time_per_iteration": 2.7442164421081543
    },
    {
      "auxiliary_loss_clip": 0.01089365,
      "auxiliary_loss_mlp": 0.01033475,
      "balance_loss_clip": 1.03775907,
      "balance_loss_mlp": 1.0260725,
      "epoch": 0.5853423916310948,
      "flos": 27271425260160.0,
      "grad_norm": 1.902790732696549,
      "language_loss": 0.79090607,
      "learning_rate": 1.5482343143540467e-06,
      "loss": 0.81213444,
      "num_input_tokens_seen": 105000020,
      "step": 4868,
      "time_per_iteration": 2.8656864166259766
    },
    {
      "auxiliary_loss_clip": 0.01137375,
      "auxiliary_loss_mlp": 0.00762379,
      "balance_loss_clip": 1.04541636,
      "balance_loss_mlp": 1.00024939,
      "epoch": 0.5854626345217339,
      "flos": 11983786611840.0,
      "grad_norm": 2.9349628190071324,
      "language_loss": 0.83326346,
      "learning_rate": 1.547475509923295e-06,
      "loss": 0.85226101,
      "num_input_tokens_seen": 105017060,
      "step": 4869,
      "time_per_iteration": 2.733159303665161
    },
    {
      "auxiliary_loss_clip": 0.01032039,
      "auxiliary_loss_mlp": 0.0100089,
      "balance_loss_clip": 1.01062,
      "balance_loss_mlp": 0.99932796,
      "epoch": 0.585582877412373,
      "flos": 64342335173760.0,
      "grad_norm": 0.7375390063164634,
      "language_loss": 0.56005239,
      "learning_rate": 1.5467167741401495e-06,
      "loss": 0.58038163,
      "num_input_tokens_seen": 105078540,
      "step": 4870,
      "time_per_iteration": 3.279709815979004
    },
    {
      "auxiliary_loss_clip": 0.01147132,
      "auxiliary_loss_mlp": 0.01029377,
      "balance_loss_clip": 1.04452181,
      "balance_loss_mlp": 1.02158093,
      "epoch": 0.5857031203030121,
      "flos": 17011926103680.0,
      "grad_norm": 2.396602183375033,
      "language_loss": 0.71902311,
      "learning_rate": 1.5459581071197083e-06,
      "loss": 0.74078828,
      "num_input_tokens_seen": 105094200,
      "step": 4871,
      "time_per_iteration": 2.7566263675689697
    },
    {
      "auxiliary_loss_clip": 0.01169157,
      "auxiliary_loss_mlp": 0.01027255,
      "balance_loss_clip": 1.05324793,
      "balance_loss_mlp": 1.01958358,
      "epoch": 0.5858233631936511,
      "flos": 20885682303360.0,
      "grad_norm": 2.048398434874254,
      "language_loss": 0.83093679,
      "learning_rate": 1.5451995089770624e-06,
      "loss": 0.85290086,
      "num_input_tokens_seen": 105113985,
      "step": 4872,
      "time_per_iteration": 2.677126169204712
    },
    {
      "auxiliary_loss_clip": 0.01175727,
      "auxiliary_loss_mlp": 0.01025464,
      "balance_loss_clip": 1.05097032,
      "balance_loss_mlp": 1.01841331,
      "epoch": 0.5859436060842903,
      "flos": 23191902000000.0,
      "grad_norm": 1.3783326311436244,
      "language_loss": 0.71918356,
      "learning_rate": 1.5444409798272885e-06,
      "loss": 0.74119544,
      "num_input_tokens_seen": 105138075,
      "step": 4873,
      "time_per_iteration": 3.6415350437164307
    },
    {
      "auxiliary_loss_clip": 0.01134155,
      "auxiliary_loss_mlp": 0.01033337,
      "balance_loss_clip": 1.04340005,
      "balance_loss_mlp": 1.02503371,
      "epoch": 0.5860638489749294,
      "flos": 22492648961280.0,
      "grad_norm": 2.2647533303839147,
      "language_loss": 0.81057906,
      "learning_rate": 1.543682519785456e-06,
      "loss": 0.83225399,
      "num_input_tokens_seen": 105156555,
      "step": 4874,
      "time_per_iteration": 4.581416606903076
    },
    {
      "auxiliary_loss_clip": 0.01147777,
      "auxiliary_loss_mlp": 0.01027988,
      "balance_loss_clip": 1.04628372,
      "balance_loss_mlp": 1.02074003,
      "epoch": 0.5861840918655684,
      "flos": 17566243764480.0,
      "grad_norm": 2.41548406527139,
      "language_loss": 0.80605137,
      "learning_rate": 1.5429241289666219e-06,
      "loss": 0.82780904,
      "num_input_tokens_seen": 105174055,
      "step": 4875,
      "time_per_iteration": 2.7073020935058594
    },
    {
      "auxiliary_loss_clip": 0.01141776,
      "auxiliary_loss_mlp": 0.01023555,
      "balance_loss_clip": 1.04656863,
      "balance_loss_mlp": 1.01615238,
      "epoch": 0.5863043347562076,
      "flos": 25556152118400.0,
      "grad_norm": 2.1651528219474248,
      "language_loss": 0.69708472,
      "learning_rate": 1.5421658074858342e-06,
      "loss": 0.71873808,
      "num_input_tokens_seen": 105192160,
      "step": 4876,
      "time_per_iteration": 3.641961097717285
    },
    {
      "auxiliary_loss_clip": 0.01144479,
      "auxiliary_loss_mlp": 0.01032734,
      "balance_loss_clip": 1.04801714,
      "balance_loss_mlp": 1.02463961,
      "epoch": 0.5864245776468466,
      "flos": 20667525050880.0,
      "grad_norm": 2.294436141596231,
      "language_loss": 0.66547793,
      "learning_rate": 1.5414075554581298e-06,
      "loss": 0.68725002,
      "num_input_tokens_seen": 105210205,
      "step": 4877,
      "time_per_iteration": 2.721721887588501
    },
    {
      "auxiliary_loss_clip": 0.0118009,
      "auxiliary_loss_mlp": 0.01030779,
      "balance_loss_clip": 1.05042291,
      "balance_loss_mlp": 1.02305412,
      "epoch": 0.5865448205374857,
      "flos": 28913907490560.0,
      "grad_norm": 3.094594771709935,
      "language_loss": 0.78146303,
      "learning_rate": 1.5406493729985348e-06,
      "loss": 0.8035717,
      "num_input_tokens_seen": 105229400,
      "step": 4878,
      "time_per_iteration": 2.736154317855835
    },
    {
      "auxiliary_loss_clip": 0.0112676,
      "auxiliary_loss_mlp": 0.00762911,
      "balance_loss_clip": 1.04672551,
      "balance_loss_mlp": 1.00028408,
      "epoch": 0.5866650634281249,
      "flos": 25842575168640.0,
      "grad_norm": 2.258639984969489,
      "language_loss": 0.71790022,
      "learning_rate": 1.5398912602220644e-06,
      "loss": 0.73679698,
      "num_input_tokens_seen": 105248675,
      "step": 4879,
      "time_per_iteration": 2.7802629470825195
    },
    {
      "auxiliary_loss_clip": 0.01130757,
      "auxiliary_loss_mlp": 0.01031489,
      "balance_loss_clip": 1.04451823,
      "balance_loss_mlp": 1.02344203,
      "epoch": 0.5867853063187639,
      "flos": 17052325925760.0,
      "grad_norm": 2.0021809142525355,
      "language_loss": 0.78226912,
      "learning_rate": 1.539133217243724e-06,
      "loss": 0.80389154,
      "num_input_tokens_seen": 105265695,
      "step": 4880,
      "time_per_iteration": 2.745915174484253
    },
    {
      "auxiliary_loss_clip": 0.01139778,
      "auxiliary_loss_mlp": 0.01025145,
      "balance_loss_clip": 1.04634535,
      "balance_loss_mlp": 1.01661575,
      "epoch": 0.586905549209403,
      "flos": 24645026707200.0,
      "grad_norm": 2.174076298590697,
      "language_loss": 0.75996447,
      "learning_rate": 1.5383752441785081e-06,
      "loss": 0.78161371,
      "num_input_tokens_seen": 105284920,
      "step": 4881,
      "time_per_iteration": 2.727510690689087
    },
    {
      "auxiliary_loss_clip": 0.01169218,
      "auxiliary_loss_mlp": 0.01029351,
      "balance_loss_clip": 1.0503068,
      "balance_loss_mlp": 1.02147985,
      "epoch": 0.5870257921000421,
      "flos": 14720538723840.0,
      "grad_norm": 2.132414687422191,
      "language_loss": 0.85416007,
      "learning_rate": 1.5376173411414003e-06,
      "loss": 0.87614572,
      "num_input_tokens_seen": 105302960,
      "step": 4882,
      "time_per_iteration": 2.6639392375946045
    },
    {
      "auxiliary_loss_clip": 0.01148191,
      "auxiliary_loss_mlp": 0.01031938,
      "balance_loss_clip": 1.04354846,
      "balance_loss_mlp": 1.02417755,
      "epoch": 0.5871460349906812,
      "flos": 23914998691200.0,
      "grad_norm": 2.3908345344027877,
      "language_loss": 0.78946471,
      "learning_rate": 1.5368595082473753e-06,
      "loss": 0.81126606,
      "num_input_tokens_seen": 105321260,
      "step": 4883,
      "time_per_iteration": 2.722381830215454
    },
    {
      "auxiliary_loss_clip": 0.01166381,
      "auxiliary_loss_mlp": 0.01024678,
      "balance_loss_clip": 1.04916143,
      "balance_loss_mlp": 1.0176034,
      "epoch": 0.5872662778813202,
      "flos": 22164174063360.0,
      "grad_norm": 1.7575592420640564,
      "language_loss": 0.78024781,
      "learning_rate": 1.5361017456113935e-06,
      "loss": 0.80215836,
      "num_input_tokens_seen": 105341610,
      "step": 4884,
      "time_per_iteration": 2.770350933074951
    },
    {
      "auxiliary_loss_clip": 0.01166021,
      "auxiliary_loss_mlp": 0.01028591,
      "balance_loss_clip": 1.04845893,
      "balance_loss_mlp": 1.02069354,
      "epoch": 0.5873865207719594,
      "flos": 18441925430400.0,
      "grad_norm": 2.018502238382439,
      "language_loss": 0.86187297,
      "learning_rate": 1.5353440533484085e-06,
      "loss": 0.8838191,
      "num_input_tokens_seen": 105360465,
      "step": 4885,
      "time_per_iteration": 2.592813491821289
    },
    {
      "auxiliary_loss_clip": 0.0115423,
      "auxiliary_loss_mlp": 0.01034073,
      "balance_loss_clip": 1.04866517,
      "balance_loss_mlp": 1.02588367,
      "epoch": 0.5875067636625985,
      "flos": 54015321427200.0,
      "grad_norm": 1.9171547990166948,
      "language_loss": 0.65921986,
      "learning_rate": 1.534586431573361e-06,
      "loss": 0.68110293,
      "num_input_tokens_seen": 105385405,
      "step": 4886,
      "time_per_iteration": 3.0140206813812256
    },
    {
      "auxiliary_loss_clip": 0.01105396,
      "auxiliary_loss_mlp": 0.01029527,
      "balance_loss_clip": 1.0393337,
      "balance_loss_mlp": 1.02094412,
      "epoch": 0.5876270065532375,
      "flos": 27995707100160.0,
      "grad_norm": 1.8282061836307484,
      "language_loss": 0.79051685,
      "learning_rate": 1.5338288804011817e-06,
      "loss": 0.8118661,
      "num_input_tokens_seen": 105404905,
      "step": 4887,
      "time_per_iteration": 2.9724414348602295
    },
    {
      "auxiliary_loss_clip": 0.01144558,
      "auxiliary_loss_mlp": 0.01030943,
      "balance_loss_clip": 1.04511142,
      "balance_loss_mlp": 1.02282536,
      "epoch": 0.5877472494438767,
      "flos": 21361462876800.0,
      "grad_norm": 1.9772791264624847,
      "language_loss": 0.71108431,
      "learning_rate": 1.533071399946791e-06,
      "loss": 0.73283935,
      "num_input_tokens_seen": 105423650,
      "step": 4888,
      "time_per_iteration": 2.7016940116882324
    },
    {
      "auxiliary_loss_clip": 0.01150393,
      "auxiliary_loss_mlp": 0.01027211,
      "balance_loss_clip": 1.04580975,
      "balance_loss_mlp": 1.0193851,
      "epoch": 0.5878674923345157,
      "flos": 22383013674240.0,
      "grad_norm": 1.9241131294930416,
      "language_loss": 0.57447851,
      "learning_rate": 1.5323139903250977e-06,
      "loss": 0.59625459,
      "num_input_tokens_seen": 105444255,
      "step": 4889,
      "time_per_iteration": 2.770261287689209
    },
    {
      "auxiliary_loss_clip": 0.01153826,
      "auxiliary_loss_mlp": 0.01033139,
      "balance_loss_clip": 1.05153656,
      "balance_loss_mlp": 1.02526546,
      "epoch": 0.5879877352251548,
      "flos": 21868664872320.0,
      "grad_norm": 1.6462654499554887,
      "language_loss": 0.76992106,
      "learning_rate": 1.5315566516510002e-06,
      "loss": 0.79179072,
      "num_input_tokens_seen": 105462425,
      "step": 4890,
      "time_per_iteration": 2.6702401638031006
    },
    {
      "auxiliary_loss_clip": 0.01177036,
      "auxiliary_loss_mlp": 0.01027899,
      "balance_loss_clip": 1.05088675,
      "balance_loss_mlp": 1.01954865,
      "epoch": 0.5881079781157939,
      "flos": 17493811989120.0,
      "grad_norm": 1.740136546501356,
      "language_loss": 0.67610127,
      "learning_rate": 1.5307993840393857e-06,
      "loss": 0.69815058,
      "num_input_tokens_seen": 105480505,
      "step": 4891,
      "time_per_iteration": 2.6186251640319824
    },
    {
      "auxiliary_loss_clip": 0.0117642,
      "auxiliary_loss_mlp": 0.01028234,
      "balance_loss_clip": 1.04962587,
      "balance_loss_mlp": 1.02096796,
      "epoch": 0.588228221006433,
      "flos": 22601853285120.0,
      "grad_norm": 1.8055722014968005,
      "language_loss": 0.8039943,
      "learning_rate": 1.530042187605132e-06,
      "loss": 0.82604086,
      "num_input_tokens_seen": 105499760,
      "step": 4892,
      "time_per_iteration": 2.594836950302124
    },
    {
      "auxiliary_loss_clip": 0.01163768,
      "auxiliary_loss_mlp": 0.00761743,
      "balance_loss_clip": 1.05042768,
      "balance_loss_mlp": 1.00035954,
      "epoch": 0.5883484638970721,
      "flos": 26176939896960.0,
      "grad_norm": 1.4150408690419767,
      "language_loss": 0.84261572,
      "learning_rate": 1.5292850624631044e-06,
      "loss": 0.86187088,
      "num_input_tokens_seen": 105521955,
      "step": 4893,
      "time_per_iteration": 2.7300970554351807
    },
    {
      "auxiliary_loss_clip": 0.01160627,
      "auxiliary_loss_mlp": 0.01028033,
      "balance_loss_clip": 1.05011797,
      "balance_loss_mlp": 1.02040935,
      "epoch": 0.5884687067877111,
      "flos": 30443737691520.0,
      "grad_norm": 2.018345781513594,
      "language_loss": 0.80277693,
      "learning_rate": 1.5285280087281593e-06,
      "loss": 0.82466352,
      "num_input_tokens_seen": 105542685,
      "step": 4894,
      "time_per_iteration": 2.6546895503997803
    },
    {
      "auxiliary_loss_clip": 0.01051908,
      "auxiliary_loss_mlp": 0.01001132,
      "balance_loss_clip": 1.01054955,
      "balance_loss_mlp": 0.99970764,
      "epoch": 0.5885889496783503,
      "flos": 70507550580480.0,
      "grad_norm": 0.6358789448993087,
      "language_loss": 0.56587797,
      "learning_rate": 1.5277710265151398e-06,
      "loss": 0.58640838,
      "num_input_tokens_seen": 105612165,
      "step": 4895,
      "time_per_iteration": 3.4039573669433594
    },
    {
      "auxiliary_loss_clip": 0.01162819,
      "auxiliary_loss_mlp": 0.01032156,
      "balance_loss_clip": 1.048195,
      "balance_loss_mlp": 1.02380586,
      "epoch": 0.5887091925689893,
      "flos": 19098767485440.0,
      "grad_norm": 2.7807579329350807,
      "language_loss": 0.77577746,
      "learning_rate": 1.5270141159388803e-06,
      "loss": 0.79772723,
      "num_input_tokens_seen": 105629185,
      "step": 4896,
      "time_per_iteration": 2.709073305130005
    },
    {
      "auxiliary_loss_clip": 0.01177912,
      "auxiliary_loss_mlp": 0.01027159,
      "balance_loss_clip": 1.04981828,
      "balance_loss_mlp": 1.01954758,
      "epoch": 0.5888294354596284,
      "flos": 23294282739840.0,
      "grad_norm": 1.7935306765304087,
      "language_loss": 0.8040964,
      "learning_rate": 1.526257277114203e-06,
      "loss": 0.82614714,
      "num_input_tokens_seen": 105650260,
      "step": 4897,
      "time_per_iteration": 2.633680820465088
    },
    {
      "auxiliary_loss_clip": 0.01145715,
      "auxiliary_loss_mlp": 0.01027383,
      "balance_loss_clip": 1.04839361,
      "balance_loss_mlp": 1.01958919,
      "epoch": 0.5889496783502676,
      "flos": 21981532383360.0,
      "grad_norm": 2.1112585436742797,
      "language_loss": 0.79186845,
      "learning_rate": 1.5255005101559201e-06,
      "loss": 0.81359947,
      "num_input_tokens_seen": 105667870,
      "step": 4898,
      "time_per_iteration": 2.6852736473083496
    },
    {
      "auxiliary_loss_clip": 0.01165181,
      "auxiliary_loss_mlp": 0.01022731,
      "balance_loss_clip": 1.04886043,
      "balance_loss_mlp": 1.01518834,
      "epoch": 0.5890699212409066,
      "flos": 21685233093120.0,
      "grad_norm": 2.290658708120417,
      "language_loss": 0.76640528,
      "learning_rate": 1.524743815178833e-06,
      "loss": 0.78828442,
      "num_input_tokens_seen": 105685830,
      "step": 4899,
      "time_per_iteration": 3.539773464202881
    },
    {
      "auxiliary_loss_clip": 0.011465,
      "auxiliary_loss_mlp": 0.01030106,
      "balance_loss_clip": 1.04282618,
      "balance_loss_mlp": 1.02295291,
      "epoch": 0.5891901641315457,
      "flos": 19464553635840.0,
      "grad_norm": 2.1326674253242395,
      "language_loss": 0.80868244,
      "learning_rate": 1.5239871922977315e-06,
      "loss": 0.83044851,
      "num_input_tokens_seen": 105705745,
      "step": 4900,
      "time_per_iteration": 4.574876308441162
    },
    {
      "auxiliary_loss_clip": 0.01146709,
      "auxiliary_loss_mlp": 0.01029448,
      "balance_loss_clip": 1.04512024,
      "balance_loss_mlp": 1.02192307,
      "epoch": 0.5893104070221848,
      "flos": 19609884063360.0,
      "grad_norm": 1.925169698697133,
      "language_loss": 0.89542276,
      "learning_rate": 1.523230641627394e-06,
      "loss": 0.91718435,
      "num_input_tokens_seen": 105724730,
      "step": 4901,
      "time_per_iteration": 2.739413261413574
    },
    {
      "auxiliary_loss_clip": 0.01122173,
      "auxiliary_loss_mlp": 0.01029063,
      "balance_loss_clip": 1.04100609,
      "balance_loss_mlp": 1.02143931,
      "epoch": 0.5894306499128239,
      "flos": 29060063930880.0,
      "grad_norm": 1.8915484756659846,
      "language_loss": 0.72912544,
      "learning_rate": 1.5224741632825888e-06,
      "loss": 0.75063783,
      "num_input_tokens_seen": 105744920,
      "step": 4902,
      "time_per_iteration": 3.7648000717163086
    },
    {
      "auxiliary_loss_clip": 0.01182866,
      "auxiliary_loss_mlp": 0.01034713,
      "balance_loss_clip": 1.05384839,
      "balance_loss_mlp": 1.02732766,
      "epoch": 0.589550892803463,
      "flos": 42298890721920.0,
      "grad_norm": 1.833636762237935,
      "language_loss": 0.69204426,
      "learning_rate": 1.521717757378074e-06,
      "loss": 0.71422005,
      "num_input_tokens_seen": 105765465,
      "step": 4903,
      "time_per_iteration": 2.8651678562164307
    },
    {
      "auxiliary_loss_clip": 0.01167432,
      "auxiliary_loss_mlp": 0.01026972,
      "balance_loss_clip": 1.0501194,
      "balance_loss_mlp": 1.0192827,
      "epoch": 0.5896711356941021,
      "flos": 14137062197760.0,
      "grad_norm": 1.7163651512574383,
      "language_loss": 0.68985522,
      "learning_rate": 1.5209614240285943e-06,
      "loss": 0.71179926,
      "num_input_tokens_seen": 105783120,
      "step": 4904,
      "time_per_iteration": 2.6265976428985596
    },
    {
      "auxiliary_loss_clip": 0.01175242,
      "auxiliary_loss_mlp": 0.0076233,
      "balance_loss_clip": 1.04918015,
      "balance_loss_mlp": 1.00035501,
      "epoch": 0.5897913785847412,
      "flos": 17201355454080.0,
      "grad_norm": 2.1005658132755274,
      "language_loss": 0.8489967,
      "learning_rate": 1.520205163348887e-06,
      "loss": 0.86837244,
      "num_input_tokens_seen": 105801055,
      "step": 4905,
      "time_per_iteration": 2.5811798572540283
    },
    {
      "auxiliary_loss_clip": 0.01041798,
      "auxiliary_loss_mlp": 0.01001614,
      "balance_loss_clip": 1.00921726,
      "balance_loss_mlp": 1.00014818,
      "epoch": 0.5899116214753802,
      "flos": 48794164202880.0,
      "grad_norm": 0.7261276204999577,
      "language_loss": 0.56957895,
      "learning_rate": 1.519448975453674e-06,
      "loss": 0.59001309,
      "num_input_tokens_seen": 105856155,
      "step": 4906,
      "time_per_iteration": 3.1605563163757324
    },
    {
      "auxiliary_loss_clip": 0.01165105,
      "auxiliary_loss_mlp": 0.00763033,
      "balance_loss_clip": 1.05154741,
      "balance_loss_mlp": 1.00043154,
      "epoch": 0.5900318643660194,
      "flos": 21103659987840.0,
      "grad_norm": 3.8432336646899756,
      "language_loss": 0.75601995,
      "learning_rate": 1.5186928604576696e-06,
      "loss": 0.77530134,
      "num_input_tokens_seen": 105873350,
      "step": 4907,
      "time_per_iteration": 2.608877420425415
    },
    {
      "auxiliary_loss_clip": 0.01148758,
      "auxiliary_loss_mlp": 0.01024283,
      "balance_loss_clip": 1.04598475,
      "balance_loss_mlp": 1.01702309,
      "epoch": 0.5901521072566585,
      "flos": 21178390233600.0,
      "grad_norm": 2.058329891845543,
      "language_loss": 0.7744028,
      "learning_rate": 1.5179368184755752e-06,
      "loss": 0.79613316,
      "num_input_tokens_seen": 105891435,
      "step": 4908,
      "time_per_iteration": 2.6908013820648193
    },
    {
      "auxiliary_loss_clip": 0.01147499,
      "auxiliary_loss_mlp": 0.01025387,
      "balance_loss_clip": 1.04723144,
      "balance_loss_mlp": 1.0181632,
      "epoch": 0.5902723501472975,
      "flos": 20225967160320.0,
      "grad_norm": 1.887621407512652,
      "language_loss": 0.82585454,
      "learning_rate": 1.5171808496220821e-06,
      "loss": 0.84758341,
      "num_input_tokens_seen": 105910190,
      "step": 4909,
      "time_per_iteration": 2.6149702072143555
    },
    {
      "auxiliary_loss_clip": 0.01152232,
      "auxiliary_loss_mlp": 0.01024601,
      "balance_loss_clip": 1.04662359,
      "balance_loss_mlp": 1.01751375,
      "epoch": 0.5903925930379367,
      "flos": 22964407211520.0,
      "grad_norm": 1.7860935293149574,
      "language_loss": 0.81485391,
      "learning_rate": 1.5164249540118708e-06,
      "loss": 0.83662224,
      "num_input_tokens_seen": 105929315,
      "step": 4910,
      "time_per_iteration": 2.7021431922912598
    },
    {
      "auxiliary_loss_clip": 0.01109853,
      "auxiliary_loss_mlp": 0.0102498,
      "balance_loss_clip": 1.04196167,
      "balance_loss_mlp": 1.01764011,
      "epoch": 0.5905128359285757,
      "flos": 23367720096000.0,
      "grad_norm": 1.886022707984701,
      "language_loss": 0.83078909,
      "learning_rate": 1.5156691317596093e-06,
      "loss": 0.85213745,
      "num_input_tokens_seen": 105950740,
      "step": 4911,
      "time_per_iteration": 2.777646541595459
    },
    {
      "auxiliary_loss_clip": 0.01166011,
      "auxiliary_loss_mlp": 0.00762197,
      "balance_loss_clip": 1.04849696,
      "balance_loss_mlp": 1.00036001,
      "epoch": 0.5906330788192148,
      "flos": 28032335994240.0,
      "grad_norm": 2.2961028825192056,
      "language_loss": 0.67074621,
      "learning_rate": 1.5149133829799556e-06,
      "loss": 0.69002825,
      "num_input_tokens_seen": 105968735,
      "step": 4912,
      "time_per_iteration": 2.699658155441284
    },
    {
      "auxiliary_loss_clip": 0.01156572,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.04766798,
      "balance_loss_mlp": 1.01787746,
      "epoch": 0.590753321709854,
      "flos": 18477943793280.0,
      "grad_norm": 2.018146604083056,
      "language_loss": 0.80802184,
      "learning_rate": 1.5141577077875556e-06,
      "loss": 0.82984328,
      "num_input_tokens_seen": 105986060,
      "step": 4913,
      "time_per_iteration": 2.79193377494812
    },
    {
      "auxiliary_loss_clip": 0.0116595,
      "auxiliary_loss_mlp": 0.01022296,
      "balance_loss_clip": 1.04949164,
      "balance_loss_mlp": 1.01510501,
      "epoch": 0.590873564600493,
      "flos": 16873706568960.0,
      "grad_norm": 1.9865414966011417,
      "language_loss": 0.72472531,
      "learning_rate": 1.5134021062970451e-06,
      "loss": 0.74660778,
      "num_input_tokens_seen": 106004440,
      "step": 4914,
      "time_per_iteration": 2.6209969520568848
    },
    {
      "auxiliary_loss_clip": 0.01128058,
      "auxiliary_loss_mlp": 0.01029723,
      "balance_loss_clip": 1.04965913,
      "balance_loss_mlp": 1.02274966,
      "epoch": 0.5909938074911321,
      "flos": 13516166678400.0,
      "grad_norm": 2.098946525312931,
      "language_loss": 0.81389976,
      "learning_rate": 1.5126465786230483e-06,
      "loss": 0.83547753,
      "num_input_tokens_seen": 106021215,
      "step": 4915,
      "time_per_iteration": 2.693582534790039
    },
    {
      "auxiliary_loss_clip": 0.01176807,
      "auxiliary_loss_mlp": 0.01024616,
      "balance_loss_clip": 1.05118632,
      "balance_loss_mlp": 1.01747251,
      "epoch": 0.5911140503817712,
      "flos": 26024067613440.0,
      "grad_norm": 1.6509665836582343,
      "language_loss": 0.82004726,
      "learning_rate": 1.5118911248801787e-06,
      "loss": 0.84206146,
      "num_input_tokens_seen": 106039225,
      "step": 4916,
      "time_per_iteration": 2.644364833831787
    },
    {
      "auxiliary_loss_clip": 0.01155007,
      "auxiliary_loss_mlp": 0.01026253,
      "balance_loss_clip": 1.04569554,
      "balance_loss_mlp": 1.0192914,
      "epoch": 0.5912342932724103,
      "flos": 23258731253760.0,
      "grad_norm": 2.000289607260502,
      "language_loss": 0.79614651,
      "learning_rate": 1.5111357451830364e-06,
      "loss": 0.81795913,
      "num_input_tokens_seen": 106057920,
      "step": 4917,
      "time_per_iteration": 2.6517252922058105
    },
    {
      "auxiliary_loss_clip": 0.01160195,
      "auxiliary_loss_mlp": 0.01027498,
      "balance_loss_clip": 1.04629052,
      "balance_loss_mlp": 1.02046514,
      "epoch": 0.5913545361630493,
      "flos": 19573039687680.0,
      "grad_norm": 2.1177890270499704,
      "language_loss": 0.71032518,
      "learning_rate": 1.5103804396462131e-06,
      "loss": 0.73220211,
      "num_input_tokens_seen": 106077855,
      "step": 4918,
      "time_per_iteration": 2.6090574264526367
    },
    {
      "auxiliary_loss_clip": 0.01164928,
      "auxiliary_loss_mlp": 0.01028274,
      "balance_loss_clip": 1.04671907,
      "balance_loss_mlp": 1.02043056,
      "epoch": 0.5914747790536885,
      "flos": 26213532877440.0,
      "grad_norm": 2.1281737835293675,
      "language_loss": 0.80131173,
      "learning_rate": 1.5096252083842877e-06,
      "loss": 0.82324386,
      "num_input_tokens_seen": 106097065,
      "step": 4919,
      "time_per_iteration": 2.747882843017578
    },
    {
      "auxiliary_loss_clip": 0.0115797,
      "auxiliary_loss_mlp": 0.01027714,
      "balance_loss_clip": 1.04389763,
      "balance_loss_mlp": 1.02044857,
      "epoch": 0.5915950219443276,
      "flos": 27417545786880.0,
      "grad_norm": 1.7752138463293996,
      "language_loss": 0.85031021,
      "learning_rate": 1.5088700515118285e-06,
      "loss": 0.87216711,
      "num_input_tokens_seen": 106116385,
      "step": 4920,
      "time_per_iteration": 2.7212016582489014
    },
    {
      "auxiliary_loss_clip": 0.01132175,
      "auxiliary_loss_mlp": 0.01026164,
      "balance_loss_clip": 1.04796767,
      "balance_loss_mlp": 1.01828766,
      "epoch": 0.5917152648349666,
      "flos": 21907879545600.0,
      "grad_norm": 1.8616438125388663,
      "language_loss": 0.66644305,
      "learning_rate": 1.508114969143392e-06,
      "loss": 0.68802643,
      "num_input_tokens_seen": 106136370,
      "step": 4921,
      "time_per_iteration": 2.708624839782715
    },
    {
      "auxiliary_loss_clip": 0.01148213,
      "auxiliary_loss_mlp": 0.01029241,
      "balance_loss_clip": 1.04404163,
      "balance_loss_mlp": 1.02233303,
      "epoch": 0.5918355077256057,
      "flos": 28109185142400.0,
      "grad_norm": 1.528157886800069,
      "language_loss": 0.77279973,
      "learning_rate": 1.5073599613935238e-06,
      "loss": 0.79457426,
      "num_input_tokens_seen": 106158490,
      "step": 4922,
      "time_per_iteration": 2.7133371829986572
    },
    {
      "auxiliary_loss_clip": 0.01148971,
      "auxiliary_loss_mlp": 0.01023706,
      "balance_loss_clip": 1.04647875,
      "balance_loss_mlp": 1.0164727,
      "epoch": 0.5919557506162448,
      "flos": 28183807647360.0,
      "grad_norm": 2.7564064509657142,
      "language_loss": 0.57801664,
      "learning_rate": 1.5066050283767574e-06,
      "loss": 0.59974337,
      "num_input_tokens_seen": 106179170,
      "step": 4923,
      "time_per_iteration": 2.7322793006896973
    },
    {
      "auxiliary_loss_clip": 0.0114453,
      "auxiliary_loss_mlp": 0.01024185,
      "balance_loss_clip": 1.04692149,
      "balance_loss_mlp": 1.01744032,
      "epoch": 0.5920759935068839,
      "flos": 12094355652480.0,
      "grad_norm": 1.9951460732544322,
      "language_loss": 0.82682478,
      "learning_rate": 1.505850170207616e-06,
      "loss": 0.84851193,
      "num_input_tokens_seen": 106196035,
      "step": 4924,
      "time_per_iteration": 3.5459909439086914
    },
    {
      "auxiliary_loss_clip": 0.01147451,
      "auxiliary_loss_mlp": 0.01029908,
      "balance_loss_clip": 1.04619503,
      "balance_loss_mlp": 1.02268445,
      "epoch": 0.592196236397523,
      "flos": 29424772673280.0,
      "grad_norm": 2.1070301641481244,
      "language_loss": 0.7789886,
      "learning_rate": 1.505095387000611e-06,
      "loss": 0.80076224,
      "num_input_tokens_seen": 106218335,
      "step": 4925,
      "time_per_iteration": 3.671443462371826
    },
    {
      "auxiliary_loss_clip": 0.01139889,
      "auxiliary_loss_mlp": 0.01026907,
      "balance_loss_clip": 1.04665816,
      "balance_loss_mlp": 1.01957595,
      "epoch": 0.5923164792881621,
      "flos": 24384709866240.0,
      "grad_norm": 1.9849252037190008,
      "language_loss": 0.74387193,
      "learning_rate": 1.504340678870242e-06,
      "loss": 0.76553988,
      "num_input_tokens_seen": 106236550,
      "step": 4926,
      "time_per_iteration": 2.7268197536468506
    },
    {
      "auxiliary_loss_clip": 0.01158723,
      "auxiliary_loss_mlp": 0.01025898,
      "balance_loss_clip": 1.04713011,
      "balance_loss_mlp": 1.01853693,
      "epoch": 0.5924367221788012,
      "flos": 24024238928640.0,
      "grad_norm": 2.1989745276419184,
      "language_loss": 0.89875001,
      "learning_rate": 1.5035860459309989e-06,
      "loss": 0.92059624,
      "num_input_tokens_seen": 106254265,
      "step": 4927,
      "time_per_iteration": 3.714533567428589
    },
    {
      "auxiliary_loss_clip": 0.01143474,
      "auxiliary_loss_mlp": 0.01027204,
      "balance_loss_clip": 1.04550195,
      "balance_loss_mlp": 1.01916885,
      "epoch": 0.5925569650694402,
      "flos": 26870590414080.0,
      "grad_norm": 1.7374245404197874,
      "language_loss": 0.63824964,
      "learning_rate": 1.5028314882973568e-06,
      "loss": 0.65995634,
      "num_input_tokens_seen": 106274670,
      "step": 4928,
      "time_per_iteration": 3.6416409015655518
    },
    {
      "auxiliary_loss_clip": 0.01146805,
      "auxiliary_loss_mlp": 0.01034879,
      "balance_loss_clip": 1.0461942,
      "balance_loss_mlp": 1.02706468,
      "epoch": 0.5926772079600794,
      "flos": 22302788647680.0,
      "grad_norm": 1.8351227075015912,
      "language_loss": 0.84849453,
      "learning_rate": 1.502077006083783e-06,
      "loss": 0.87031138,
      "num_input_tokens_seen": 106293330,
      "step": 4929,
      "time_per_iteration": 2.684135913848877
    },
    {
      "auxiliary_loss_clip": 0.01166988,
      "auxiliary_loss_mlp": 0.00762061,
      "balance_loss_clip": 1.05127478,
      "balance_loss_mlp": 1.00030804,
      "epoch": 0.5927974508507184,
      "flos": 19865244827520.0,
      "grad_norm": 1.929855842175919,
      "language_loss": 0.76628655,
      "learning_rate": 1.5013225994047315e-06,
      "loss": 0.78557706,
      "num_input_tokens_seen": 106310960,
      "step": 4930,
      "time_per_iteration": 2.664365291595459
    },
    {
      "auxiliary_loss_clip": 0.01164566,
      "auxiliary_loss_mlp": 0.00762391,
      "balance_loss_clip": 1.05053985,
      "balance_loss_mlp": 1.00041056,
      "epoch": 0.5929176937413575,
      "flos": 15776743167360.0,
      "grad_norm": 1.5601533452751828,
      "language_loss": 0.80694544,
      "learning_rate": 1.5005682683746452e-06,
      "loss": 0.82621503,
      "num_input_tokens_seen": 106329475,
      "step": 4931,
      "time_per_iteration": 2.655264139175415
    },
    {
      "auxiliary_loss_clip": 0.01165671,
      "auxiliary_loss_mlp": 0.01023696,
      "balance_loss_clip": 1.05232048,
      "balance_loss_mlp": 1.01661181,
      "epoch": 0.5930379366319967,
      "flos": 17601472028160.0,
      "grad_norm": 3.473202059069269,
      "language_loss": 0.73014712,
      "learning_rate": 1.4998140131079553e-06,
      "loss": 0.75204086,
      "num_input_tokens_seen": 106345565,
      "step": 4932,
      "time_per_iteration": 2.645610809326172
    },
    {
      "auxiliary_loss_clip": 0.01102625,
      "auxiliary_loss_mlp": 0.00762328,
      "balance_loss_clip": 1.04204106,
      "balance_loss_mlp": 1.00031996,
      "epoch": 0.5931581795226357,
      "flos": 17704283731200.0,
      "grad_norm": 1.7952848941786759,
      "language_loss": 0.73371428,
      "learning_rate": 1.4990598337190821e-06,
      "loss": 0.7523638,
      "num_input_tokens_seen": 106361920,
      "step": 4933,
      "time_per_iteration": 2.869096517562866
    },
    {
      "auxiliary_loss_clip": 0.01174698,
      "auxiliary_loss_mlp": 0.0076212,
      "balance_loss_clip": 1.04931426,
      "balance_loss_mlp": 1.00038075,
      "epoch": 0.5932784224132748,
      "flos": 24280102483200.0,
      "grad_norm": 1.8609330816637781,
      "language_loss": 0.68177021,
      "learning_rate": 1.4983057303224338e-06,
      "loss": 0.70113838,
      "num_input_tokens_seen": 106381735,
      "step": 4934,
      "time_per_iteration": 2.7019543647766113
    },
    {
      "auxiliary_loss_clip": 0.01117831,
      "auxiliary_loss_mlp": 0.01027254,
      "balance_loss_clip": 1.04206979,
      "balance_loss_mlp": 1.02008104,
      "epoch": 0.5933986653039139,
      "flos": 22926700909440.0,
      "grad_norm": 1.6330041852577895,
      "language_loss": 0.87480175,
      "learning_rate": 1.4975517030324072e-06,
      "loss": 0.89625263,
      "num_input_tokens_seen": 106399745,
      "step": 4935,
      "time_per_iteration": 2.879046678543091
    },
    {
      "auxiliary_loss_clip": 0.0106941,
      "auxiliary_loss_mlp": 0.00753789,
      "balance_loss_clip": 1.01016378,
      "balance_loss_mlp": 1.00033402,
      "epoch": 0.593518908194553,
      "flos": 71121730256640.0,
      "grad_norm": 0.7834552091149868,
      "language_loss": 0.61766863,
      "learning_rate": 1.4967977519633882e-06,
      "loss": 0.63590062,
      "num_input_tokens_seen": 106457205,
      "step": 4936,
      "time_per_iteration": 3.375359058380127
    },
    {
      "auxiliary_loss_clip": 0.01133161,
      "auxiliary_loss_mlp": 0.01025138,
      "balance_loss_clip": 1.04591322,
      "balance_loss_mlp": 1.01717472,
      "epoch": 0.593639151085192,
      "flos": 20448649526400.0,
      "grad_norm": 1.8996652854987477,
      "language_loss": 0.78097266,
      "learning_rate": 1.4960438772297494e-06,
      "loss": 0.80255568,
      "num_input_tokens_seen": 106474250,
      "step": 4937,
      "time_per_iteration": 2.631826162338257
    },
    {
      "auxiliary_loss_clip": 0.01151835,
      "auxiliary_loss_mlp": 0.01025471,
      "balance_loss_clip": 1.04609859,
      "balance_loss_mlp": 1.01803565,
      "epoch": 0.5937593939758312,
      "flos": 30883428074880.0,
      "grad_norm": 2.406045999110213,
      "language_loss": 0.73724508,
      "learning_rate": 1.495290078945855e-06,
      "loss": 0.75901812,
      "num_input_tokens_seen": 106494015,
      "step": 4938,
      "time_per_iteration": 2.7481181621551514
    },
    {
      "auxiliary_loss_clip": 0.01175109,
      "auxiliary_loss_mlp": 0.01029255,
      "balance_loss_clip": 1.05077052,
      "balance_loss_mlp": 1.02201557,
      "epoch": 0.5938796368664703,
      "flos": 36898069668480.0,
      "grad_norm": 3.1565330882268703,
      "language_loss": 0.74553788,
      "learning_rate": 1.4945363572260529e-06,
      "loss": 0.76758146,
      "num_input_tokens_seen": 106515010,
      "step": 4939,
      "time_per_iteration": 2.7523868083953857
    },
    {
      "auxiliary_loss_clip": 0.0115798,
      "auxiliary_loss_mlp": 0.01026481,
      "balance_loss_clip": 1.04702652,
      "balance_loss_mlp": 1.01943564,
      "epoch": 0.5939998797571093,
      "flos": 23842926051840.0,
      "grad_norm": 2.07074245523757,
      "language_loss": 0.67881203,
      "learning_rate": 1.4937827121846845e-06,
      "loss": 0.70065665,
      "num_input_tokens_seen": 106535265,
      "step": 4940,
      "time_per_iteration": 2.6404669284820557
    },
    {
      "auxiliary_loss_clip": 0.0112937,
      "auxiliary_loss_mlp": 0.01029876,
      "balance_loss_clip": 1.04728198,
      "balance_loss_mlp": 1.02221644,
      "epoch": 0.5941201226477485,
      "flos": 25191407462400.0,
      "grad_norm": 1.528919300996716,
      "language_loss": 0.73509103,
      "learning_rate": 1.4930291439360755e-06,
      "loss": 0.75668347,
      "num_input_tokens_seen": 106557830,
      "step": 4941,
      "time_per_iteration": 2.7574355602264404
    },
    {
      "auxiliary_loss_clip": 0.01162922,
      "auxiliary_loss_mlp": 0.01032008,
      "balance_loss_clip": 1.04863036,
      "balance_loss_mlp": 1.02381253,
      "epoch": 0.5942403655383875,
      "flos": 22418996123520.0,
      "grad_norm": 1.8821525518320121,
      "language_loss": 0.79288542,
      "learning_rate": 1.4922756525945427e-06,
      "loss": 0.81483471,
      "num_input_tokens_seen": 106577140,
      "step": 4942,
      "time_per_iteration": 2.6362712383270264
    },
    {
      "auxiliary_loss_clip": 0.01059847,
      "auxiliary_loss_mlp": 0.01000862,
      "balance_loss_clip": 1.00977492,
      "balance_loss_mlp": 0.99940729,
      "epoch": 0.5943606084290266,
      "flos": 67629310796160.0,
      "grad_norm": 0.7745454027183941,
      "language_loss": 0.59575236,
      "learning_rate": 1.4915222382743894e-06,
      "loss": 0.61635947,
      "num_input_tokens_seen": 106635975,
      "step": 4943,
      "time_per_iteration": 3.3216826915740967
    },
    {
      "auxiliary_loss_clip": 0.01164709,
      "auxiliary_loss_mlp": 0.01031785,
      "balance_loss_clip": 1.05090916,
      "balance_loss_mlp": 1.02457869,
      "epoch": 0.5944808513196658,
      "flos": 18223157646720.0,
      "grad_norm": 2.4145949082932643,
      "language_loss": 0.71859765,
      "learning_rate": 1.4907689010899085e-06,
      "loss": 0.74056262,
      "num_input_tokens_seen": 106653555,
      "step": 4944,
      "time_per_iteration": 2.696063995361328
    },
    {
      "auxiliary_loss_clip": 0.01147255,
      "auxiliary_loss_mlp": 0.01022284,
      "balance_loss_clip": 1.04736722,
      "balance_loss_mlp": 1.01478899,
      "epoch": 0.5946010942103048,
      "flos": 24790824011520.0,
      "grad_norm": 1.9970512468044177,
      "language_loss": 0.62627369,
      "learning_rate": 1.4900156411553804e-06,
      "loss": 0.64796913,
      "num_input_tokens_seen": 106673385,
      "step": 4945,
      "time_per_iteration": 2.755937099456787
    },
    {
      "auxiliary_loss_clip": 0.01151707,
      "auxiliary_loss_mlp": 0.01033008,
      "balance_loss_clip": 1.04963422,
      "balance_loss_mlp": 1.02566469,
      "epoch": 0.5947213371009439,
      "flos": 15231619388160.0,
      "grad_norm": 2.217266874433972,
      "language_loss": 0.85504138,
      "learning_rate": 1.4892624585850739e-06,
      "loss": 0.87688857,
      "num_input_tokens_seen": 106691740,
      "step": 4946,
      "time_per_iteration": 2.626225471496582
    },
    {
      "auxiliary_loss_clip": 0.01178728,
      "auxiliary_loss_mlp": 0.01025544,
      "balance_loss_clip": 1.05095673,
      "balance_loss_mlp": 1.01769459,
      "epoch": 0.594841579991583,
      "flos": 25848069949440.0,
      "grad_norm": 3.7621189102549693,
      "language_loss": 0.7968986,
      "learning_rate": 1.4885093534932465e-06,
      "loss": 0.81894135,
      "num_input_tokens_seen": 106709705,
      "step": 4947,
      "time_per_iteration": 2.6524109840393066
    },
    {
      "auxiliary_loss_clip": 0.01149233,
      "auxiliary_loss_mlp": 0.01031908,
      "balance_loss_clip": 1.05021203,
      "balance_loss_mlp": 1.02427268,
      "epoch": 0.5949618228822221,
      "flos": 23981109672960.0,
      "grad_norm": 2.17079063834989,
      "language_loss": 0.71543622,
      "learning_rate": 1.4877563259941433e-06,
      "loss": 0.73724759,
      "num_input_tokens_seen": 106727560,
      "step": 4948,
      "time_per_iteration": 2.695841073989868
    },
    {
      "auxiliary_loss_clip": 0.01173635,
      "auxiliary_loss_mlp": 0.01027662,
      "balance_loss_clip": 1.05417943,
      "balance_loss_mlp": 1.01959193,
      "epoch": 0.5950820657728612,
      "flos": 40547491476480.0,
      "grad_norm": 2.5168648517024703,
      "language_loss": 0.68135923,
      "learning_rate": 1.4870033762019988e-06,
      "loss": 0.70337224,
      "num_input_tokens_seen": 106747725,
      "step": 4949,
      "time_per_iteration": 2.7337679862976074
    },
    {
      "auxiliary_loss_clip": 0.01144199,
      "auxiliary_loss_mlp": 0.01024535,
      "balance_loss_clip": 1.04590869,
      "balance_loss_mlp": 1.01675701,
      "epoch": 0.5952023086635003,
      "flos": 23184467884800.0,
      "grad_norm": 1.5581550825109456,
      "language_loss": 0.73364913,
      "learning_rate": 1.4862505042310334e-06,
      "loss": 0.75533652,
      "num_input_tokens_seen": 106767010,
      "step": 4950,
      "time_per_iteration": 3.6990439891815186
    },
    {
      "auxiliary_loss_clip": 0.0113992,
      "auxiliary_loss_mlp": 0.01028723,
      "balance_loss_clip": 1.04666436,
      "balance_loss_mlp": 1.02115345,
      "epoch": 0.5953225515541394,
      "flos": 33653289548160.0,
      "grad_norm": 1.6561506438949212,
      "language_loss": 0.69615495,
      "learning_rate": 1.4854977101954587e-06,
      "loss": 0.71784139,
      "num_input_tokens_seen": 106789230,
      "step": 4951,
      "time_per_iteration": 3.6728219985961914
    },
    {
      "auxiliary_loss_clip": 0.01161938,
      "auxiliary_loss_mlp": 0.0102755,
      "balance_loss_clip": 1.04527819,
      "balance_loss_mlp": 1.01977444,
      "epoch": 0.5954427944447784,
      "flos": 24459619680000.0,
      "grad_norm": 1.8769045931652921,
      "language_loss": 0.86637646,
      "learning_rate": 1.4847449942094716e-06,
      "loss": 0.88827133,
      "num_input_tokens_seen": 106808110,
      "step": 4952,
      "time_per_iteration": 3.5909407138824463
    },
    {
      "auxiliary_loss_clip": 0.01142818,
      "auxiliary_loss_mlp": 0.0102228,
      "balance_loss_clip": 1.04623818,
      "balance_loss_mlp": 1.01478219,
      "epoch": 0.5955630373354175,
      "flos": 18551848026240.0,
      "grad_norm": 2.338284585982738,
      "language_loss": 0.8639586,
      "learning_rate": 1.4839923563872598e-06,
      "loss": 0.88560957,
      "num_input_tokens_seen": 106826650,
      "step": 4953,
      "time_per_iteration": 2.668748140335083
    },
    {
      "auxiliary_loss_clip": 0.01134512,
      "auxiliary_loss_mlp": 0.01032523,
      "balance_loss_clip": 1.04762721,
      "balance_loss_mlp": 1.02506077,
      "epoch": 0.5956832802260567,
      "flos": 19791699730560.0,
      "grad_norm": 1.6809302522707876,
      "language_loss": 0.75966096,
      "learning_rate": 1.483239796842997e-06,
      "loss": 0.7813313,
      "num_input_tokens_seen": 106844680,
      "step": 4954,
      "time_per_iteration": 2.658351421356201
    },
    {
      "auxiliary_loss_clip": 0.01134189,
      "auxiliary_loss_mlp": 0.01028189,
      "balance_loss_clip": 1.04570949,
      "balance_loss_mlp": 1.02060688,
      "epoch": 0.5958035231166957,
      "flos": 19750868945280.0,
      "grad_norm": 1.622825129218234,
      "language_loss": 0.83730525,
      "learning_rate": 1.4824873156908462e-06,
      "loss": 0.85892904,
      "num_input_tokens_seen": 106862605,
      "step": 4955,
      "time_per_iteration": 3.5932064056396484
    },
    {
      "auxiliary_loss_clip": 0.01166817,
      "auxiliary_loss_mlp": 0.00763133,
      "balance_loss_clip": 1.05168653,
      "balance_loss_mlp": 1.00038528,
      "epoch": 0.5959237660073348,
      "flos": 21652806090240.0,
      "grad_norm": 1.5194661797305027,
      "language_loss": 0.75879288,
      "learning_rate": 1.4817349130449584e-06,
      "loss": 0.77809238,
      "num_input_tokens_seen": 106882325,
      "step": 4956,
      "time_per_iteration": 2.6862120628356934
    },
    {
      "auxiliary_loss_clip": 0.01159185,
      "auxiliary_loss_mlp": 0.01025453,
      "balance_loss_clip": 1.04763031,
      "balance_loss_mlp": 1.01839542,
      "epoch": 0.5960440088979739,
      "flos": 21171207513600.0,
      "grad_norm": 2.6989827599012455,
      "language_loss": 0.82674837,
      "learning_rate": 1.4809825890194717e-06,
      "loss": 0.84859478,
      "num_input_tokens_seen": 106900995,
      "step": 4957,
      "time_per_iteration": 2.630902051925659
    },
    {
      "auxiliary_loss_clip": 0.01142236,
      "auxiliary_loss_mlp": 0.01026093,
      "balance_loss_clip": 1.04552639,
      "balance_loss_mlp": 1.0185771,
      "epoch": 0.596164251788613,
      "flos": 14757526753920.0,
      "grad_norm": 1.8605523188671738,
      "language_loss": 0.77308142,
      "learning_rate": 1.4802303437285139e-06,
      "loss": 0.7947647,
      "num_input_tokens_seen": 106918265,
      "step": 4958,
      "time_per_iteration": 2.8147284984588623
    },
    {
      "auxiliary_loss_clip": 0.01144943,
      "auxiliary_loss_mlp": 0.01022297,
      "balance_loss_clip": 1.04495668,
      "balance_loss_mlp": 1.01477754,
      "epoch": 0.596284494679252,
      "flos": 20485924865280.0,
      "grad_norm": 3.1164179526804237,
      "language_loss": 0.80827016,
      "learning_rate": 1.4794781772861994e-06,
      "loss": 0.82994258,
      "num_input_tokens_seen": 106934760,
      "step": 4959,
      "time_per_iteration": 2.6869027614593506
    },
    {
      "auxiliary_loss_clip": 0.01146989,
      "auxiliary_loss_mlp": 0.00762262,
      "balance_loss_clip": 1.04625249,
      "balance_loss_mlp": 1.00038064,
      "epoch": 0.5964047375698912,
      "flos": 31212262108800.0,
      "grad_norm": 1.948902827572513,
      "language_loss": 0.66892266,
      "learning_rate": 1.4787260898066324e-06,
      "loss": 0.68801522,
      "num_input_tokens_seen": 106954760,
      "step": 4960,
      "time_per_iteration": 2.771940231323242
    },
    {
      "auxiliary_loss_clip": 0.01174262,
      "auxiliary_loss_mlp": 0.01024873,
      "balance_loss_clip": 1.0502882,
      "balance_loss_mlp": 1.01764321,
      "epoch": 0.5965249804605303,
      "flos": 27483620855040.0,
      "grad_norm": 2.1605123964985116,
      "language_loss": 0.85705084,
      "learning_rate": 1.4779740814039023e-06,
      "loss": 0.87904227,
      "num_input_tokens_seen": 106974845,
      "step": 4961,
      "time_per_iteration": 2.6879870891571045
    },
    {
      "auxiliary_loss_clip": 0.01174596,
      "auxiliary_loss_mlp": 0.01021949,
      "balance_loss_clip": 1.04902303,
      "balance_loss_mlp": 1.01386094,
      "epoch": 0.5966452233511693,
      "flos": 30773936442240.0,
      "grad_norm": 1.9437018735303446,
      "language_loss": 0.6871255,
      "learning_rate": 1.4772221521920894e-06,
      "loss": 0.70909095,
      "num_input_tokens_seen": 106994870,
      "step": 4962,
      "time_per_iteration": 2.7051305770874023
    },
    {
      "auxiliary_loss_clip": 0.01147628,
      "auxiliary_loss_mlp": 0.01029238,
      "balance_loss_clip": 1.04878104,
      "balance_loss_mlp": 1.02145314,
      "epoch": 0.5967654662418085,
      "flos": 25481170477440.0,
      "grad_norm": 3.496652556396209,
      "language_loss": 0.74320066,
      "learning_rate": 1.4764703022852598e-06,
      "loss": 0.76496929,
      "num_input_tokens_seen": 107015390,
      "step": 4963,
      "time_per_iteration": 2.7165348529815674
    },
    {
      "auxiliary_loss_clip": 0.01092845,
      "auxiliary_loss_mlp": 0.0102811,
      "balance_loss_clip": 1.03999472,
      "balance_loss_mlp": 1.02144647,
      "epoch": 0.5968857091324475,
      "flos": 19099126621440.0,
      "grad_norm": 1.9836007547313463,
      "language_loss": 0.77198255,
      "learning_rate": 1.4757185317974696e-06,
      "loss": 0.79319215,
      "num_input_tokens_seen": 107033775,
      "step": 4964,
      "time_per_iteration": 2.7762391567230225
    },
    {
      "auxiliary_loss_clip": 0.01163314,
      "auxiliary_loss_mlp": 0.01026683,
      "balance_loss_clip": 1.04860091,
      "balance_loss_mlp": 1.01860702,
      "epoch": 0.5970059520230866,
      "flos": 23692711374720.0,
      "grad_norm": 2.2274890348095715,
      "language_loss": 0.70681584,
      "learning_rate": 1.474966840842761e-06,
      "loss": 0.72871578,
      "num_input_tokens_seen": 107053355,
      "step": 4965,
      "time_per_iteration": 2.6759989261627197
    },
    {
      "auxiliary_loss_clip": 0.01162876,
      "auxiliary_loss_mlp": 0.010249,
      "balance_loss_clip": 1.04797935,
      "balance_loss_mlp": 1.01751161,
      "epoch": 0.5971261949137258,
      "flos": 23185545292800.0,
      "grad_norm": 1.8216510722485046,
      "language_loss": 0.87092376,
      "learning_rate": 1.4742152295351655e-06,
      "loss": 0.89280146,
      "num_input_tokens_seen": 107072510,
      "step": 4966,
      "time_per_iteration": 2.662252187728882
    },
    {
      "auxiliary_loss_clip": 0.01162492,
      "auxiliary_loss_mlp": 0.00762976,
      "balance_loss_clip": 1.0482918,
      "balance_loss_mlp": 1.00030446,
      "epoch": 0.5972464378043648,
      "flos": 20557710195840.0,
      "grad_norm": 2.310148593338156,
      "language_loss": 0.64144266,
      "learning_rate": 1.4734636979887016e-06,
      "loss": 0.66069734,
      "num_input_tokens_seen": 107089970,
      "step": 4967,
      "time_per_iteration": 2.656702995300293
    },
    {
      "auxiliary_loss_clip": 0.01138861,
      "auxiliary_loss_mlp": 0.01024341,
      "balance_loss_clip": 1.04556227,
      "balance_loss_mlp": 1.01686072,
      "epoch": 0.5973666806950039,
      "flos": 29387030457600.0,
      "grad_norm": 1.9570763154506534,
      "language_loss": 0.9030174,
      "learning_rate": 1.4727122463173755e-06,
      "loss": 0.92464948,
      "num_input_tokens_seen": 107108500,
      "step": 4968,
      "time_per_iteration": 2.8027780055999756
    },
    {
      "auxiliary_loss_clip": 0.01145023,
      "auxiliary_loss_mlp": 0.01025826,
      "balance_loss_clip": 1.0479207,
      "balance_loss_mlp": 1.01850414,
      "epoch": 0.597486923585643,
      "flos": 22273522041600.0,
      "grad_norm": 1.8692096463166925,
      "language_loss": 0.64752495,
      "learning_rate": 1.471960874635183e-06,
      "loss": 0.66923344,
      "num_input_tokens_seen": 107128060,
      "step": 4969,
      "time_per_iteration": 2.91204571723938
    },
    {
      "auxiliary_loss_clip": 0.01140117,
      "auxiliary_loss_mlp": 0.01023938,
      "balance_loss_clip": 1.04353905,
      "balance_loss_mlp": 1.01637435,
      "epoch": 0.5976071664762821,
      "flos": 13772461196160.0,
      "grad_norm": 2.4784733591170847,
      "language_loss": 0.70519096,
      "learning_rate": 1.4712095830561055e-06,
      "loss": 0.7268315,
      "num_input_tokens_seen": 107146550,
      "step": 4970,
      "time_per_iteration": 2.671529769897461
    },
    {
      "auxiliary_loss_clip": 0.01146363,
      "auxiliary_loss_mlp": 0.01022786,
      "balance_loss_clip": 1.04487586,
      "balance_loss_mlp": 1.01563644,
      "epoch": 0.5977274093669211,
      "flos": 19098623831040.0,
      "grad_norm": 1.7337669579252672,
      "language_loss": 0.80786115,
      "learning_rate": 1.4704583716941147e-06,
      "loss": 0.82955265,
      "num_input_tokens_seen": 107165415,
      "step": 4971,
      "time_per_iteration": 2.663783073425293
    },
    {
      "auxiliary_loss_clip": 0.01155808,
      "auxiliary_loss_mlp": 0.01026793,
      "balance_loss_clip": 1.04874134,
      "balance_loss_mlp": 1.01850832,
      "epoch": 0.5978476522575603,
      "flos": 20376002269440.0,
      "grad_norm": 1.7485822365449648,
      "language_loss": 0.72292298,
      "learning_rate": 1.4697072406631672e-06,
      "loss": 0.74474895,
      "num_input_tokens_seen": 107185320,
      "step": 4972,
      "time_per_iteration": 2.6577064990997314
    },
    {
      "auxiliary_loss_clip": 0.01126295,
      "auxiliary_loss_mlp": 0.01037081,
      "balance_loss_clip": 1.04964566,
      "balance_loss_mlp": 1.02931464,
      "epoch": 0.5979678951481994,
      "flos": 29023147728000.0,
      "grad_norm": 1.6442295515928538,
      "language_loss": 0.7258296,
      "learning_rate": 1.4689561900772097e-06,
      "loss": 0.74746335,
      "num_input_tokens_seen": 107205380,
      "step": 4973,
      "time_per_iteration": 2.7852942943573
    },
    {
      "auxiliary_loss_clip": 0.01144557,
      "auxiliary_loss_mlp": 0.01034888,
      "balance_loss_clip": 1.04388666,
      "balance_loss_mlp": 1.02733016,
      "epoch": 0.5980881380388384,
      "flos": 17967689141760.0,
      "grad_norm": 2.3590652531219525,
      "language_loss": 0.72582763,
      "learning_rate": 1.4682052200501758e-06,
      "loss": 0.74762213,
      "num_input_tokens_seen": 107222585,
      "step": 4974,
      "time_per_iteration": 2.664034128189087
    },
    {
      "auxiliary_loss_clip": 0.01173494,
      "auxiliary_loss_mlp": 0.01029593,
      "balance_loss_clip": 1.04869485,
      "balance_loss_mlp": 1.02242565,
      "epoch": 0.5982083809294776,
      "flos": 22962827013120.0,
      "grad_norm": 1.742813449056899,
      "language_loss": 0.80094767,
      "learning_rate": 1.4674543306959876e-06,
      "loss": 0.8229785,
      "num_input_tokens_seen": 107242055,
      "step": 4975,
      "time_per_iteration": 2.72810959815979
    },
    {
      "auxiliary_loss_clip": 0.0115124,
      "auxiliary_loss_mlp": 0.01028598,
      "balance_loss_clip": 1.04658675,
      "balance_loss_mlp": 1.02006805,
      "epoch": 0.5983286238201166,
      "flos": 20991941712000.0,
      "grad_norm": 2.2135592536200397,
      "language_loss": 0.84194541,
      "learning_rate": 1.4667035221285535e-06,
      "loss": 0.8637439,
      "num_input_tokens_seen": 107259695,
      "step": 4976,
      "time_per_iteration": 3.6211395263671875
    },
    {
      "auxiliary_loss_clip": 0.01159763,
      "auxiliary_loss_mlp": 0.0102519,
      "balance_loss_clip": 1.04996049,
      "balance_loss_mlp": 1.01801991,
      "epoch": 0.5984488667107557,
      "flos": 28183448511360.0,
      "grad_norm": 1.8937107112323335,
      "language_loss": 0.74422824,
      "learning_rate": 1.4659527944617715e-06,
      "loss": 0.76607776,
      "num_input_tokens_seen": 107279640,
      "step": 4977,
      "time_per_iteration": 3.544301986694336
    },
    {
      "auxiliary_loss_clip": 0.01098939,
      "auxiliary_loss_mlp": 0.01027912,
      "balance_loss_clip": 1.04045033,
      "balance_loss_mlp": 1.0204674,
      "epoch": 0.5985691096013949,
      "flos": 16471794314880.0,
      "grad_norm": 1.9855193615362312,
      "language_loss": 0.76120651,
      "learning_rate": 1.465202147809526e-06,
      "loss": 0.78247505,
      "num_input_tokens_seen": 107298135,
      "step": 4978,
      "time_per_iteration": 3.659019947052002
    },
    {
      "auxiliary_loss_clip": 0.01176185,
      "auxiliary_loss_mlp": 0.01026107,
      "balance_loss_clip": 1.05002856,
      "balance_loss_mlp": 1.01907349,
      "epoch": 0.5986893524920339,
      "flos": 26719046933760.0,
      "grad_norm": 2.126129869938563,
      "language_loss": 0.76334977,
      "learning_rate": 1.4644515822856888e-06,
      "loss": 0.78537267,
      "num_input_tokens_seen": 107316570,
      "step": 4979,
      "time_per_iteration": 2.617230176925659
    },
    {
      "auxiliary_loss_clip": 0.01037436,
      "auxiliary_loss_mlp": 0.0100291,
      "balance_loss_clip": 1.00956309,
      "balance_loss_mlp": 1.00145578,
      "epoch": 0.598809595382673,
      "flos": 61608061100160.0,
      "grad_norm": 0.7577283473028841,
      "language_loss": 0.56525201,
      "learning_rate": 1.4637010980041215e-06,
      "loss": 0.58565545,
      "num_input_tokens_seen": 107378680,
      "step": 4980,
      "time_per_iteration": 4.172622442245483
    },
    {
      "auxiliary_loss_clip": 0.01177762,
      "auxiliary_loss_mlp": 0.01032541,
      "balance_loss_clip": 1.05156636,
      "balance_loss_mlp": 1.02447009,
      "epoch": 0.5989298382733121,
      "flos": 11801719549440.0,
      "grad_norm": 2.2015109682280674,
      "language_loss": 0.89729786,
      "learning_rate": 1.4629506950786707e-06,
      "loss": 0.91940081,
      "num_input_tokens_seen": 107394860,
      "step": 4981,
      "time_per_iteration": 2.6379129886627197
    },
    {
      "auxiliary_loss_clip": 0.01069123,
      "auxiliary_loss_mlp": 0.01001658,
      "balance_loss_clip": 1.01047826,
      "balance_loss_mlp": 1.00028121,
      "epoch": 0.5990500811639512,
      "flos": 60025800021120.0,
      "grad_norm": 0.8155934540320864,
      "language_loss": 0.56046015,
      "learning_rate": 1.4622003736231733e-06,
      "loss": 0.58116794,
      "num_input_tokens_seen": 107453850,
      "step": 4982,
      "time_per_iteration": 3.2368595600128174
    },
    {
      "auxiliary_loss_clip": 0.0115867,
      "auxiliary_loss_mlp": 0.01026704,
      "balance_loss_clip": 1.04893994,
      "balance_loss_mlp": 1.0190922,
      "epoch": 0.5991703240545903,
      "flos": 18222726683520.0,
      "grad_norm": 1.8040983779012478,
      "language_loss": 0.80395114,
      "learning_rate": 1.461450133751451e-06,
      "loss": 0.82580489,
      "num_input_tokens_seen": 107471920,
      "step": 4983,
      "time_per_iteration": 2.6203761100769043
    },
    {
      "auxiliary_loss_clip": 0.01164115,
      "auxiliary_loss_mlp": 0.0102446,
      "balance_loss_clip": 1.04836094,
      "balance_loss_mlp": 1.01723886,
      "epoch": 0.5992905669452293,
      "flos": 27709894581120.0,
      "grad_norm": 1.699729858697749,
      "language_loss": 0.75607479,
      "learning_rate": 1.4606999755773153e-06,
      "loss": 0.77796054,
      "num_input_tokens_seen": 107493125,
      "step": 4984,
      "time_per_iteration": 2.673377275466919
    },
    {
      "auxiliary_loss_clip": 0.01174456,
      "auxiliary_loss_mlp": 0.01025381,
      "balance_loss_clip": 1.05015087,
      "balance_loss_mlp": 1.01764393,
      "epoch": 0.5994108098358685,
      "flos": 20449008662400.0,
      "grad_norm": 1.7105095461652378,
      "language_loss": 0.82045698,
      "learning_rate": 1.4599498992145643e-06,
      "loss": 0.84245539,
      "num_input_tokens_seen": 107513150,
      "step": 4985,
      "time_per_iteration": 2.657057046890259
    },
    {
      "auxiliary_loss_clip": 0.01151239,
      "auxiliary_loss_mlp": 0.0076247,
      "balance_loss_clip": 1.04745626,
      "balance_loss_mlp": 1.00044823,
      "epoch": 0.5995310527265075,
      "flos": 22269966595200.0,
      "grad_norm": 2.821284256225329,
      "language_loss": 0.70468926,
      "learning_rate": 1.4591999047769846e-06,
      "loss": 0.72382629,
      "num_input_tokens_seen": 107532005,
      "step": 4986,
      "time_per_iteration": 2.6651856899261475
    },
    {
      "auxiliary_loss_clip": 0.01098931,
      "auxiliary_loss_mlp": 0.01028446,
      "balance_loss_clip": 1.03836322,
      "balance_loss_mlp": 1.02073312,
      "epoch": 0.5996512956171466,
      "flos": 18916951818240.0,
      "grad_norm": 1.6655067413566855,
      "language_loss": 0.7570219,
      "learning_rate": 1.4584499923783486e-06,
      "loss": 0.77829564,
      "num_input_tokens_seen": 107550585,
      "step": 4987,
      "time_per_iteration": 2.6733927726745605
    },
    {
      "auxiliary_loss_clip": 0.01148702,
      "auxiliary_loss_mlp": 0.01027656,
      "balance_loss_clip": 1.04833603,
      "balance_loss_mlp": 1.0199734,
      "epoch": 0.5997715385077858,
      "flos": 15370916330880.0,
      "grad_norm": 2.4458279903328397,
      "language_loss": 0.75861722,
      "learning_rate": 1.457700162132419e-06,
      "loss": 0.78038085,
      "num_input_tokens_seen": 107567575,
      "step": 4988,
      "time_per_iteration": 2.647803783416748
    },
    {
      "auxiliary_loss_clip": 0.01116751,
      "auxiliary_loss_mlp": 0.01025305,
      "balance_loss_clip": 1.04427564,
      "balance_loss_mlp": 1.01784861,
      "epoch": 0.5998917813984248,
      "flos": 25264844818560.0,
      "grad_norm": 2.0274138832116915,
      "language_loss": 0.72447371,
      "learning_rate": 1.4569504141529433e-06,
      "loss": 0.74589431,
      "num_input_tokens_seen": 107585410,
      "step": 4989,
      "time_per_iteration": 2.7681922912597656
    },
    {
      "auxiliary_loss_clip": 0.01160683,
      "auxiliary_loss_mlp": 0.01024908,
      "balance_loss_clip": 1.05025816,
      "balance_loss_mlp": 1.01673007,
      "epoch": 0.6000120242890639,
      "flos": 22054502862720.0,
      "grad_norm": 2.229542913503612,
      "language_loss": 0.72293025,
      "learning_rate": 1.456200748553658e-06,
      "loss": 0.74478614,
      "num_input_tokens_seen": 107603405,
      "step": 4990,
      "time_per_iteration": 2.7586252689361572
    },
    {
      "auxiliary_loss_clip": 0.01177822,
      "auxiliary_loss_mlp": 0.01034564,
      "balance_loss_clip": 1.05173874,
      "balance_loss_mlp": 1.02647591,
      "epoch": 0.600132267179703,
      "flos": 29863421562240.0,
      "grad_norm": 1.8311269296003176,
      "language_loss": 0.78879142,
      "learning_rate": 1.455451165448287e-06,
      "loss": 0.81091529,
      "num_input_tokens_seen": 107626060,
      "step": 4991,
      "time_per_iteration": 2.7449238300323486
    },
    {
      "auxiliary_loss_clip": 0.01146881,
      "auxiliary_loss_mlp": 0.01031124,
      "balance_loss_clip": 1.04713464,
      "balance_loss_mlp": 1.0230056,
      "epoch": 0.6002525100703421,
      "flos": 25045358762880.0,
      "grad_norm": 2.8142612497468256,
      "language_loss": 0.73995018,
      "learning_rate": 1.4547016649505407e-06,
      "loss": 0.76173019,
      "num_input_tokens_seen": 107644070,
      "step": 4992,
      "time_per_iteration": 2.712965250015259
    },
    {
      "auxiliary_loss_clip": 0.01129807,
      "auxiliary_loss_mlp": 0.01030074,
      "balance_loss_clip": 1.04079795,
      "balance_loss_mlp": 1.02222979,
      "epoch": 0.6003727529609811,
      "flos": 20849592113280.0,
      "grad_norm": 2.066832787943725,
      "language_loss": 0.84808391,
      "learning_rate": 1.4539522471741193e-06,
      "loss": 0.86968267,
      "num_input_tokens_seen": 107661495,
      "step": 4993,
      "time_per_iteration": 2.9016408920288086
    },
    {
      "auxiliary_loss_clip": 0.01168101,
      "auxiliary_loss_mlp": 0.01031301,
      "balance_loss_clip": 1.05004084,
      "balance_loss_mlp": 1.02296257,
      "epoch": 0.6004929958516203,
      "flos": 15594604277760.0,
      "grad_norm": 2.1687986983771927,
      "language_loss": 0.71459174,
      "learning_rate": 1.4532029122327067e-06,
      "loss": 0.73658574,
      "num_input_tokens_seen": 107678280,
      "step": 4994,
      "time_per_iteration": 2.6248929500579834
    },
    {
      "auxiliary_loss_clip": 0.01128294,
      "auxiliary_loss_mlp": 0.01028773,
      "balance_loss_clip": 1.04761577,
      "balance_loss_mlp": 1.02173376,
      "epoch": 0.6006132387422594,
      "flos": 21763267390080.0,
      "grad_norm": 1.9993562021188211,
      "language_loss": 0.75652814,
      "learning_rate": 1.4524536602399783e-06,
      "loss": 0.77809882,
      "num_input_tokens_seen": 107697370,
      "step": 4995,
      "time_per_iteration": 2.6576426029205322
    },
    {
      "auxiliary_loss_clip": 0.0114764,
      "auxiliary_loss_mlp": 0.01027208,
      "balance_loss_clip": 1.05090714,
      "balance_loss_mlp": 1.01967132,
      "epoch": 0.6007334816328984,
      "flos": 22858542852480.0,
      "grad_norm": 1.5962312636885647,
      "language_loss": 0.77501786,
      "learning_rate": 1.4517044913095938e-06,
      "loss": 0.79676634,
      "num_input_tokens_seen": 107717790,
      "step": 4996,
      "time_per_iteration": 2.6031856536865234
    },
    {
      "auxiliary_loss_clip": 0.01162213,
      "auxiliary_loss_mlp": 0.0102866,
      "balance_loss_clip": 1.04840517,
      "balance_loss_mlp": 1.02079201,
      "epoch": 0.6008537245235376,
      "flos": 28324577047680.0,
      "grad_norm": 2.171677519191441,
      "language_loss": 0.81636995,
      "learning_rate": 1.4509554055552022e-06,
      "loss": 0.83827877,
      "num_input_tokens_seen": 107738020,
      "step": 4997,
      "time_per_iteration": 2.614917278289795
    },
    {
      "auxiliary_loss_clip": 0.01146737,
      "auxiliary_loss_mlp": 0.01030528,
      "balance_loss_clip": 1.04679215,
      "balance_loss_mlp": 1.02280068,
      "epoch": 0.6009739674141766,
      "flos": 20886113266560.0,
      "grad_norm": 2.234984670630521,
      "language_loss": 0.83850694,
      "learning_rate": 1.450206403090439e-06,
      "loss": 0.86027956,
      "num_input_tokens_seen": 107756215,
      "step": 4998,
      "time_per_iteration": 2.6205227375030518
    },
    {
      "auxiliary_loss_clip": 0.01159455,
      "auxiliary_loss_mlp": 0.01027715,
      "balance_loss_clip": 1.04985797,
      "balance_loss_mlp": 1.02055037,
      "epoch": 0.6010942103048157,
      "flos": 20481004702080.0,
      "grad_norm": 1.923039522096354,
      "language_loss": 0.86052781,
      "learning_rate": 1.4494574840289274e-06,
      "loss": 0.8823995,
      "num_input_tokens_seen": 107773330,
      "step": 4999,
      "time_per_iteration": 2.6754627227783203
    },
    {
      "auxiliary_loss_clip": 0.0116861,
      "auxiliary_loss_mlp": 0.01032258,
      "balance_loss_clip": 1.04965281,
      "balance_loss_mlp": 1.02378213,
      "epoch": 0.6012144531954549,
      "flos": 23805973935360.0,
      "grad_norm": 1.6824643289282981,
      "language_loss": 0.73743987,
      "learning_rate": 1.4487086484842782e-06,
      "loss": 0.75944853,
      "num_input_tokens_seen": 107791975,
      "step": 5000,
      "time_per_iteration": 2.7889316082000732
    },
    {
      "auxiliary_loss_clip": 0.01173817,
      "auxiliary_loss_mlp": 0.0102346,
      "balance_loss_clip": 1.04940665,
      "balance_loss_mlp": 1.01603937,
      "epoch": 0.6013346960860939,
      "flos": 18988378012800.0,
      "grad_norm": 1.909405005234516,
      "language_loss": 0.60089952,
      "learning_rate": 1.4479598965700878e-06,
      "loss": 0.62287229,
      "num_input_tokens_seen": 107809240,
      "step": 5001,
      "time_per_iteration": 3.501450538635254
    },
    {
      "auxiliary_loss_clip": 0.01130531,
      "auxiliary_loss_mlp": 0.01029853,
      "balance_loss_clip": 1.04376078,
      "balance_loss_mlp": 1.02243805,
      "epoch": 0.601454938976733,
      "flos": 24025316336640.0,
      "grad_norm": 2.1552275029686183,
      "language_loss": 0.69116771,
      "learning_rate": 1.4472112283999427e-06,
      "loss": 0.71277153,
      "num_input_tokens_seen": 107827895,
      "step": 5002,
      "time_per_iteration": 2.737783908843994
    },
    {
      "auxiliary_loss_clip": 0.01160277,
      "auxiliary_loss_mlp": 0.01026484,
      "balance_loss_clip": 1.05239105,
      "balance_loss_mlp": 1.0189743,
      "epoch": 0.6015751818673721,
      "flos": 26427129102720.0,
      "grad_norm": 2.1411522917084462,
      "language_loss": 0.68869638,
      "learning_rate": 1.4464626440874143e-06,
      "loss": 0.71056396,
      "num_input_tokens_seen": 107847010,
      "step": 5003,
      "time_per_iteration": 3.5529987812042236
    },
    {
      "auxiliary_loss_clip": 0.01127504,
      "auxiliary_loss_mlp": 0.01028814,
      "balance_loss_clip": 1.0415628,
      "balance_loss_mlp": 1.02068698,
      "epoch": 0.6016954247580112,
      "flos": 13115260005120.0,
      "grad_norm": 2.421227621092797,
      "language_loss": 0.74458504,
      "learning_rate": 1.4457141437460636e-06,
      "loss": 0.76614821,
      "num_input_tokens_seen": 107864235,
      "step": 5004,
      "time_per_iteration": 3.5994081497192383
    },
    {
      "auxiliary_loss_clip": 0.01147862,
      "auxiliary_loss_mlp": 0.01031073,
      "balance_loss_clip": 1.04659247,
      "balance_loss_mlp": 1.02281165,
      "epoch": 0.6018156676486502,
      "flos": 23768447201280.0,
      "grad_norm": 1.9783213232247108,
      "language_loss": 0.73283482,
      "learning_rate": 1.444965727489436e-06,
      "loss": 0.75462419,
      "num_input_tokens_seen": 107883680,
      "step": 5005,
      "time_per_iteration": 2.5895891189575195
    },
    {
      "auxiliary_loss_clip": 0.01129648,
      "auxiliary_loss_mlp": 0.01029378,
      "balance_loss_clip": 1.04296708,
      "balance_loss_mlp": 1.02116489,
      "epoch": 0.6019359105392894,
      "flos": 26469360518400.0,
      "grad_norm": 2.368925015647772,
      "language_loss": 0.63183415,
      "learning_rate": 1.444217395431066e-06,
      "loss": 0.65342438,
      "num_input_tokens_seen": 107906220,
      "step": 5006,
      "time_per_iteration": 3.4323251247406006
    },
    {
      "auxiliary_loss_clip": 0.01033215,
      "auxiliary_loss_mlp": 0.01001745,
      "balance_loss_clip": 1.00969291,
      "balance_loss_mlp": 1.000458,
      "epoch": 0.6020561534299285,
      "flos": 69190849728000.0,
      "grad_norm": 0.8309960699152448,
      "language_loss": 0.55832732,
      "learning_rate": 1.4434691476844755e-06,
      "loss": 0.57867694,
      "num_input_tokens_seen": 107967195,
      "step": 5007,
      "time_per_iteration": 3.2029309272766113
    },
    {
      "auxiliary_loss_clip": 0.01146857,
      "auxiliary_loss_mlp": 0.01029797,
      "balance_loss_clip": 1.05058956,
      "balance_loss_mlp": 1.02233434,
      "epoch": 0.6021763963205675,
      "flos": 21835304115840.0,
      "grad_norm": 2.2505375170237847,
      "language_loss": 0.66802537,
      "learning_rate": 1.4427209843631729e-06,
      "loss": 0.68979192,
      "num_input_tokens_seen": 107984245,
      "step": 5008,
      "time_per_iteration": 2.703042984008789
    },
    {
      "auxiliary_loss_clip": 0.01175423,
      "auxiliary_loss_mlp": 0.00762405,
      "balance_loss_clip": 1.05111337,
      "balance_loss_mlp": 1.00055385,
      "epoch": 0.6022966392112067,
      "flos": 26578636669440.0,
      "grad_norm": 1.7959321209807007,
      "language_loss": 0.80555946,
      "learning_rate": 1.4419729055806534e-06,
      "loss": 0.82493782,
      "num_input_tokens_seen": 108003680,
      "step": 5009,
      "time_per_iteration": 2.637044668197632
    },
    {
      "auxiliary_loss_clip": 0.01141918,
      "auxiliary_loss_mlp": 0.00762525,
      "balance_loss_clip": 1.04756641,
      "balance_loss_mlp": 1.00039804,
      "epoch": 0.6024168821018457,
      "flos": 20703722981760.0,
      "grad_norm": 1.7704001005201455,
      "language_loss": 0.82227659,
      "learning_rate": 1.441224911450401e-06,
      "loss": 0.84132105,
      "num_input_tokens_seen": 108019635,
      "step": 5010,
      "time_per_iteration": 2.773890256881714
    },
    {
      "auxiliary_loss_clip": 0.01167267,
      "auxiliary_loss_mlp": 0.01028078,
      "balance_loss_clip": 1.05093133,
      "balance_loss_mlp": 1.0205915,
      "epoch": 0.6025371249924848,
      "flos": 24680973242880.0,
      "grad_norm": 1.7712176978128895,
      "language_loss": 0.82064855,
      "learning_rate": 1.4404770020858851e-06,
      "loss": 0.84260201,
      "num_input_tokens_seen": 108039120,
      "step": 5011,
      "time_per_iteration": 2.675382137298584
    },
    {
      "auxiliary_loss_clip": 0.01156024,
      "auxiliary_loss_mlp": 0.01028804,
      "balance_loss_clip": 1.0477078,
      "balance_loss_mlp": 1.02049494,
      "epoch": 0.602657367883124,
      "flos": 25955801815680.0,
      "grad_norm": 1.7996071502404007,
      "language_loss": 0.85822195,
      "learning_rate": 1.439729177600563e-06,
      "loss": 0.88007027,
      "num_input_tokens_seen": 108059615,
      "step": 5012,
      "time_per_iteration": 2.691541910171509
    },
    {
      "auxiliary_loss_clip": 0.01159479,
      "auxiliary_loss_mlp": 0.01027763,
      "balance_loss_clip": 1.04908085,
      "balance_loss_mlp": 1.02044332,
      "epoch": 0.602777610773763,
      "flos": 16690633925760.0,
      "grad_norm": 1.8141227878972637,
      "language_loss": 0.73402786,
      "learning_rate": 1.4389814381078793e-06,
      "loss": 0.75590026,
      "num_input_tokens_seen": 108078855,
      "step": 5013,
      "time_per_iteration": 2.692416191101074
    },
    {
      "auxiliary_loss_clip": 0.01068671,
      "auxiliary_loss_mlp": 0.01030576,
      "balance_loss_clip": 1.04226673,
      "balance_loss_mlp": 1.02327406,
      "epoch": 0.6028978536644021,
      "flos": 13334243270400.0,
      "grad_norm": 2.1902465090951853,
      "language_loss": 0.79799104,
      "learning_rate": 1.438233783721265e-06,
      "loss": 0.81898355,
      "num_input_tokens_seen": 108095020,
      "step": 5014,
      "time_per_iteration": 3.179271936416626
    },
    {
      "auxiliary_loss_clip": 0.01144687,
      "auxiliary_loss_mlp": 0.01030523,
      "balance_loss_clip": 1.05154192,
      "balance_loss_mlp": 1.02315021,
      "epoch": 0.6030180965550412,
      "flos": 19644825018240.0,
      "grad_norm": 3.5764874742502526,
      "language_loss": 0.78025657,
      "learning_rate": 1.43748621455414e-06,
      "loss": 0.80200863,
      "num_input_tokens_seen": 108111455,
      "step": 5015,
      "time_per_iteration": 3.1983656883239746
    },
    {
      "auxiliary_loss_clip": 0.01146933,
      "auxiliary_loss_mlp": 0.01027396,
      "balance_loss_clip": 1.04933667,
      "balance_loss_mlp": 1.01962352,
      "epoch": 0.6031383394456803,
      "flos": 14458390289280.0,
      "grad_norm": 2.8112249118798,
      "language_loss": 0.80876547,
      "learning_rate": 1.4367387307199082e-06,
      "loss": 0.83050877,
      "num_input_tokens_seen": 108128305,
      "step": 5016,
      "time_per_iteration": 2.7225799560546875
    },
    {
      "auxiliary_loss_clip": 0.01154943,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.04549909,
      "balance_loss_mlp": 1.02070808,
      "epoch": 0.6032585823363193,
      "flos": 13917791623680.0,
      "grad_norm": 1.8663937249686873,
      "language_loss": 0.82447815,
      "learning_rate": 1.4359913323319632e-06,
      "loss": 0.84630978,
      "num_input_tokens_seen": 108145475,
      "step": 5017,
      "time_per_iteration": 2.6768887042999268
    },
    {
      "auxiliary_loss_clip": 0.01093841,
      "auxiliary_loss_mlp": 0.01022309,
      "balance_loss_clip": 1.03929496,
      "balance_loss_mlp": 1.01433432,
      "epoch": 0.6033788252269584,
      "flos": 24353252530560.0,
      "grad_norm": 1.7472455463410574,
      "language_loss": 0.77557904,
      "learning_rate": 1.4352440195036847e-06,
      "loss": 0.79674053,
      "num_input_tokens_seen": 108165650,
      "step": 5018,
      "time_per_iteration": 3.2117042541503906
    },
    {
      "auxiliary_loss_clip": 0.0109535,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.0391748,
      "balance_loss_mlp": 1.01880825,
      "epoch": 0.6034990681175976,
      "flos": 25521247077120.0,
      "grad_norm": 1.5560127143919007,
      "language_loss": 0.80187523,
      "learning_rate": 1.4344967923484395e-06,
      "loss": 0.82309532,
      "num_input_tokens_seen": 108187620,
      "step": 5019,
      "time_per_iteration": 3.056807041168213
    },
    {
      "auxiliary_loss_clip": 0.01157242,
      "auxiliary_loss_mlp": 0.01026494,
      "balance_loss_clip": 1.04708815,
      "balance_loss_mlp": 1.01885271,
      "epoch": 0.6036193110082366,
      "flos": 25958387594880.0,
      "grad_norm": 2.72340220139867,
      "language_loss": 0.7199797,
      "learning_rate": 1.433749650979581e-06,
      "loss": 0.74181706,
      "num_input_tokens_seen": 108207605,
      "step": 5020,
      "time_per_iteration": 3.377457857131958
    },
    {
      "auxiliary_loss_clip": 0.01136726,
      "auxiliary_loss_mlp": 0.01026435,
      "balance_loss_clip": 1.0458076,
      "balance_loss_mlp": 1.01943731,
      "epoch": 0.6037395538988757,
      "flos": 25593427457280.0,
      "grad_norm": 1.9977401522236844,
      "language_loss": 0.68038237,
      "learning_rate": 1.433002595510451e-06,
      "loss": 0.70201397,
      "num_input_tokens_seen": 108226385,
      "step": 5021,
      "time_per_iteration": 2.830561637878418
    },
    {
      "auxiliary_loss_clip": 0.01142272,
      "auxiliary_loss_mlp": 0.0076261,
      "balance_loss_clip": 1.0456481,
      "balance_loss_mlp": 1.00040078,
      "epoch": 0.6038597967895148,
      "flos": 17816253402240.0,
      "grad_norm": 1.9160488157094748,
      "language_loss": 0.7213906,
      "learning_rate": 1.4322556260543757e-06,
      "loss": 0.74043936,
      "num_input_tokens_seen": 108242960,
      "step": 5022,
      "time_per_iteration": 2.7014102935791016
    },
    {
      "auxiliary_loss_clip": 0.01039405,
      "auxiliary_loss_mlp": 0.01002491,
      "balance_loss_clip": 1.01143622,
      "balance_loss_mlp": 1.00112629,
      "epoch": 0.6039800396801539,
      "flos": 65169213235200.0,
      "grad_norm": 0.8920757920305581,
      "language_loss": 0.62775153,
      "learning_rate": 1.4315087427246703e-06,
      "loss": 0.64817047,
      "num_input_tokens_seen": 108296785,
      "step": 5023,
      "time_per_iteration": 3.167895793914795
    },
    {
      "auxiliary_loss_clip": 0.01070121,
      "auxiliary_loss_mlp": 0.01001596,
      "balance_loss_clip": 1.01145077,
      "balance_loss_mlp": 1.00029612,
      "epoch": 0.604100282570793,
      "flos": 67386409073280.0,
      "grad_norm": 0.871283349322447,
      "language_loss": 0.5850395,
      "learning_rate": 1.4307619456346372e-06,
      "loss": 0.60575664,
      "num_input_tokens_seen": 108341090,
      "step": 5024,
      "time_per_iteration": 2.9324069023132324
    },
    {
      "auxiliary_loss_clip": 0.01164063,
      "auxiliary_loss_mlp": 0.01029076,
      "balance_loss_clip": 1.04659295,
      "balance_loss_mlp": 1.02074945,
      "epoch": 0.6042205254614321,
      "flos": 35297495631360.0,
      "grad_norm": 1.7358304422334627,
      "language_loss": 0.74176627,
      "learning_rate": 1.430015234897564e-06,
      "loss": 0.76369762,
      "num_input_tokens_seen": 108364370,
      "step": 5025,
      "time_per_iteration": 2.809781074523926
    },
    {
      "auxiliary_loss_clip": 0.01172838,
      "auxiliary_loss_mlp": 0.00762664,
      "balance_loss_clip": 1.04788756,
      "balance_loss_mlp": 1.00050139,
      "epoch": 0.6043407683520712,
      "flos": 45658262206080.0,
      "grad_norm": 1.7952190703201352,
      "language_loss": 0.66457057,
      "learning_rate": 1.4292686106267274e-06,
      "loss": 0.68392557,
      "num_input_tokens_seen": 108387220,
      "step": 5026,
      "time_per_iteration": 2.8330507278442383
    },
    {
      "auxiliary_loss_clip": 0.0116611,
      "auxiliary_loss_mlp": 0.01026289,
      "balance_loss_clip": 1.04958248,
      "balance_loss_mlp": 1.01766372,
      "epoch": 0.6044610112427102,
      "flos": 16180020138240.0,
      "grad_norm": 1.5784952717699534,
      "language_loss": 0.77065784,
      "learning_rate": 1.4285220729353876e-06,
      "loss": 0.7925818,
      "num_input_tokens_seen": 108405760,
      "step": 5027,
      "time_per_iteration": 3.78674054145813
    },
    {
      "auxiliary_loss_clip": 0.01145114,
      "auxiliary_loss_mlp": 0.0102228,
      "balance_loss_clip": 1.04386473,
      "balance_loss_mlp": 1.01489472,
      "epoch": 0.6045812541333494,
      "flos": 13804062186240.0,
      "grad_norm": 1.8915746205369093,
      "language_loss": 0.77806807,
      "learning_rate": 1.4277756219367957e-06,
      "loss": 0.79974198,
      "num_input_tokens_seen": 108422785,
      "step": 5028,
      "time_per_iteration": 3.630660057067871
    },
    {
      "auxiliary_loss_clip": 0.01142081,
      "auxiliary_loss_mlp": 0.01027116,
      "balance_loss_clip": 1.04860473,
      "balance_loss_mlp": 1.0191108,
      "epoch": 0.6047014970239885,
      "flos": 19975059682560.0,
      "grad_norm": 2.1383705176383923,
      "language_loss": 0.7957648,
      "learning_rate": 1.4270292577441864e-06,
      "loss": 0.81745678,
      "num_input_tokens_seen": 108442290,
      "step": 5029,
      "time_per_iteration": 3.69783878326416
    },
    {
      "auxiliary_loss_clip": 0.01166192,
      "auxiliary_loss_mlp": 0.01022857,
      "balance_loss_clip": 1.04838204,
      "balance_loss_mlp": 1.01523936,
      "epoch": 0.6048217399146275,
      "flos": 25337097025920.0,
      "grad_norm": 1.8633701952750807,
      "language_loss": 0.72020757,
      "learning_rate": 1.4262829804707836e-06,
      "loss": 0.74209809,
      "num_input_tokens_seen": 108464280,
      "step": 5030,
      "time_per_iteration": 2.735646963119507
    },
    {
      "auxiliary_loss_clip": 0.01162327,
      "auxiliary_loss_mlp": 0.01027867,
      "balance_loss_clip": 1.04591656,
      "balance_loss_mlp": 1.01946902,
      "epoch": 0.6049419828052667,
      "flos": 26030819370240.0,
      "grad_norm": 1.436724671093019,
      "language_loss": 0.69779861,
      "learning_rate": 1.4255367902297958e-06,
      "loss": 0.71970057,
      "num_input_tokens_seen": 108485610,
      "step": 5031,
      "time_per_iteration": 2.6805918216705322
    },
    {
      "auxiliary_loss_clip": 0.01173022,
      "auxiliary_loss_mlp": 0.01028795,
      "balance_loss_clip": 1.04991245,
      "balance_loss_mlp": 1.02146673,
      "epoch": 0.6050622256959057,
      "flos": 14648106948480.0,
      "grad_norm": 2.3174314075281623,
      "language_loss": 0.78477192,
      "learning_rate": 1.4247906871344215e-06,
      "loss": 0.80679011,
      "num_input_tokens_seen": 108501005,
      "step": 5032,
      "time_per_iteration": 3.5489003658294678
    },
    {
      "auxiliary_loss_clip": 0.01142258,
      "auxiliary_loss_mlp": 0.0102801,
      "balance_loss_clip": 1.04445446,
      "balance_loss_mlp": 1.02055359,
      "epoch": 0.6051824685865448,
      "flos": 23331450337920.0,
      "grad_norm": 2.219451323212368,
      "language_loss": 0.75520462,
      "learning_rate": 1.4240446712978415e-06,
      "loss": 0.77690732,
      "num_input_tokens_seen": 108519990,
      "step": 5033,
      "time_per_iteration": 2.7841176986694336
    },
    {
      "auxiliary_loss_clip": 0.01163547,
      "auxiliary_loss_mlp": 0.01031102,
      "balance_loss_clip": 1.04678488,
      "balance_loss_mlp": 1.02282894,
      "epoch": 0.605302711477184,
      "flos": 27563307177600.0,
      "grad_norm": 2.580800873220544,
      "language_loss": 0.74725306,
      "learning_rate": 1.423298742833227e-06,
      "loss": 0.76919961,
      "num_input_tokens_seen": 108538650,
      "step": 5034,
      "time_per_iteration": 2.6969282627105713
    },
    {
      "auxiliary_loss_clip": 0.01138149,
      "auxiliary_loss_mlp": 0.0102633,
      "balance_loss_clip": 1.04428577,
      "balance_loss_mlp": 1.01853657,
      "epoch": 0.605422954367823,
      "flos": 15154698412800.0,
      "grad_norm": 2.1444130029078923,
      "language_loss": 0.7167697,
      "learning_rate": 1.4225529018537352e-06,
      "loss": 0.73841447,
      "num_input_tokens_seen": 108554155,
      "step": 5035,
      "time_per_iteration": 2.683643341064453
    },
    {
      "auxiliary_loss_clip": 0.01173727,
      "auxiliary_loss_mlp": 0.01029431,
      "balance_loss_clip": 1.04915404,
      "balance_loss_mlp": 1.02258897,
      "epoch": 0.6055431972584621,
      "flos": 27673912131840.0,
      "grad_norm": 1.516246265500322,
      "language_loss": 0.77885175,
      "learning_rate": 1.4218071484725082e-06,
      "loss": 0.80088329,
      "num_input_tokens_seen": 108576275,
      "step": 5036,
      "time_per_iteration": 2.6514716148376465
    },
    {
      "auxiliary_loss_clip": 0.01145106,
      "auxiliary_loss_mlp": 0.01031715,
      "balance_loss_clip": 1.04908001,
      "balance_loss_mlp": 1.02429152,
      "epoch": 0.6056634401491012,
      "flos": 19387489006080.0,
      "grad_norm": 2.0720834826300414,
      "language_loss": 0.76459736,
      "learning_rate": 1.4210614828026786e-06,
      "loss": 0.78636557,
      "num_input_tokens_seen": 108594125,
      "step": 5037,
      "time_per_iteration": 2.6622183322906494
    },
    {
      "auxiliary_loss_clip": 0.01171795,
      "auxiliary_loss_mlp": 0.01024076,
      "balance_loss_clip": 1.04733539,
      "balance_loss_mlp": 1.01644039,
      "epoch": 0.6057836830397403,
      "flos": 24789459294720.0,
      "grad_norm": 1.524051763954337,
      "language_loss": 0.7449702,
      "learning_rate": 1.4203159049573605e-06,
      "loss": 0.76692891,
      "num_input_tokens_seen": 108615360,
      "step": 5038,
      "time_per_iteration": 2.600682258605957
    },
    {
      "auxiliary_loss_clip": 0.0115294,
      "auxiliary_loss_mlp": 0.01027105,
      "balance_loss_clip": 1.04479194,
      "balance_loss_mlp": 1.01910925,
      "epoch": 0.6059039259303793,
      "flos": 20558248899840.0,
      "grad_norm": 2.9608739983378656,
      "language_loss": 0.87261182,
      "learning_rate": 1.4195704150496593e-06,
      "loss": 0.89441228,
      "num_input_tokens_seen": 108633075,
      "step": 5039,
      "time_per_iteration": 2.6481616497039795
    },
    {
      "auxiliary_loss_clip": 0.01148466,
      "auxiliary_loss_mlp": 0.01024184,
      "balance_loss_clip": 1.04885888,
      "balance_loss_mlp": 1.01651621,
      "epoch": 0.6060241688210185,
      "flos": 21069724613760.0,
      "grad_norm": 1.7074773104448981,
      "language_loss": 0.73631048,
      "learning_rate": 1.4188250131926639e-06,
      "loss": 0.75803697,
      "num_input_tokens_seen": 108651875,
      "step": 5040,
      "time_per_iteration": 2.671645164489746
    },
    {
      "auxiliary_loss_clip": 0.01147493,
      "auxiliary_loss_mlp": 0.01029991,
      "balance_loss_clip": 1.04692221,
      "balance_loss_mlp": 1.02231133,
      "epoch": 0.6061444117116576,
      "flos": 16361081619840.0,
      "grad_norm": 1.7665053704913436,
      "language_loss": 0.80622667,
      "learning_rate": 1.4180796994994525e-06,
      "loss": 0.8280015,
      "num_input_tokens_seen": 108669290,
      "step": 5041,
      "time_per_iteration": 2.643491744995117
    },
    {
      "auxiliary_loss_clip": 0.01143462,
      "auxiliary_loss_mlp": 0.01025126,
      "balance_loss_clip": 1.04528451,
      "balance_loss_mlp": 1.01740694,
      "epoch": 0.6062646546022966,
      "flos": 21507296094720.0,
      "grad_norm": 1.7459314059254187,
      "language_loss": 0.71880239,
      "learning_rate": 1.4173344740830877e-06,
      "loss": 0.74048823,
      "num_input_tokens_seen": 108688420,
      "step": 5042,
      "time_per_iteration": 2.678483724594116
    },
    {
      "auxiliary_loss_clip": 0.01142361,
      "auxiliary_loss_mlp": 0.01028009,
      "balance_loss_clip": 1.0493567,
      "balance_loss_mlp": 1.02086878,
      "epoch": 0.6063848974929358,
      "flos": 38983151283840.0,
      "grad_norm": 1.4407724471679442,
      "language_loss": 0.704252,
      "learning_rate": 1.4165893370566206e-06,
      "loss": 0.72595561,
      "num_input_tokens_seen": 108712175,
      "step": 5043,
      "time_per_iteration": 2.8046371936798096
    },
    {
      "auxiliary_loss_clip": 0.01156498,
      "auxiliary_loss_mlp": 0.01031443,
      "balance_loss_clip": 1.04605865,
      "balance_loss_mlp": 1.02392054,
      "epoch": 0.6065051403835748,
      "flos": 19646584784640.0,
      "grad_norm": 1.8692023053347995,
      "language_loss": 0.77849877,
      "learning_rate": 1.4158442885330865e-06,
      "loss": 0.8003782,
      "num_input_tokens_seen": 108730745,
      "step": 5044,
      "time_per_iteration": 2.7018954753875732
    },
    {
      "auxiliary_loss_clip": 0.01154779,
      "auxiliary_loss_mlp": 0.01027086,
      "balance_loss_clip": 1.04605782,
      "balance_loss_mlp": 1.01946294,
      "epoch": 0.6066253832742139,
      "flos": 23513086437120.0,
      "grad_norm": 2.109540476549584,
      "language_loss": 0.78734219,
      "learning_rate": 1.4150993286255094e-06,
      "loss": 0.80916083,
      "num_input_tokens_seen": 108749995,
      "step": 5045,
      "time_per_iteration": 2.67759108543396
    },
    {
      "auxiliary_loss_clip": 0.01174402,
      "auxiliary_loss_mlp": 0.0102799,
      "balance_loss_clip": 1.04901123,
      "balance_loss_mlp": 1.02096295,
      "epoch": 0.6067456261648531,
      "flos": 19133708440320.0,
      "grad_norm": 1.8607932955318516,
      "language_loss": 0.79663706,
      "learning_rate": 1.4143544574468993e-06,
      "loss": 0.81866097,
      "num_input_tokens_seen": 108768355,
      "step": 5046,
      "time_per_iteration": 2.552079200744629
    },
    {
      "auxiliary_loss_clip": 0.01157104,
      "auxiliary_loss_mlp": 0.01024584,
      "balance_loss_clip": 1.04752088,
      "balance_loss_mlp": 1.01752138,
      "epoch": 0.6068658690554921,
      "flos": 20520614424960.0,
      "grad_norm": 1.7135791702772045,
      "language_loss": 0.82690108,
      "learning_rate": 1.4136096751102523e-06,
      "loss": 0.84871793,
      "num_input_tokens_seen": 108786685,
      "step": 5047,
      "time_per_iteration": 2.6374855041503906
    },
    {
      "auxiliary_loss_clip": 0.01148636,
      "auxiliary_loss_mlp": 0.01026794,
      "balance_loss_clip": 1.0469892,
      "balance_loss_mlp": 1.01933193,
      "epoch": 0.6069861119461312,
      "flos": 27374560185600.0,
      "grad_norm": 2.3090417849371567,
      "language_loss": 0.82972741,
      "learning_rate": 1.4128649817285516e-06,
      "loss": 0.8514818,
      "num_input_tokens_seen": 108804820,
      "step": 5048,
      "time_per_iteration": 2.7611138820648193
    },
    {
      "auxiliary_loss_clip": 0.01148089,
      "auxiliary_loss_mlp": 0.01031954,
      "balance_loss_clip": 1.04520452,
      "balance_loss_mlp": 1.02516532,
      "epoch": 0.6071063548367702,
      "flos": 25626500904960.0,
      "grad_norm": 1.7521350697924771,
      "language_loss": 0.63174343,
      "learning_rate": 1.412120377414766e-06,
      "loss": 0.65354389,
      "num_input_tokens_seen": 108825010,
      "step": 5049,
      "time_per_iteration": 2.704341173171997
    },
    {
      "auxiliary_loss_clip": 0.01174819,
      "auxiliary_loss_mlp": 0.01025709,
      "balance_loss_clip": 1.05021524,
      "balance_loss_mlp": 1.01901865,
      "epoch": 0.6072265977274094,
      "flos": 24460517520000.0,
      "grad_norm": 1.6190112310216156,
      "language_loss": 0.71375883,
      "learning_rate": 1.4113758622818522e-06,
      "loss": 0.73576403,
      "num_input_tokens_seen": 108845075,
      "step": 5050,
      "time_per_iteration": 2.6372625827789307
    },
    {
      "auxiliary_loss_clip": 0.01153387,
      "auxiliary_loss_mlp": 0.00762509,
      "balance_loss_clip": 1.04972267,
      "balance_loss_mlp": 1.00047648,
      "epoch": 0.6073468406180484,
      "flos": 18149253413760.0,
      "grad_norm": 1.9603853427703537,
      "language_loss": 0.83132088,
      "learning_rate": 1.410631436442751e-06,
      "loss": 0.85047984,
      "num_input_tokens_seen": 108863870,
      "step": 5051,
      "time_per_iteration": 2.694211483001709
    },
    {
      "auxiliary_loss_clip": 0.01164616,
      "auxiliary_loss_mlp": 0.01029381,
      "balance_loss_clip": 1.04818773,
      "balance_loss_mlp": 1.02175772,
      "epoch": 0.6074670835086875,
      "flos": 20697617669760.0,
      "grad_norm": 2.453851160473683,
      "language_loss": 0.86446846,
      "learning_rate": 1.4098871000103936e-06,
      "loss": 0.88640845,
      "num_input_tokens_seen": 108882470,
      "step": 5052,
      "time_per_iteration": 2.6802048683166504
    },
    {
      "auxiliary_loss_clip": 0.01147787,
      "auxiliary_loss_mlp": 0.01026399,
      "balance_loss_clip": 1.04683042,
      "balance_loss_mlp": 1.01904345,
      "epoch": 0.6075873263993267,
      "flos": 23769955572480.0,
      "grad_norm": 1.787758566022627,
      "language_loss": 0.82585704,
      "learning_rate": 1.409142853097693e-06,
      "loss": 0.84759879,
      "num_input_tokens_seen": 108902710,
      "step": 5053,
      "time_per_iteration": 3.6108081340789795
    },
    {
      "auxiliary_loss_clip": 0.01150986,
      "auxiliary_loss_mlp": 0.01026829,
      "balance_loss_clip": 1.05033302,
      "balance_loss_mlp": 1.01953888,
      "epoch": 0.6077075692899657,
      "flos": 24454484035200.0,
      "grad_norm": 2.344193048659633,
      "language_loss": 0.79538631,
      "learning_rate": 1.408398695817553e-06,
      "loss": 0.81716442,
      "num_input_tokens_seen": 108919935,
      "step": 5054,
      "time_per_iteration": 2.7133491039276123
    },
    {
      "auxiliary_loss_clip": 0.0114766,
      "auxiliary_loss_mlp": 0.01029482,
      "balance_loss_clip": 1.04584837,
      "balance_loss_mlp": 1.02116668,
      "epoch": 0.6078278121806048,
      "flos": 27382102041600.0,
      "grad_norm": 1.8877527791977893,
      "language_loss": 0.70078266,
      "learning_rate": 1.4076546282828593e-06,
      "loss": 0.72255403,
      "num_input_tokens_seen": 108942790,
      "step": 5055,
      "time_per_iteration": 3.7269246578216553
    },
    {
      "auxiliary_loss_clip": 0.01149758,
      "auxiliary_loss_mlp": 0.01028353,
      "balance_loss_clip": 1.04358816,
      "balance_loss_mlp": 1.02080429,
      "epoch": 0.6079480550712439,
      "flos": 38436447306240.0,
      "grad_norm": 2.249194779709867,
      "language_loss": 0.6628319,
      "learning_rate": 1.4069106506064874e-06,
      "loss": 0.68461305,
      "num_input_tokens_seen": 108964215,
      "step": 5056,
      "time_per_iteration": 3.8192877769470215
    },
    {
      "auxiliary_loss_clip": 0.01143703,
      "auxiliary_loss_mlp": 0.01029376,
      "balance_loss_clip": 1.04892921,
      "balance_loss_mlp": 1.02214003,
      "epoch": 0.608068297961883,
      "flos": 25336271013120.0,
      "grad_norm": 1.7641820975107054,
      "language_loss": 0.7845726,
      "learning_rate": 1.4061667629012989e-06,
      "loss": 0.80630344,
      "num_input_tokens_seen": 108984885,
      "step": 5057,
      "time_per_iteration": 2.7895429134368896
    },
    {
      "auxiliary_loss_clip": 0.01137217,
      "auxiliary_loss_mlp": 0.01026114,
      "balance_loss_clip": 1.04505432,
      "balance_loss_mlp": 1.01847243,
      "epoch": 0.608188540852522,
      "flos": 24202463235840.0,
      "grad_norm": 1.6356214173453085,
      "language_loss": 0.8318857,
      "learning_rate": 1.40542296528014e-06,
      "loss": 0.85351896,
      "num_input_tokens_seen": 109004545,
      "step": 5058,
      "time_per_iteration": 3.6207730770111084
    },
    {
      "auxiliary_loss_clip": 0.01161907,
      "auxiliary_loss_mlp": 0.01030035,
      "balance_loss_clip": 1.04841208,
      "balance_loss_mlp": 1.02209568,
      "epoch": 0.6083087837431612,
      "flos": 21284146851840.0,
      "grad_norm": 2.321070280158195,
      "language_loss": 0.76475817,
      "learning_rate": 1.4046792578558452e-06,
      "loss": 0.7866776,
      "num_input_tokens_seen": 109022440,
      "step": 5059,
      "time_per_iteration": 2.5691776275634766
    },
    {
      "auxiliary_loss_clip": 0.01143611,
      "auxiliary_loss_mlp": 0.01026184,
      "balance_loss_clip": 1.04632556,
      "balance_loss_mlp": 1.01861715,
      "epoch": 0.6084290266338003,
      "flos": 16471435178880.0,
      "grad_norm": 2.271413106923837,
      "language_loss": 0.76071185,
      "learning_rate": 1.4039356407412325e-06,
      "loss": 0.78240979,
      "num_input_tokens_seen": 109035680,
      "step": 5060,
      "time_per_iteration": 2.6533749103546143
    },
    {
      "auxiliary_loss_clip": 0.0106177,
      "auxiliary_loss_mlp": 0.01001418,
      "balance_loss_clip": 1.01234055,
      "balance_loss_mlp": 1.00008917,
      "epoch": 0.6085492695244393,
      "flos": 66443574931200.0,
      "grad_norm": 0.7748442699825008,
      "language_loss": 0.57030928,
      "learning_rate": 1.40319211404911e-06,
      "loss": 0.59094119,
      "num_input_tokens_seen": 109090680,
      "step": 5061,
      "time_per_iteration": 3.1475319862365723
    },
    {
      "auxiliary_loss_clip": 0.0117584,
      "auxiliary_loss_mlp": 0.01028767,
      "balance_loss_clip": 1.05060458,
      "balance_loss_mlp": 1.02113783,
      "epoch": 0.6086695124150785,
      "flos": 23618986709760.0,
      "grad_norm": 1.9276146166952546,
      "language_loss": 0.90525162,
      "learning_rate": 1.4024486778922691e-06,
      "loss": 0.92729777,
      "num_input_tokens_seen": 109108995,
      "step": 5062,
      "time_per_iteration": 2.609812021255493
    },
    {
      "auxiliary_loss_clip": 0.0115025,
      "auxiliary_loss_mlp": 0.01025512,
      "balance_loss_clip": 1.04448915,
      "balance_loss_mlp": 1.01795375,
      "epoch": 0.6087897553057176,
      "flos": 20157054917760.0,
      "grad_norm": 1.842942928713752,
      "language_loss": 0.77807277,
      "learning_rate": 1.4017053323834884e-06,
      "loss": 0.79983038,
      "num_input_tokens_seen": 109128825,
      "step": 5063,
      "time_per_iteration": 2.672015905380249
    },
    {
      "auxiliary_loss_clip": 0.01149884,
      "auxiliary_loss_mlp": 0.01021811,
      "balance_loss_clip": 1.04573703,
      "balance_loss_mlp": 1.01462245,
      "epoch": 0.6089099981963566,
      "flos": 25482535194240.0,
      "grad_norm": 1.7970566532738677,
      "language_loss": 0.75936437,
      "learning_rate": 1.4009620776355333e-06,
      "loss": 0.78108132,
      "num_input_tokens_seen": 109150425,
      "step": 5064,
      "time_per_iteration": 2.6676270961761475
    },
    {
      "auxiliary_loss_clip": 0.01157104,
      "auxiliary_loss_mlp": 0.01026599,
      "balance_loss_clip": 1.04716527,
      "balance_loss_mlp": 1.01939559,
      "epoch": 0.6090302410869958,
      "flos": 25332895134720.0,
      "grad_norm": 2.594178801401346,
      "language_loss": 0.7913332,
      "learning_rate": 1.4002189137611553e-06,
      "loss": 0.81317019,
      "num_input_tokens_seen": 109169765,
      "step": 5065,
      "time_per_iteration": 2.722817897796631
    },
    {
      "auxiliary_loss_clip": 0.01158854,
      "auxiliary_loss_mlp": 0.01026248,
      "balance_loss_clip": 1.04678607,
      "balance_loss_mlp": 1.01905417,
      "epoch": 0.6091504839776348,
      "flos": 23987358639360.0,
      "grad_norm": 1.8833345537736377,
      "language_loss": 0.69624871,
      "learning_rate": 1.3994758408730901e-06,
      "loss": 0.71809971,
      "num_input_tokens_seen": 109188950,
      "step": 5066,
      "time_per_iteration": 2.6628191471099854
    },
    {
      "auxiliary_loss_clip": 0.01149809,
      "auxiliary_loss_mlp": 0.01028617,
      "balance_loss_clip": 1.049577,
      "balance_loss_mlp": 1.01989055,
      "epoch": 0.6092707268682739,
      "flos": 29643037666560.0,
      "grad_norm": 2.162622121316223,
      "language_loss": 0.77105331,
      "learning_rate": 1.3987328590840629e-06,
      "loss": 0.79283762,
      "num_input_tokens_seen": 109209895,
      "step": 5067,
      "time_per_iteration": 2.7814409732818604
    },
    {
      "auxiliary_loss_clip": 0.011569,
      "auxiliary_loss_mlp": 0.01024915,
      "balance_loss_clip": 1.04719818,
      "balance_loss_mlp": 1.01671338,
      "epoch": 0.609390969758913,
      "flos": 24024957200640.0,
      "grad_norm": 2.0301867093589117,
      "language_loss": 0.8616277,
      "learning_rate": 1.397989968506783e-06,
      "loss": 0.88344586,
      "num_input_tokens_seen": 109228905,
      "step": 5068,
      "time_per_iteration": 2.6547365188598633
    },
    {
      "auxiliary_loss_clip": 0.01181412,
      "auxiliary_loss_mlp": 0.01027678,
      "balance_loss_clip": 1.0524075,
      "balance_loss_mlp": 1.0201081,
      "epoch": 0.6095112126495521,
      "flos": 11102143288320.0,
      "grad_norm": 2.1621678281581276,
      "language_loss": 0.72205263,
      "learning_rate": 1.3972471692539458e-06,
      "loss": 0.74414355,
      "num_input_tokens_seen": 109243620,
      "step": 5069,
      "time_per_iteration": 2.622789144515991
    },
    {
      "auxiliary_loss_clip": 0.01147022,
      "auxiliary_loss_mlp": 0.01029499,
      "balance_loss_clip": 1.05071831,
      "balance_loss_mlp": 1.02187002,
      "epoch": 0.6096314555401912,
      "flos": 17265491187840.0,
      "grad_norm": 2.031216404375011,
      "language_loss": 0.75300646,
      "learning_rate": 1.3965044614382348e-06,
      "loss": 0.77477163,
      "num_input_tokens_seen": 109259070,
      "step": 5070,
      "time_per_iteration": 2.7041192054748535
    },
    {
      "auxiliary_loss_clip": 0.01180223,
      "auxiliary_loss_mlp": 0.01030644,
      "balance_loss_clip": 1.05330431,
      "balance_loss_mlp": 1.02234149,
      "epoch": 0.6097516984308303,
      "flos": 21645910679040.0,
      "grad_norm": 2.7998282383892255,
      "language_loss": 0.75246078,
      "learning_rate": 1.3957618451723162e-06,
      "loss": 0.77456951,
      "num_input_tokens_seen": 109275100,
      "step": 5071,
      "time_per_iteration": 2.677096366882324
    },
    {
      "auxiliary_loss_clip": 0.01149189,
      "auxiliary_loss_mlp": 0.01034728,
      "balance_loss_clip": 1.047755,
      "balance_loss_mlp": 1.02676439,
      "epoch": 0.6098719413214694,
      "flos": 27199208966400.0,
      "grad_norm": 2.0534722133160237,
      "language_loss": 0.71625698,
      "learning_rate": 1.3950193205688457e-06,
      "loss": 0.73809612,
      "num_input_tokens_seen": 109294825,
      "step": 5072,
      "time_per_iteration": 2.692213535308838
    },
    {
      "auxiliary_loss_clip": 0.011461,
      "auxiliary_loss_mlp": 0.0102669,
      "balance_loss_clip": 1.04750872,
      "balance_loss_mlp": 1.01937652,
      "epoch": 0.6099921842121084,
      "flos": 20412954385920.0,
      "grad_norm": 1.7917431273889666,
      "language_loss": 0.83486664,
      "learning_rate": 1.3942768877404627e-06,
      "loss": 0.85659456,
      "num_input_tokens_seen": 109313790,
      "step": 5073,
      "time_per_iteration": 2.6535658836364746
    },
    {
      "auxiliary_loss_clip": 0.01176467,
      "auxiliary_loss_mlp": 0.0102377,
      "balance_loss_clip": 1.05078244,
      "balance_loss_mlp": 1.01623857,
      "epoch": 0.6101124271027476,
      "flos": 23366139897600.0,
      "grad_norm": 1.5057480141620876,
      "language_loss": 0.73475677,
      "learning_rate": 1.393534546799795e-06,
      "loss": 0.75675917,
      "num_input_tokens_seen": 109333490,
      "step": 5074,
      "time_per_iteration": 2.6030752658843994
    },
    {
      "auxiliary_loss_clip": 0.01138666,
      "auxiliary_loss_mlp": 0.01025661,
      "balance_loss_clip": 1.04588211,
      "balance_loss_mlp": 1.0178647,
      "epoch": 0.6102326699933867,
      "flos": 26687840993280.0,
      "grad_norm": 1.9068559341800866,
      "language_loss": 0.68138719,
      "learning_rate": 1.3927922978594536e-06,
      "loss": 0.70303047,
      "num_input_tokens_seen": 109354575,
      "step": 5075,
      "time_per_iteration": 2.6772894859313965
    },
    {
      "auxiliary_loss_clip": 0.01055281,
      "auxiliary_loss_mlp": 0.01001862,
      "balance_loss_clip": 1.00990701,
      "balance_loss_mlp": 1.00052643,
      "epoch": 0.6103529128840257,
      "flos": 60644612551680.0,
      "grad_norm": 0.7702837971750532,
      "language_loss": 0.57381576,
      "learning_rate": 1.3920501410320387e-06,
      "loss": 0.59438723,
      "num_input_tokens_seen": 109410690,
      "step": 5076,
      "time_per_iteration": 3.154648780822754
    },
    {
      "auxiliary_loss_clip": 0.01147369,
      "auxiliary_loss_mlp": 0.01027679,
      "balance_loss_clip": 1.04604065,
      "balance_loss_mlp": 1.01955533,
      "epoch": 0.6104731557746649,
      "flos": 19021307806080.0,
      "grad_norm": 2.39989036774582,
      "language_loss": 0.76184165,
      "learning_rate": 1.3913080764301333e-06,
      "loss": 0.7835921,
      "num_input_tokens_seen": 109427650,
      "step": 5077,
      "time_per_iteration": 2.6623387336730957
    },
    {
      "auxiliary_loss_clip": 0.01129943,
      "auxiliary_loss_mlp": 0.01027616,
      "balance_loss_clip": 1.04371595,
      "balance_loss_mlp": 1.01984918,
      "epoch": 0.6105933986653039,
      "flos": 23366894083200.0,
      "grad_norm": 2.0208728968143848,
      "language_loss": 0.71377897,
      "learning_rate": 1.3905661041663085e-06,
      "loss": 0.73535454,
      "num_input_tokens_seen": 109448835,
      "step": 5078,
      "time_per_iteration": 2.8191845417022705
    },
    {
      "auxiliary_loss_clip": 0.01164369,
      "auxiliary_loss_mlp": 0.01028451,
      "balance_loss_clip": 1.05028796,
      "balance_loss_mlp": 1.0204165,
      "epoch": 0.610713641555943,
      "flos": 34637565006720.0,
      "grad_norm": 2.270242190261541,
      "language_loss": 0.65188146,
      "learning_rate": 1.389824224353122e-06,
      "loss": 0.67380965,
      "num_input_tokens_seen": 109470425,
      "step": 5079,
      "time_per_iteration": 3.7860398292541504
    },
    {
      "auxiliary_loss_clip": 0.01162766,
      "auxiliary_loss_mlp": 0.01032972,
      "balance_loss_clip": 1.05072021,
      "balance_loss_mlp": 1.02565265,
      "epoch": 0.610833884446582,
      "flos": 26646471504000.0,
      "grad_norm": 1.653751615546049,
      "language_loss": 0.77024817,
      "learning_rate": 1.389082437103115e-06,
      "loss": 0.79220557,
      "num_input_tokens_seen": 109489695,
      "step": 5080,
      "time_per_iteration": 3.5545904636383057
    },
    {
      "auxiliary_loss_clip": 0.01132976,
      "auxiliary_loss_mlp": 0.0102374,
      "balance_loss_clip": 1.04384708,
      "balance_loss_mlp": 1.01531768,
      "epoch": 0.6109541273372212,
      "flos": 21215126868480.0,
      "grad_norm": 2.7956545335016814,
      "language_loss": 0.78495049,
      "learning_rate": 1.3883407425288172e-06,
      "loss": 0.8065176,
      "num_input_tokens_seen": 109510030,
      "step": 5081,
      "time_per_iteration": 3.6652467250823975
    },
    {
      "auxiliary_loss_clip": 0.01141827,
      "auxiliary_loss_mlp": 0.01021814,
      "balance_loss_clip": 1.04370642,
      "balance_loss_mlp": 1.01345181,
      "epoch": 0.6110743702278603,
      "flos": 20084084438400.0,
      "grad_norm": 2.482333674879216,
      "language_loss": 0.79745674,
      "learning_rate": 1.3875991407427417e-06,
      "loss": 0.81909323,
      "num_input_tokens_seen": 109528255,
      "step": 5082,
      "time_per_iteration": 2.70442795753479
    },
    {
      "auxiliary_loss_clip": 0.01040299,
      "auxiliary_loss_mlp": 0.01003432,
      "balance_loss_clip": 1.00915051,
      "balance_loss_mlp": 1.00205481,
      "epoch": 0.6111946131184993,
      "flos": 68302957438080.0,
      "grad_norm": 0.769428636143955,
      "language_loss": 0.58209747,
      "learning_rate": 1.38685763185739e-06,
      "loss": 0.60253477,
      "num_input_tokens_seen": 109581915,
      "step": 5083,
      "time_per_iteration": 3.323903799057007
    },
    {
      "auxiliary_loss_clip": 0.01173394,
      "auxiliary_loss_mlp": 0.01026508,
      "balance_loss_clip": 1.0482645,
      "balance_loss_mlp": 1.01920342,
      "epoch": 0.6113148560091385,
      "flos": 19937676602880.0,
      "grad_norm": 2.4608168522728966,
      "language_loss": 0.67672396,
      "learning_rate": 1.3861162159852476e-06,
      "loss": 0.69872296,
      "num_input_tokens_seen": 109600050,
      "step": 5084,
      "time_per_iteration": 3.6494317054748535
    },
    {
      "auxiliary_loss_clip": 0.01153066,
      "auxiliary_loss_mlp": 0.01025282,
      "balance_loss_clip": 1.04939508,
      "balance_loss_mlp": 1.01726544,
      "epoch": 0.6114350988997775,
      "flos": 23731854220800.0,
      "grad_norm": 1.6287432313150219,
      "language_loss": 0.80081099,
      "learning_rate": 1.3853748932387875e-06,
      "loss": 0.82259452,
      "num_input_tokens_seen": 109620690,
      "step": 5085,
      "time_per_iteration": 2.806335210800171
    },
    {
      "auxiliary_loss_clip": 0.01136545,
      "auxiliary_loss_mlp": 0.01023107,
      "balance_loss_clip": 1.04368234,
      "balance_loss_mlp": 1.01506972,
      "epoch": 0.6115553417904166,
      "flos": 24023700224640.0,
      "grad_norm": 2.4478155433616817,
      "language_loss": 0.74719644,
      "learning_rate": 1.3846336637304671e-06,
      "loss": 0.76879299,
      "num_input_tokens_seen": 109638960,
      "step": 5086,
      "time_per_iteration": 2.724245548248291
    },
    {
      "auxiliary_loss_clip": 0.01139949,
      "auxiliary_loss_mlp": 0.01024411,
      "balance_loss_clip": 1.04611671,
      "balance_loss_mlp": 1.01659691,
      "epoch": 0.6116755846810558,
      "flos": 23733542160000.0,
      "grad_norm": 2.3017752554114383,
      "language_loss": 0.83155388,
      "learning_rate": 1.3838925275727316e-06,
      "loss": 0.85319746,
      "num_input_tokens_seen": 109659700,
      "step": 5087,
      "time_per_iteration": 2.7186381816864014
    },
    {
      "auxiliary_loss_clip": 0.01176454,
      "auxiliary_loss_mlp": 0.01026411,
      "balance_loss_clip": 1.05118906,
      "balance_loss_mlp": 1.0190382,
      "epoch": 0.6117958275716948,
      "flos": 18661626967680.0,
      "grad_norm": 2.0108947975046227,
      "language_loss": 0.78928006,
      "learning_rate": 1.3831514848780089e-06,
      "loss": 0.81130874,
      "num_input_tokens_seen": 109679275,
      "step": 5088,
      "time_per_iteration": 2.6995601654052734
    },
    {
      "auxiliary_loss_clip": 0.0115396,
      "auxiliary_loss_mlp": 0.01024832,
      "balance_loss_clip": 1.04651999,
      "balance_loss_mlp": 1.01781046,
      "epoch": 0.6119160704623339,
      "flos": 16471183783680.0,
      "grad_norm": 2.4922608323486832,
      "language_loss": 0.92110914,
      "learning_rate": 1.3824105357587152e-06,
      "loss": 0.94289702,
      "num_input_tokens_seen": 109696380,
      "step": 5089,
      "time_per_iteration": 2.6134634017944336
    },
    {
      "auxiliary_loss_clip": 0.01139732,
      "auxiliary_loss_mlp": 0.01022683,
      "balance_loss_clip": 1.04336262,
      "balance_loss_mlp": 1.0157094,
      "epoch": 0.612036313352973,
      "flos": 23915465568000.0,
      "grad_norm": 1.5731645071657891,
      "language_loss": 0.82481557,
      "learning_rate": 1.381669680327253e-06,
      "loss": 0.84643972,
      "num_input_tokens_seen": 109718060,
      "step": 5090,
      "time_per_iteration": 2.6760735511779785
    },
    {
      "auxiliary_loss_clip": 0.01141311,
      "auxiliary_loss_mlp": 0.01028626,
      "balance_loss_clip": 1.04810929,
      "balance_loss_mlp": 1.02051735,
      "epoch": 0.6121565562436121,
      "flos": 26974766833920.0,
      "grad_norm": 2.097683624069703,
      "language_loss": 0.70889509,
      "learning_rate": 1.380928918696008e-06,
      "loss": 0.7305944,
      "num_input_tokens_seen": 109736830,
      "step": 5091,
      "time_per_iteration": 2.739025592803955
    },
    {
      "auxiliary_loss_clip": 0.0116169,
      "auxiliary_loss_mlp": 0.01027159,
      "balance_loss_clip": 1.04928732,
      "balance_loss_mlp": 1.01893318,
      "epoch": 0.6122767991342511,
      "flos": 15668867646720.0,
      "grad_norm": 2.4341000921912648,
      "language_loss": 0.71157807,
      "learning_rate": 1.3801882509773548e-06,
      "loss": 0.73346651,
      "num_input_tokens_seen": 109754690,
      "step": 5092,
      "time_per_iteration": 2.592524528503418
    },
    {
      "auxiliary_loss_clip": 0.01154421,
      "auxiliary_loss_mlp": 0.01022444,
      "balance_loss_clip": 1.0443747,
      "balance_loss_mlp": 1.01509476,
      "epoch": 0.6123970420248903,
      "flos": 27964321591680.0,
      "grad_norm": 1.6194449802237743,
      "language_loss": 0.81641567,
      "learning_rate": 1.3794476772836503e-06,
      "loss": 0.8381843,
      "num_input_tokens_seen": 109775790,
      "step": 5093,
      "time_per_iteration": 2.6756527423858643
    },
    {
      "auxiliary_loss_clip": 0.011261,
      "auxiliary_loss_mlp": 0.01024741,
      "balance_loss_clip": 1.04578793,
      "balance_loss_mlp": 1.01699877,
      "epoch": 0.6125172849155294,
      "flos": 21468727866240.0,
      "grad_norm": 1.7176692259859787,
      "language_loss": 0.84585953,
      "learning_rate": 1.3787071977272402e-06,
      "loss": 0.86736798,
      "num_input_tokens_seen": 109795050,
      "step": 5094,
      "time_per_iteration": 2.724889039993286
    },
    {
      "auxiliary_loss_clip": 0.01119136,
      "auxiliary_loss_mlp": 0.0102821,
      "balance_loss_clip": 1.04806328,
      "balance_loss_mlp": 1.02052438,
      "epoch": 0.6126375278061684,
      "flos": 16248321849600.0,
      "grad_norm": 2.510817459522866,
      "language_loss": 0.7186352,
      "learning_rate": 1.3779668124204535e-06,
      "loss": 0.74010867,
      "num_input_tokens_seen": 109811465,
      "step": 5095,
      "time_per_iteration": 2.8334970474243164
    },
    {
      "auxiliary_loss_clip": 0.0114417,
      "auxiliary_loss_mlp": 0.01025372,
      "balance_loss_clip": 1.05047488,
      "balance_loss_mlp": 1.01695633,
      "epoch": 0.6127577706968076,
      "flos": 20448865008000.0,
      "grad_norm": 1.6342850178437835,
      "language_loss": 0.80676025,
      "learning_rate": 1.3772265214756074e-06,
      "loss": 0.82845569,
      "num_input_tokens_seen": 109831225,
      "step": 5096,
      "time_per_iteration": 2.7553205490112305
    },
    {
      "auxiliary_loss_clip": 0.01164355,
      "auxiliary_loss_mlp": 0.0102673,
      "balance_loss_clip": 1.04688895,
      "balance_loss_mlp": 1.01851106,
      "epoch": 0.6128780135874466,
      "flos": 18260397072000.0,
      "grad_norm": 1.8692480290703006,
      "language_loss": 0.75007999,
      "learning_rate": 1.3764863250050025e-06,
      "loss": 0.77199078,
      "num_input_tokens_seen": 109849465,
      "step": 5097,
      "time_per_iteration": 2.6290128231048584
    },
    {
      "auxiliary_loss_clip": 0.01132758,
      "auxiliary_loss_mlp": 0.01026601,
      "balance_loss_clip": 1.04358256,
      "balance_loss_mlp": 1.0193944,
      "epoch": 0.6129982564780857,
      "flos": 24937088192640.0,
      "grad_norm": 1.871237289758781,
      "language_loss": 0.80239207,
      "learning_rate": 1.3757462231209272e-06,
      "loss": 0.82398564,
      "num_input_tokens_seen": 109869770,
      "step": 5098,
      "time_per_iteration": 2.75150465965271
    },
    {
      "auxiliary_loss_clip": 0.01140279,
      "auxiliary_loss_mlp": 0.01030947,
      "balance_loss_clip": 1.04477584,
      "balance_loss_mlp": 1.0231564,
      "epoch": 0.6131184993687249,
      "flos": 22492038430080.0,
      "grad_norm": 2.324017785406244,
      "language_loss": 0.88887429,
      "learning_rate": 1.3750062159356525e-06,
      "loss": 0.91058648,
      "num_input_tokens_seen": 109889120,
      "step": 5099,
      "time_per_iteration": 2.9186325073242188
    },
    {
      "auxiliary_loss_clip": 0.01122898,
      "auxiliary_loss_mlp": 0.01030371,
      "balance_loss_clip": 1.04250598,
      "balance_loss_mlp": 1.02265787,
      "epoch": 0.6132387422593639,
      "flos": 15885839750400.0,
      "grad_norm": 1.737963350495826,
      "language_loss": 0.83472532,
      "learning_rate": 1.3742663035614382e-06,
      "loss": 0.85625803,
      "num_input_tokens_seen": 109906490,
      "step": 5100,
      "time_per_iteration": 2.8609015941619873
    },
    {
      "auxiliary_loss_clip": 0.01176137,
      "auxiliary_loss_mlp": 0.01033252,
      "balance_loss_clip": 1.05032015,
      "balance_loss_mlp": 1.02523565,
      "epoch": 0.613358985150003,
      "flos": 25411539962880.0,
      "grad_norm": 2.0833523320931184,
      "language_loss": 0.79908377,
      "learning_rate": 1.3735264861105283e-06,
      "loss": 0.82117772,
      "num_input_tokens_seen": 109927130,
      "step": 5101,
      "time_per_iteration": 2.795558214187622
    },
    {
      "auxiliary_loss_clip": 0.01133871,
      "auxiliary_loss_mlp": 0.01023488,
      "balance_loss_clip": 1.04453278,
      "balance_loss_mlp": 1.01651382,
      "epoch": 0.6134792280406421,
      "flos": 21361283308800.0,
      "grad_norm": 1.9460032681867656,
      "language_loss": 0.78066307,
      "learning_rate": 1.372786763695152e-06,
      "loss": 0.80223668,
      "num_input_tokens_seen": 109945890,
      "step": 5102,
      "time_per_iteration": 2.713202953338623
    },
    {
      "auxiliary_loss_clip": 0.01160516,
      "auxiliary_loss_mlp": 0.01023166,
      "balance_loss_clip": 1.04578972,
      "balance_loss_mlp": 1.01459455,
      "epoch": 0.6135994709312812,
      "flos": 21211248199680.0,
      "grad_norm": 1.885747920346991,
      "language_loss": 0.7741785,
      "learning_rate": 1.3720471364275257e-06,
      "loss": 0.79601526,
      "num_input_tokens_seen": 109965535,
      "step": 5103,
      "time_per_iteration": 2.6315817832946777
    },
    {
      "auxiliary_loss_clip": 0.01128751,
      "auxiliary_loss_mlp": 0.00762925,
      "balance_loss_clip": 1.04404199,
      "balance_loss_mlp": 1.00054479,
      "epoch": 0.6137197138219203,
      "flos": 14794047907200.0,
      "grad_norm": 2.078619914517779,
      "language_loss": 0.78045285,
      "learning_rate": 1.3713076044198486e-06,
      "loss": 0.79936957,
      "num_input_tokens_seen": 109982345,
      "step": 5104,
      "time_per_iteration": 3.7225918769836426
    },
    {
      "auxiliary_loss_clip": 0.01140278,
      "auxiliary_loss_mlp": 0.01024253,
      "balance_loss_clip": 1.04538965,
      "balance_loss_mlp": 1.01722848,
      "epoch": 0.6138399567125594,
      "flos": 20084515401600.0,
      "grad_norm": 2.249859640336811,
      "language_loss": 0.8084814,
      "learning_rate": 1.3705681677843086e-06,
      "loss": 0.8301267,
      "num_input_tokens_seen": 110000940,
      "step": 5105,
      "time_per_iteration": 2.668070077896118
    },
    {
      "auxiliary_loss_clip": 0.01068321,
      "auxiliary_loss_mlp": 0.01001546,
      "balance_loss_clip": 1.00991249,
      "balance_loss_mlp": 1.00017476,
      "epoch": 0.6139601996031985,
      "flos": 60123838193280.0,
      "grad_norm": 0.7707510659849622,
      "language_loss": 0.60561806,
      "learning_rate": 1.3698288266330768e-06,
      "loss": 0.62631673,
      "num_input_tokens_seen": 110061565,
      "step": 5106,
      "time_per_iteration": 3.295222043991089
    },
    {
      "auxiliary_loss_clip": 0.01143712,
      "auxiliary_loss_mlp": 0.01023923,
      "balance_loss_clip": 1.04974627,
      "balance_loss_mlp": 1.01676416,
      "epoch": 0.6140804424938375,
      "flos": 23586703361280.0,
      "grad_norm": 2.1696506615365614,
      "language_loss": 0.72314322,
      "learning_rate": 1.3690895810783113e-06,
      "loss": 0.74481964,
      "num_input_tokens_seen": 110080360,
      "step": 5107,
      "time_per_iteration": 4.557645559310913
    },
    {
      "auxiliary_loss_clip": 0.01110479,
      "auxiliary_loss_mlp": 0.00762466,
      "balance_loss_clip": 1.04161072,
      "balance_loss_mlp": 1.00047958,
      "epoch": 0.6142006853844767,
      "flos": 21398199511680.0,
      "grad_norm": 2.1720846546211283,
      "language_loss": 0.71711397,
      "learning_rate": 1.3683504312321543e-06,
      "loss": 0.73584342,
      "num_input_tokens_seen": 110100695,
      "step": 5108,
      "time_per_iteration": 2.819596529006958
    },
    {
      "auxiliary_loss_clip": 0.0116452,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.04813349,
      "balance_loss_mlp": 1.02166224,
      "epoch": 0.6143209282751158,
      "flos": 12057367622400.0,
      "grad_norm": 2.30332009779576,
      "language_loss": 0.79731083,
      "learning_rate": 1.3676113772067355e-06,
      "loss": 0.81924939,
      "num_input_tokens_seen": 110117750,
      "step": 5109,
      "time_per_iteration": 2.6225333213806152
    },
    {
      "auxiliary_loss_clip": 0.01123948,
      "auxiliary_loss_mlp": 0.01030795,
      "balance_loss_clip": 1.04387593,
      "balance_loss_mlp": 1.02310598,
      "epoch": 0.6144411711657548,
      "flos": 25082274965760.0,
      "grad_norm": 2.2727888892407364,
      "language_loss": 0.72822255,
      "learning_rate": 1.3668724191141671e-06,
      "loss": 0.74976999,
      "num_input_tokens_seen": 110137020,
      "step": 5110,
      "time_per_iteration": 3.8047516345977783
    },
    {
      "auxiliary_loss_clip": 0.01129745,
      "auxiliary_loss_mlp": 0.01030089,
      "balance_loss_clip": 1.05066681,
      "balance_loss_mlp": 1.0224719,
      "epoch": 0.6145614140563939,
      "flos": 20114069316480.0,
      "grad_norm": 2.3710671031094783,
      "language_loss": 0.66740304,
      "learning_rate": 1.3661335570665493e-06,
      "loss": 0.68900138,
      "num_input_tokens_seen": 110154930,
      "step": 5111,
      "time_per_iteration": 2.732600212097168
    },
    {
      "auxiliary_loss_clip": 0.01150803,
      "auxiliary_loss_mlp": 0.01022768,
      "balance_loss_clip": 1.04987216,
      "balance_loss_mlp": 1.01472449,
      "epoch": 0.614681656947033,
      "flos": 16800376953600.0,
      "grad_norm": 2.454539298026532,
      "language_loss": 0.69910383,
      "learning_rate": 1.3653947911759676e-06,
      "loss": 0.72083956,
      "num_input_tokens_seen": 110172480,
      "step": 5112,
      "time_per_iteration": 2.7454352378845215
    },
    {
      "auxiliary_loss_clip": 0.01112345,
      "auxiliary_loss_mlp": 0.01032091,
      "balance_loss_clip": 1.04450762,
      "balance_loss_mlp": 1.02378798,
      "epoch": 0.6148018998376721,
      "flos": 38801587011840.0,
      "grad_norm": 1.6626911804290145,
      "language_loss": 0.74597621,
      "learning_rate": 1.3646561215544904e-06,
      "loss": 0.76742059,
      "num_input_tokens_seen": 110197120,
      "step": 5113,
      "time_per_iteration": 2.903869152069092
    },
    {
      "auxiliary_loss_clip": 0.0116132,
      "auxiliary_loss_mlp": 0.01029143,
      "balance_loss_clip": 1.04836977,
      "balance_loss_mlp": 1.02139699,
      "epoch": 0.6149221427283111,
      "flos": 23327032965120.0,
      "grad_norm": 2.155760607759631,
      "language_loss": 0.79388016,
      "learning_rate": 1.363917548314176e-06,
      "loss": 0.81578481,
      "num_input_tokens_seen": 110216385,
      "step": 5114,
      "time_per_iteration": 2.6495800018310547
    },
    {
      "auxiliary_loss_clip": 0.01167049,
      "auxiliary_loss_mlp": 0.01024809,
      "balance_loss_clip": 1.0486176,
      "balance_loss_mlp": 1.01697111,
      "epoch": 0.6150423856189503,
      "flos": 22379494141440.0,
      "grad_norm": 1.8074752968636894,
      "language_loss": 0.7340253,
      "learning_rate": 1.3631790715670626e-06,
      "loss": 0.75594389,
      "num_input_tokens_seen": 110234790,
      "step": 5115,
      "time_per_iteration": 2.7059402465820312
    },
    {
      "auxiliary_loss_clip": 0.01081114,
      "auxiliary_loss_mlp": 0.01024788,
      "balance_loss_clip": 1.04267943,
      "balance_loss_mlp": 1.01799858,
      "epoch": 0.6151626285095894,
      "flos": 18692078722560.0,
      "grad_norm": 2.0241632661351323,
      "language_loss": 0.85746992,
      "learning_rate": 1.3624406914251783e-06,
      "loss": 0.87852895,
      "num_input_tokens_seen": 110251910,
      "step": 5116,
      "time_per_iteration": 2.946479558944702
    },
    {
      "auxiliary_loss_clip": 0.01162808,
      "auxiliary_loss_mlp": 0.01027112,
      "balance_loss_clip": 1.04800773,
      "balance_loss_mlp": 1.01993012,
      "epoch": 0.6152828714002284,
      "flos": 15851688894720.0,
      "grad_norm": 2.264189305527427,
      "language_loss": 0.88305652,
      "learning_rate": 1.3617024080005335e-06,
      "loss": 0.90495569,
      "num_input_tokens_seen": 110268810,
      "step": 5117,
      "time_per_iteration": 2.9764926433563232
    },
    {
      "auxiliary_loss_clip": 0.01150594,
      "auxiliary_loss_mlp": 0.00762533,
      "balance_loss_clip": 1.04590368,
      "balance_loss_mlp": 1.00051939,
      "epoch": 0.6154031142908676,
      "flos": 24869792062080.0,
      "grad_norm": 1.5552460830434998,
      "language_loss": 0.74627233,
      "learning_rate": 1.3609642214051266e-06,
      "loss": 0.76540357,
      "num_input_tokens_seen": 110293035,
      "step": 5118,
      "time_per_iteration": 2.760932207107544
    },
    {
      "auxiliary_loss_clip": 0.01142714,
      "auxiliary_loss_mlp": 0.0102906,
      "balance_loss_clip": 1.04994798,
      "balance_loss_mlp": 1.0214988,
      "epoch": 0.6155233571815066,
      "flos": 19244744357760.0,
      "grad_norm": 2.8883832057335392,
      "language_loss": 0.66001266,
      "learning_rate": 1.3602261317509385e-06,
      "loss": 0.68173045,
      "num_input_tokens_seen": 110309695,
      "step": 5119,
      "time_per_iteration": 2.714329719543457
    },
    {
      "auxiliary_loss_clip": 0.01161808,
      "auxiliary_loss_mlp": 0.01027964,
      "balance_loss_clip": 1.04709339,
      "balance_loss_mlp": 1.01938105,
      "epoch": 0.6156436000721457,
      "flos": 18770077105920.0,
      "grad_norm": 3.5471802356062923,
      "language_loss": 0.82612252,
      "learning_rate": 1.3594881391499387e-06,
      "loss": 0.8480202,
      "num_input_tokens_seen": 110328610,
      "step": 5120,
      "time_per_iteration": 2.6424176692962646
    },
    {
      "auxiliary_loss_clip": 0.01148762,
      "auxiliary_loss_mlp": 0.01023737,
      "balance_loss_clip": 1.04847682,
      "balance_loss_mlp": 1.015571,
      "epoch": 0.6157638429627849,
      "flos": 18041198325120.0,
      "grad_norm": 1.6626917953618883,
      "language_loss": 0.7944755,
      "learning_rate": 1.3587502437140778e-06,
      "loss": 0.81620049,
      "num_input_tokens_seen": 110346775,
      "step": 5121,
      "time_per_iteration": 2.727005958557129
    },
    {
      "auxiliary_loss_clip": 0.01150782,
      "auxiliary_loss_mlp": 0.01029938,
      "balance_loss_clip": 1.04663897,
      "balance_loss_mlp": 1.02268457,
      "epoch": 0.6158840858534239,
      "flos": 25556726736000.0,
      "grad_norm": 2.579619000881958,
      "language_loss": 0.85027325,
      "learning_rate": 1.3580124455552952e-06,
      "loss": 0.87208045,
      "num_input_tokens_seen": 110366140,
      "step": 5122,
      "time_per_iteration": 2.7320878505706787
    },
    {
      "auxiliary_loss_clip": 0.01161399,
      "auxiliary_loss_mlp": 0.00762314,
      "balance_loss_clip": 1.05099094,
      "balance_loss_mlp": 1.00053,
      "epoch": 0.616004328744063,
      "flos": 24640788902400.0,
      "grad_norm": 1.8620717534832194,
      "language_loss": 0.87536311,
      "learning_rate": 1.3572747447855148e-06,
      "loss": 0.89460027,
      "num_input_tokens_seen": 110386550,
      "step": 5123,
      "time_per_iteration": 2.6589739322662354
    },
    {
      "auxiliary_loss_clip": 0.01177117,
      "auxiliary_loss_mlp": 0.01024955,
      "balance_loss_clip": 1.05057037,
      "balance_loss_mlp": 1.01718283,
      "epoch": 0.6161245716347021,
      "flos": 21689686379520.0,
      "grad_norm": 3.702941604161871,
      "language_loss": 0.6914674,
      "learning_rate": 1.356537141516644e-06,
      "loss": 0.7134881,
      "num_input_tokens_seen": 110403970,
      "step": 5124,
      "time_per_iteration": 2.632977247238159
    },
    {
      "auxiliary_loss_clip": 0.01162501,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.05165327,
      "balance_loss_mlp": 1.02010226,
      "epoch": 0.6162448145253412,
      "flos": 35189225061120.0,
      "grad_norm": 1.867792406663666,
      "language_loss": 0.61886847,
      "learning_rate": 1.3557996358605775e-06,
      "loss": 0.64076531,
      "num_input_tokens_seen": 110423890,
      "step": 5125,
      "time_per_iteration": 2.7215487957000732
    },
    {
      "auxiliary_loss_clip": 0.01159071,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.04682517,
      "balance_loss_mlp": 1.01922572,
      "epoch": 0.6163650574159802,
      "flos": 21615279356160.0,
      "grad_norm": 2.1043129745283395,
      "language_loss": 0.70459473,
      "learning_rate": 1.3550622279291941e-06,
      "loss": 0.72644913,
      "num_input_tokens_seen": 110442035,
      "step": 5126,
      "time_per_iteration": 2.7399542331695557
    },
    {
      "auxiliary_loss_clip": 0.01107545,
      "auxiliary_loss_mlp": 0.01025932,
      "balance_loss_clip": 1.03985453,
      "balance_loss_mlp": 1.01851678,
      "epoch": 0.6164853003066194,
      "flos": 24572163968640.0,
      "grad_norm": 1.4354359396100267,
      "language_loss": 0.83301657,
      "learning_rate": 1.354324917834358e-06,
      "loss": 0.85435128,
      "num_input_tokens_seen": 110463280,
      "step": 5127,
      "time_per_iteration": 2.8185513019561768
    },
    {
      "auxiliary_loss_clip": 0.01099227,
      "auxiliary_loss_mlp": 0.00762034,
      "balance_loss_clip": 1.04035056,
      "balance_loss_mlp": 1.0004878,
      "epoch": 0.6166055431972585,
      "flos": 21835986474240.0,
      "grad_norm": 1.7987626333025601,
      "language_loss": 0.77002358,
      "learning_rate": 1.353587705687918e-06,
      "loss": 0.78863621,
      "num_input_tokens_seen": 110481455,
      "step": 5128,
      "time_per_iteration": 2.834465265274048
    },
    {
      "auxiliary_loss_clip": 0.01152916,
      "auxiliary_loss_mlp": 0.01027727,
      "balance_loss_clip": 1.05034673,
      "balance_loss_mlp": 1.01952493,
      "epoch": 0.6167257860878975,
      "flos": 17785262943360.0,
      "grad_norm": 2.6202025211606315,
      "language_loss": 0.71916336,
      "learning_rate": 1.3528505916017096e-06,
      "loss": 0.74096978,
      "num_input_tokens_seen": 110499155,
      "step": 5129,
      "time_per_iteration": 2.698446273803711
    },
    {
      "auxiliary_loss_clip": 0.0116033,
      "auxiliary_loss_mlp": 0.01026512,
      "balance_loss_clip": 1.04727125,
      "balance_loss_mlp": 1.01884079,
      "epoch": 0.6168460289785367,
      "flos": 23214811898880.0,
      "grad_norm": 2.2220748649869635,
      "language_loss": 0.88745308,
      "learning_rate": 1.3521135756875514e-06,
      "loss": 0.90932149,
      "num_input_tokens_seen": 110515470,
      "step": 5130,
      "time_per_iteration": 3.5330092906951904
    },
    {
      "auxiliary_loss_clip": 0.01097084,
      "auxiliary_loss_mlp": 0.01030017,
      "balance_loss_clip": 1.04067314,
      "balance_loss_mlp": 1.02290964,
      "epoch": 0.6169662718691757,
      "flos": 26213281482240.0,
      "grad_norm": 1.4608311824689204,
      "language_loss": 0.86417389,
      "learning_rate": 1.3513766580572496e-06,
      "loss": 0.88544488,
      "num_input_tokens_seen": 110538290,
      "step": 5131,
      "time_per_iteration": 2.853149175643921
    },
    {
      "auxiliary_loss_clip": 0.01161211,
      "auxiliary_loss_mlp": 0.01029321,
      "balance_loss_clip": 1.05042601,
      "balance_loss_mlp": 1.02235329,
      "epoch": 0.6170865147598148,
      "flos": 19026120228480.0,
      "grad_norm": 2.009584382608736,
      "language_loss": 0.77313226,
      "learning_rate": 1.3506398388225924e-06,
      "loss": 0.79503757,
      "num_input_tokens_seen": 110555610,
      "step": 5132,
      "time_per_iteration": 2.675783634185791
    },
    {
      "auxiliary_loss_clip": 0.01172692,
      "auxiliary_loss_mlp": 0.01028692,
      "balance_loss_clip": 1.05023241,
      "balance_loss_mlp": 1.02124763,
      "epoch": 0.617206757650454,
      "flos": 18260361158400.0,
      "grad_norm": 1.7631389328091054,
      "language_loss": 0.71904421,
      "learning_rate": 1.349903118095355e-06,
      "loss": 0.74105799,
      "num_input_tokens_seen": 110574745,
      "step": 5133,
      "time_per_iteration": 4.711911916732788
    },
    {
      "auxiliary_loss_clip": 0.01165047,
      "auxiliary_loss_mlp": 0.01020641,
      "balance_loss_clip": 1.0494678,
      "balance_loss_mlp": 1.01375365,
      "epoch": 0.617327000541093,
      "flos": 18186959715840.0,
      "grad_norm": 1.6579777536516793,
      "language_loss": 0.73521966,
      "learning_rate": 1.349166495987298e-06,
      "loss": 0.75707662,
      "num_input_tokens_seen": 110593310,
      "step": 5134,
      "time_per_iteration": 2.6439762115478516
    },
    {
      "auxiliary_loss_clip": 0.01060006,
      "auxiliary_loss_mlp": 0.01007764,
      "balance_loss_clip": 1.02513003,
      "balance_loss_mlp": 1.00639307,
      "epoch": 0.6174472434317321,
      "flos": 61833796122240.0,
      "grad_norm": 0.8653160513871059,
      "language_loss": 0.60845083,
      "learning_rate": 1.348429972610166e-06,
      "loss": 0.62912852,
      "num_input_tokens_seen": 110657615,
      "step": 5135,
      "time_per_iteration": 3.3337912559509277
    },
    {
      "auxiliary_loss_clip": 0.01030661,
      "auxiliary_loss_mlp": 0.01004099,
      "balance_loss_clip": 1.02168107,
      "balance_loss_mlp": 1.00278139,
      "epoch": 0.6175674863223712,
      "flos": 71230970494080.0,
      "grad_norm": 0.8468687473811993,
      "language_loss": 0.57846498,
      "learning_rate": 1.3476935480756897e-06,
      "loss": 0.59881258,
      "num_input_tokens_seen": 110714365,
      "step": 5136,
      "time_per_iteration": 4.095571517944336
    },
    {
      "auxiliary_loss_clip": 0.01125402,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.04418933,
      "balance_loss_mlp": 1.01957095,
      "epoch": 0.6176877292130103,
      "flos": 21835447770240.0,
      "grad_norm": 2.457109644593805,
      "language_loss": 0.75352645,
      "learning_rate": 1.346957222495583e-06,
      "loss": 0.77505076,
      "num_input_tokens_seen": 110732160,
      "step": 5137,
      "time_per_iteration": 2.793405294418335
    },
    {
      "auxiliary_loss_clip": 0.01154325,
      "auxiliary_loss_mlp": 0.00762366,
      "balance_loss_clip": 1.05132031,
      "balance_loss_mlp": 1.0005796,
      "epoch": 0.6178079721036493,
      "flos": 17741738638080.0,
      "grad_norm": 2.382259330288585,
      "language_loss": 0.71433252,
      "learning_rate": 1.3462209959815466e-06,
      "loss": 0.73349941,
      "num_input_tokens_seen": 110746900,
      "step": 5138,
      "time_per_iteration": 2.6927008628845215
    },
    {
      "auxiliary_loss_clip": 0.01149046,
      "auxiliary_loss_mlp": 0.0102464,
      "balance_loss_clip": 1.04829824,
      "balance_loss_mlp": 1.01737666,
      "epoch": 0.6179282149942885,
      "flos": 22633131052800.0,
      "grad_norm": 1.9263983054730647,
      "language_loss": 0.74305111,
      "learning_rate": 1.345484868645265e-06,
      "loss": 0.76478791,
      "num_input_tokens_seen": 110765710,
      "step": 5139,
      "time_per_iteration": 2.680694341659546
    },
    {
      "auxiliary_loss_clip": 0.01139694,
      "auxiliary_loss_mlp": 0.01027052,
      "balance_loss_clip": 1.04524279,
      "balance_loss_mlp": 1.01967323,
      "epoch": 0.6180484578849276,
      "flos": 22310330503680.0,
      "grad_norm": 1.9648647488793296,
      "language_loss": 0.78480124,
      "learning_rate": 1.3447488405984088e-06,
      "loss": 0.80646867,
      "num_input_tokens_seen": 110783970,
      "step": 5140,
      "time_per_iteration": 2.714421510696411
    },
    {
      "auxiliary_loss_clip": 0.01145149,
      "auxiliary_loss_mlp": 0.01031502,
      "balance_loss_clip": 1.0466876,
      "balance_loss_mlp": 1.02375305,
      "epoch": 0.6181687007755666,
      "flos": 35225458905600.0,
      "grad_norm": 4.9941183153593975,
      "language_loss": 0.69981349,
      "learning_rate": 1.3440129119526322e-06,
      "loss": 0.72158003,
      "num_input_tokens_seen": 110806395,
      "step": 5141,
      "time_per_iteration": 2.8164448738098145
    },
    {
      "auxiliary_loss_clip": 0.01072076,
      "auxiliary_loss_mlp": 0.01000906,
      "balance_loss_clip": 1.01357698,
      "balance_loss_mlp": 0.99964792,
      "epoch": 0.6182889436662057,
      "flos": 61547370094080.0,
      "grad_norm": 0.8323921930977222,
      "language_loss": 0.51232994,
      "learning_rate": 1.3432770828195762e-06,
      "loss": 0.53305972,
      "num_input_tokens_seen": 110867380,
      "step": 5142,
      "time_per_iteration": 3.3947131633758545
    },
    {
      "auxiliary_loss_clip": 0.01123162,
      "auxiliary_loss_mlp": 0.01031551,
      "balance_loss_clip": 1.04216599,
      "balance_loss_mlp": 1.02389216,
      "epoch": 0.6184091865568448,
      "flos": 19609991804160.0,
      "grad_norm": 2.7321161052111216,
      "language_loss": 0.70511115,
      "learning_rate": 1.3425413533108635e-06,
      "loss": 0.72665828,
      "num_input_tokens_seen": 110885980,
      "step": 5143,
      "time_per_iteration": 2.7420411109924316
    },
    {
      "auxiliary_loss_clip": 0.01120259,
      "auxiliary_loss_mlp": 0.01025662,
      "balance_loss_clip": 1.04775834,
      "balance_loss_mlp": 1.01791334,
      "epoch": 0.6185294294474839,
      "flos": 23586882929280.0,
      "grad_norm": 2.737316670582121,
      "language_loss": 0.70746505,
      "learning_rate": 1.341805723538105e-06,
      "loss": 0.72892427,
      "num_input_tokens_seen": 110906085,
      "step": 5144,
      "time_per_iteration": 2.7991676330566406
    },
    {
      "auxiliary_loss_clip": 0.01153917,
      "auxiliary_loss_mlp": 0.01030303,
      "balance_loss_clip": 1.04771972,
      "balance_loss_mlp": 1.02249515,
      "epoch": 0.618649672338123,
      "flos": 26762032535040.0,
      "grad_norm": 1.541682840690723,
      "language_loss": 0.77484572,
      "learning_rate": 1.3410701936128948e-06,
      "loss": 0.7966879,
      "num_input_tokens_seen": 110928865,
      "step": 5145,
      "time_per_iteration": 2.706430435180664
    },
    {
      "auxiliary_loss_clip": 0.01161378,
      "auxiliary_loss_mlp": 0.01024312,
      "balance_loss_clip": 1.05000198,
      "balance_loss_mlp": 1.01665878,
      "epoch": 0.6187699152287621,
      "flos": 14456630522880.0,
      "grad_norm": 3.820765396652624,
      "language_loss": 0.84926009,
      "learning_rate": 1.340334763646812e-06,
      "loss": 0.871117,
      "num_input_tokens_seen": 110943000,
      "step": 5146,
      "time_per_iteration": 2.6816675662994385
    },
    {
      "auxiliary_loss_clip": 0.01176813,
      "auxiliary_loss_mlp": 0.010299,
      "balance_loss_clip": 1.05001867,
      "balance_loss_mlp": 1.02202344,
      "epoch": 0.6188901581194012,
      "flos": 20084766796800.0,
      "grad_norm": 1.8258164335627944,
      "language_loss": 0.74608648,
      "learning_rate": 1.3395994337514218e-06,
      "loss": 0.76815367,
      "num_input_tokens_seen": 110963170,
      "step": 5147,
      "time_per_iteration": 2.6112060546875
    },
    {
      "auxiliary_loss_clip": 0.01151305,
      "auxiliary_loss_mlp": 0.01024837,
      "balance_loss_clip": 1.04456997,
      "balance_loss_mlp": 1.01725543,
      "epoch": 0.6190104010100402,
      "flos": 25700728360320.0,
      "grad_norm": 3.0328764077602215,
      "language_loss": 0.78643072,
      "learning_rate": 1.3388642040382725e-06,
      "loss": 0.80819213,
      "num_input_tokens_seen": 110983595,
      "step": 5148,
      "time_per_iteration": 2.791912317276001
    },
    {
      "auxiliary_loss_clip": 0.01133079,
      "auxiliary_loss_mlp": 0.01029548,
      "balance_loss_clip": 1.0410192,
      "balance_loss_mlp": 1.02157903,
      "epoch": 0.6191306439006794,
      "flos": 30442372974720.0,
      "grad_norm": 1.6924065875120395,
      "language_loss": 0.84288317,
      "learning_rate": 1.3381290746188975e-06,
      "loss": 0.86450946,
      "num_input_tokens_seen": 111002965,
      "step": 5149,
      "time_per_iteration": 2.743757486343384
    },
    {
      "auxiliary_loss_clip": 0.0116188,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.05081749,
      "balance_loss_mlp": 1.01903272,
      "epoch": 0.6192508867913185,
      "flos": 26685793918080.0,
      "grad_norm": 5.877035890663604,
      "language_loss": 0.6691348,
      "learning_rate": 1.3373940456048152e-06,
      "loss": 0.69101948,
      "num_input_tokens_seen": 111022990,
      "step": 5150,
      "time_per_iteration": 2.6878085136413574
    },
    {
      "auxiliary_loss_clip": 0.01174737,
      "auxiliary_loss_mlp": 0.01026089,
      "balance_loss_clip": 1.05128264,
      "balance_loss_mlp": 1.01905012,
      "epoch": 0.6193711296819575,
      "flos": 36722036090880.0,
      "grad_norm": 1.645514157551669,
      "language_loss": 0.59308267,
      "learning_rate": 1.3366591171075299e-06,
      "loss": 0.61509091,
      "num_input_tokens_seen": 111046495,
      "step": 5151,
      "time_per_iteration": 2.712066650390625
    },
    {
      "auxiliary_loss_clip": 0.0114515,
      "auxiliary_loss_mlp": 0.0102444,
      "balance_loss_clip": 1.04765224,
      "balance_loss_mlp": 1.01746583,
      "epoch": 0.6194913725725967,
      "flos": 25192556697600.0,
      "grad_norm": 1.9540831206298006,
      "language_loss": 0.91262162,
      "learning_rate": 1.335924289238529e-06,
      "loss": 0.93431753,
      "num_input_tokens_seen": 111065705,
      "step": 5152,
      "time_per_iteration": 2.7269067764282227
    },
    {
      "auxiliary_loss_clip": 0.01163307,
      "auxiliary_loss_mlp": 0.00762953,
      "balance_loss_clip": 1.05466592,
      "balance_loss_mlp": 1.00056088,
      "epoch": 0.6196116154632357,
      "flos": 21178821196800.0,
      "grad_norm": 1.7868151830255712,
      "language_loss": 0.76638234,
      "learning_rate": 1.3351895621092859e-06,
      "loss": 0.78564501,
      "num_input_tokens_seen": 111086050,
      "step": 5153,
      "time_per_iteration": 2.6874210834503174
    },
    {
      "auxiliary_loss_clip": 0.01063397,
      "auxiliary_loss_mlp": 0.01026038,
      "balance_loss_clip": 1.03350067,
      "balance_loss_mlp": 1.01854539,
      "epoch": 0.6197318583538748,
      "flos": 16253744803200.0,
      "grad_norm": 2.0866012968309926,
      "language_loss": 0.76468241,
      "learning_rate": 1.3344549358312567e-06,
      "loss": 0.7855767,
      "num_input_tokens_seen": 111104450,
      "step": 5154,
      "time_per_iteration": 3.1413373947143555
    },
    {
      "auxiliary_loss_clip": 0.01166316,
      "auxiliary_loss_mlp": 0.01025907,
      "balance_loss_clip": 1.05267453,
      "balance_loss_mlp": 1.01863837,
      "epoch": 0.619852101244514,
      "flos": 24425612478720.0,
      "grad_norm": 2.06225268436024,
      "language_loss": 0.78358239,
      "learning_rate": 1.3337204105158852e-06,
      "loss": 0.80550462,
      "num_input_tokens_seen": 111123320,
      "step": 5155,
      "time_per_iteration": 3.256554126739502
    },
    {
      "auxiliary_loss_clip": 0.01118667,
      "auxiliary_loss_mlp": 0.0102378,
      "balance_loss_clip": 1.03822875,
      "balance_loss_mlp": 1.01642776,
      "epoch": 0.619972344135153,
      "flos": 16727298733440.0,
      "grad_norm": 2.365523723372955,
      "language_loss": 0.72707713,
      "learning_rate": 1.332985986274597e-06,
      "loss": 0.74850154,
      "num_input_tokens_seen": 111140950,
      "step": 5156,
      "time_per_iteration": 4.3413472175598145
    },
    {
      "auxiliary_loss_clip": 0.01096911,
      "auxiliary_loss_mlp": 0.00762288,
      "balance_loss_clip": 1.04372835,
      "balance_loss_mlp": 1.00052714,
      "epoch": 0.6200925870257921,
      "flos": 12495190498560.0,
      "grad_norm": 2.0140275519881734,
      "language_loss": 0.7538327,
      "learning_rate": 1.3322516632188047e-06,
      "loss": 0.7724247,
      "num_input_tokens_seen": 111157845,
      "step": 5157,
      "time_per_iteration": 2.9348976612091064
    },
    {
      "auxiliary_loss_clip": 0.01130172,
      "auxiliary_loss_mlp": 0.01029314,
      "balance_loss_clip": 1.0449698,
      "balance_loss_mlp": 1.0216161,
      "epoch": 0.6202128299164312,
      "flos": 26539350168960.0,
      "grad_norm": 1.812767283664021,
      "language_loss": 0.67067498,
      "learning_rate": 1.3315174414599045e-06,
      "loss": 0.6922698,
      "num_input_tokens_seen": 111179165,
      "step": 5158,
      "time_per_iteration": 3.926490068435669
    },
    {
      "auxiliary_loss_clip": 0.01155158,
      "auxiliary_loss_mlp": 0.0102782,
      "balance_loss_clip": 1.04667997,
      "balance_loss_mlp": 1.0199703,
      "epoch": 0.6203330728070703,
      "flos": 18770508069120.0,
      "grad_norm": 1.8924795807099115,
      "language_loss": 0.75291669,
      "learning_rate": 1.3307833211092768e-06,
      "loss": 0.77474654,
      "num_input_tokens_seen": 111197830,
      "step": 5159,
      "time_per_iteration": 3.597700834274292
    },
    {
      "auxiliary_loss_clip": 0.01176576,
      "auxiliary_loss_mlp": 0.01027594,
      "balance_loss_clip": 1.05287266,
      "balance_loss_mlp": 1.02007794,
      "epoch": 0.6204533156977093,
      "flos": 20629782835200.0,
      "grad_norm": 1.570854426727099,
      "language_loss": 0.75267327,
      "learning_rate": 1.3300493022782873e-06,
      "loss": 0.77471501,
      "num_input_tokens_seen": 111218400,
      "step": 5160,
      "time_per_iteration": 2.6456613540649414
    },
    {
      "auxiliary_loss_clip": 0.01107017,
      "auxiliary_loss_mlp": 0.00762731,
      "balance_loss_clip": 1.04198146,
      "balance_loss_mlp": 1.00056338,
      "epoch": 0.6205735585883485,
      "flos": 17348050598400.0,
      "grad_norm": 2.0307614736758546,
      "language_loss": 0.72599739,
      "learning_rate": 1.3293153850782855e-06,
      "loss": 0.74469483,
      "num_input_tokens_seen": 111236720,
      "step": 5161,
      "time_per_iteration": 2.7428035736083984
    },
    {
      "auxiliary_loss_clip": 0.01123322,
      "auxiliary_loss_mlp": 0.01030347,
      "balance_loss_clip": 1.04416251,
      "balance_loss_mlp": 1.02244997,
      "epoch": 0.6206938014789876,
      "flos": 22965017742720.0,
      "grad_norm": 1.8346152092028212,
      "language_loss": 0.71383893,
      "learning_rate": 1.3285815696206069e-06,
      "loss": 0.73537564,
      "num_input_tokens_seen": 111258265,
      "step": 5162,
      "time_per_iteration": 3.656492233276367
    },
    {
      "auxiliary_loss_clip": 0.01133309,
      "auxiliary_loss_mlp": 0.01026491,
      "balance_loss_clip": 1.04159045,
      "balance_loss_mlp": 1.01884413,
      "epoch": 0.6208140443696266,
      "flos": 23983192661760.0,
      "grad_norm": 2.2130039300193154,
      "language_loss": 0.76883346,
      "learning_rate": 1.32784785601657e-06,
      "loss": 0.7904315,
      "num_input_tokens_seen": 111277675,
      "step": 5163,
      "time_per_iteration": 2.732396125793457
    },
    {
      "auxiliary_loss_clip": 0.01146921,
      "auxiliary_loss_mlp": 0.01022169,
      "balance_loss_clip": 1.04423451,
      "balance_loss_mlp": 1.01469469,
      "epoch": 0.6209342872602658,
      "flos": 35077291303680.0,
      "grad_norm": 1.8443039108672714,
      "language_loss": 0.73685312,
      "learning_rate": 1.3271142443774798e-06,
      "loss": 0.75854409,
      "num_input_tokens_seen": 111299910,
      "step": 5164,
      "time_per_iteration": 2.831143617630005
    },
    {
      "auxiliary_loss_clip": 0.0114633,
      "auxiliary_loss_mlp": 0.01026232,
      "balance_loss_clip": 1.05025291,
      "balance_loss_mlp": 1.01880503,
      "epoch": 0.6210545301509048,
      "flos": 26979327861120.0,
      "grad_norm": 2.3081532449357764,
      "language_loss": 0.81722677,
      "learning_rate": 1.3263807348146228e-06,
      "loss": 0.83895236,
      "num_input_tokens_seen": 111319765,
      "step": 5165,
      "time_per_iteration": 2.6854279041290283
    },
    {
      "auxiliary_loss_clip": 0.01142811,
      "auxiliary_loss_mlp": 0.01021469,
      "balance_loss_clip": 1.04205418,
      "balance_loss_mlp": 1.01388168,
      "epoch": 0.6211747730415439,
      "flos": 33618240852480.0,
      "grad_norm": 2.181876568621823,
      "language_loss": 0.73586744,
      "learning_rate": 1.3256473274392733e-06,
      "loss": 0.7575103,
      "num_input_tokens_seen": 111341110,
      "step": 5166,
      "time_per_iteration": 2.748460292816162
    },
    {
      "auxiliary_loss_clip": 0.01174755,
      "auxiliary_loss_mlp": 0.01024205,
      "balance_loss_clip": 1.04950666,
      "balance_loss_mlp": 1.01690984,
      "epoch": 0.6212950159321831,
      "flos": 34167099646080.0,
      "grad_norm": 1.980631046904707,
      "language_loss": 0.70065355,
      "learning_rate": 1.3249140223626873e-06,
      "loss": 0.72264314,
      "num_input_tokens_seen": 111362730,
      "step": 5167,
      "time_per_iteration": 2.692476987838745
    },
    {
      "auxiliary_loss_clip": 0.01159617,
      "auxiliary_loss_mlp": 0.01027262,
      "balance_loss_clip": 1.04907072,
      "balance_loss_mlp": 1.01967096,
      "epoch": 0.6214152588228221,
      "flos": 27965758135680.0,
      "grad_norm": 1.5624993942910204,
      "language_loss": 0.75369394,
      "learning_rate": 1.3241808196961077e-06,
      "loss": 0.7755627,
      "num_input_tokens_seen": 111383855,
      "step": 5168,
      "time_per_iteration": 2.6380221843719482
    },
    {
      "auxiliary_loss_clip": 0.01135069,
      "auxiliary_loss_mlp": 0.01023357,
      "balance_loss_clip": 1.04453683,
      "balance_loss_mlp": 1.01642823,
      "epoch": 0.6215355017134612,
      "flos": 20230204965120.0,
      "grad_norm": 1.7124997014721468,
      "language_loss": 0.70922452,
      "learning_rate": 1.3234477195507608e-06,
      "loss": 0.73080873,
      "num_input_tokens_seen": 111402685,
      "step": 5169,
      "time_per_iteration": 2.654358148574829
    },
    {
      "auxiliary_loss_clip": 0.01129365,
      "auxiliary_loss_mlp": 0.010343,
      "balance_loss_clip": 1.04677236,
      "balance_loss_mlp": 1.02673578,
      "epoch": 0.6216557446041003,
      "flos": 41428129219200.0,
      "grad_norm": 2.2504697232453,
      "language_loss": 0.62461686,
      "learning_rate": 1.322714722037857e-06,
      "loss": 0.64625347,
      "num_input_tokens_seen": 111424130,
      "step": 5170,
      "time_per_iteration": 2.8941967487335205
    },
    {
      "auxiliary_loss_clip": 0.01139505,
      "auxiliary_loss_mlp": 0.01028087,
      "balance_loss_clip": 1.04474247,
      "balance_loss_mlp": 1.01943803,
      "epoch": 0.6217759874947394,
      "flos": 27928770105600.0,
      "grad_norm": 2.5693995895181247,
      "language_loss": 0.77336812,
      "learning_rate": 1.321981827268591e-06,
      "loss": 0.79504406,
      "num_input_tokens_seen": 111444785,
      "step": 5171,
      "time_per_iteration": 2.760979413986206
    },
    {
      "auxiliary_loss_clip": 0.01148668,
      "auxiliary_loss_mlp": 0.01026339,
      "balance_loss_clip": 1.04642022,
      "balance_loss_mlp": 1.0190165,
      "epoch": 0.6218962303853784,
      "flos": 21765673601280.0,
      "grad_norm": 1.6678952378089609,
      "language_loss": 0.81035626,
      "learning_rate": 1.3212490353541426e-06,
      "loss": 0.83210635,
      "num_input_tokens_seen": 111467045,
      "step": 5172,
      "time_per_iteration": 2.803497314453125
    },
    {
      "auxiliary_loss_clip": 0.01177497,
      "auxiliary_loss_mlp": 0.01028316,
      "balance_loss_clip": 1.05136752,
      "balance_loss_mlp": 1.02037096,
      "epoch": 0.6220164732760175,
      "flos": 21246260981760.0,
      "grad_norm": 1.9245050051904422,
      "language_loss": 0.80549318,
      "learning_rate": 1.3205163464056762e-06,
      "loss": 0.82755131,
      "num_input_tokens_seen": 111483650,
      "step": 5173,
      "time_per_iteration": 2.6149773597717285
    },
    {
      "auxiliary_loss_clip": 0.01158666,
      "auxiliary_loss_mlp": 0.01024773,
      "balance_loss_clip": 1.04725862,
      "balance_loss_mlp": 1.01745331,
      "epoch": 0.6221367161666567,
      "flos": 26136360506880.0,
      "grad_norm": 1.8001800949161348,
      "language_loss": 0.73201573,
      "learning_rate": 1.319783760534339e-06,
      "loss": 0.7538501,
      "num_input_tokens_seen": 111502895,
      "step": 5174,
      "time_per_iteration": 2.7096168994903564
    },
    {
      "auxiliary_loss_clip": 0.01162368,
      "auxiliary_loss_mlp": 0.01029032,
      "balance_loss_clip": 1.05127883,
      "balance_loss_mlp": 1.0208962,
      "epoch": 0.6222569590572957,
      "flos": 16284196558080.0,
      "grad_norm": 2.035815446723046,
      "language_loss": 0.75302672,
      "learning_rate": 1.319051277851266e-06,
      "loss": 0.77494073,
      "num_input_tokens_seen": 111519180,
      "step": 5175,
      "time_per_iteration": 2.7719216346740723
    },
    {
      "auxiliary_loss_clip": 0.01162779,
      "auxiliary_loss_mlp": 0.0102455,
      "balance_loss_clip": 1.04808187,
      "balance_loss_mlp": 1.01745141,
      "epoch": 0.6223772019479348,
      "flos": 18223840005120.0,
      "grad_norm": 2.07951943974655,
      "language_loss": 0.84048992,
      "learning_rate": 1.3183188984675716e-06,
      "loss": 0.86236322,
      "num_input_tokens_seen": 111537545,
      "step": 5176,
      "time_per_iteration": 2.5973520278930664
    },
    {
      "auxiliary_loss_clip": 0.01147935,
      "auxiliary_loss_mlp": 0.01027359,
      "balance_loss_clip": 1.04959393,
      "balance_loss_mlp": 1.01965773,
      "epoch": 0.6224974448385739,
      "flos": 27489797994240.0,
      "grad_norm": 2.5753177652824943,
      "language_loss": 0.71426469,
      "learning_rate": 1.3175866224943586e-06,
      "loss": 0.73601758,
      "num_input_tokens_seen": 111556265,
      "step": 5177,
      "time_per_iteration": 2.760357618331909
    },
    {
      "auxiliary_loss_clip": 0.01155958,
      "auxiliary_loss_mlp": 0.01023986,
      "balance_loss_clip": 1.05131912,
      "balance_loss_mlp": 1.01616549,
      "epoch": 0.622617687729213,
      "flos": 19791951125760.0,
      "grad_norm": 2.2975033495438195,
      "language_loss": 0.73807549,
      "learning_rate": 1.316854450042712e-06,
      "loss": 0.75987494,
      "num_input_tokens_seen": 111574205,
      "step": 5178,
      "time_per_iteration": 2.654616594314575
    },
    {
      "auxiliary_loss_clip": 0.01164808,
      "auxiliary_loss_mlp": 0.01029906,
      "balance_loss_clip": 1.04970396,
      "balance_loss_mlp": 1.02213669,
      "epoch": 0.622737930619852,
      "flos": 23038886062080.0,
      "grad_norm": 2.1564565912319695,
      "language_loss": 0.74528873,
      "learning_rate": 1.3161223812237024e-06,
      "loss": 0.76723582,
      "num_input_tokens_seen": 111593560,
      "step": 5179,
      "time_per_iteration": 2.644167423248291
    },
    {
      "auxiliary_loss_clip": 0.01175026,
      "auxiliary_loss_mlp": 0.01022256,
      "balance_loss_clip": 1.04954433,
      "balance_loss_mlp": 1.01454866,
      "epoch": 0.6228581735104912,
      "flos": 12634271959680.0,
      "grad_norm": 2.947748484244013,
      "language_loss": 0.85330629,
      "learning_rate": 1.3153904161483842e-06,
      "loss": 0.87527907,
      "num_input_tokens_seen": 111608860,
      "step": 5180,
      "time_per_iteration": 2.5517916679382324
    },
    {
      "auxiliary_loss_clip": 0.01130153,
      "auxiliary_loss_mlp": 0.01025272,
      "balance_loss_clip": 1.04341209,
      "balance_loss_mlp": 1.01750588,
      "epoch": 0.6229784164011303,
      "flos": 23802813538560.0,
      "grad_norm": 1.8610368907706776,
      "language_loss": 0.85174799,
      "learning_rate": 1.3146585549277953e-06,
      "loss": 0.87330222,
      "num_input_tokens_seen": 111627500,
      "step": 5181,
      "time_per_iteration": 2.672966480255127
    },
    {
      "auxiliary_loss_clip": 0.01157957,
      "auxiliary_loss_mlp": 0.01029277,
      "balance_loss_clip": 1.05072725,
      "balance_loss_mlp": 1.02143288,
      "epoch": 0.6230986592917693,
      "flos": 22414219614720.0,
      "grad_norm": 2.0581206312410703,
      "language_loss": 0.78416359,
      "learning_rate": 1.3139267976729591e-06,
      "loss": 0.80603588,
      "num_input_tokens_seen": 111647690,
      "step": 5182,
      "time_per_iteration": 3.6487162113189697
    },
    {
      "auxiliary_loss_clip": 0.0116575,
      "auxiliary_loss_mlp": 0.01024736,
      "balance_loss_clip": 1.05092168,
      "balance_loss_mlp": 1.01705289,
      "epoch": 0.6232189021824085,
      "flos": 34528217028480.0,
      "grad_norm": 1.684923432070248,
      "language_loss": 0.71615148,
      "learning_rate": 1.3131951444948815e-06,
      "loss": 0.7380563,
      "num_input_tokens_seen": 111667090,
      "step": 5183,
      "time_per_iteration": 2.6832220554351807
    },
    {
      "auxiliary_loss_clip": 0.01153532,
      "auxiliary_loss_mlp": 0.01024675,
      "balance_loss_clip": 1.05125415,
      "balance_loss_mlp": 1.01702785,
      "epoch": 0.6233391450730476,
      "flos": 22237000888320.0,
      "grad_norm": 1.7913109425262055,
      "language_loss": 0.76439404,
      "learning_rate": 1.3124635955045546e-06,
      "loss": 0.78617609,
      "num_input_tokens_seen": 111686905,
      "step": 5184,
      "time_per_iteration": 3.648263931274414
    },
    {
      "auxiliary_loss_clip": 0.01107494,
      "auxiliary_loss_mlp": 0.00762866,
      "balance_loss_clip": 1.04059398,
      "balance_loss_mlp": 1.00046921,
      "epoch": 0.6234593879636866,
      "flos": 20332693445760.0,
      "grad_norm": 1.9731241328755287,
      "language_loss": 0.84278369,
      "learning_rate": 1.3117321508129537e-06,
      "loss": 0.86148727,
      "num_input_tokens_seen": 111704985,
      "step": 5185,
      "time_per_iteration": 3.660752773284912
    },
    {
      "auxiliary_loss_clip": 0.01152168,
      "auxiliary_loss_mlp": 0.01027828,
      "balance_loss_clip": 1.05036891,
      "balance_loss_mlp": 1.01977515,
      "epoch": 0.6235796308543258,
      "flos": 20664903358080.0,
      "grad_norm": 1.5043872166648897,
      "language_loss": 0.76786864,
      "learning_rate": 1.3110008105310388e-06,
      "loss": 0.78966868,
      "num_input_tokens_seen": 111724805,
      "step": 5186,
      "time_per_iteration": 2.6329448223114014
    },
    {
      "auxiliary_loss_clip": 0.01175454,
      "auxiliary_loss_mlp": 0.01031102,
      "balance_loss_clip": 1.04825175,
      "balance_loss_mlp": 1.02335382,
      "epoch": 0.6236998737449648,
      "flos": 26618641441920.0,
      "grad_norm": 1.6734060642014357,
      "language_loss": 0.77943474,
      "learning_rate": 1.3102695747697526e-06,
      "loss": 0.80150032,
      "num_input_tokens_seen": 111747675,
      "step": 5187,
      "time_per_iteration": 2.7031948566436768
    },
    {
      "auxiliary_loss_clip": 0.01110067,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.04765773,
      "balance_loss_mlp": 1.01888275,
      "epoch": 0.6238201166356039,
      "flos": 12674599954560.0,
      "grad_norm": 7.351997014559598,
      "language_loss": 0.90637249,
      "learning_rate": 1.3095384436400237e-06,
      "loss": 0.92774487,
      "num_input_tokens_seen": 111759205,
      "step": 5188,
      "time_per_iteration": 2.764676809310913
    },
    {
      "auxiliary_loss_clip": 0.01156372,
      "auxiliary_loss_mlp": 0.01026805,
      "balance_loss_clip": 1.04867148,
      "balance_loss_mlp": 1.0187645,
      "epoch": 0.623940359526243,
      "flos": 10452160730880.0,
      "grad_norm": 2.5278912759688414,
      "language_loss": 0.82281864,
      "learning_rate": 1.3088074172527633e-06,
      "loss": 0.84465039,
      "num_input_tokens_seen": 111776335,
      "step": 5189,
      "time_per_iteration": 3.652198553085327
    },
    {
      "auxiliary_loss_clip": 0.01150537,
      "auxiliary_loss_mlp": 0.01027177,
      "balance_loss_clip": 1.04525208,
      "balance_loss_mlp": 1.01908255,
      "epoch": 0.6240606024168821,
      "flos": 29059525226880.0,
      "grad_norm": 2.481427498585943,
      "language_loss": 0.71805513,
      "learning_rate": 1.3080764957188684e-06,
      "loss": 0.73983228,
      "num_input_tokens_seen": 111796580,
      "step": 5190,
      "time_per_iteration": 2.73006534576416
    },
    {
      "auxiliary_loss_clip": 0.01123631,
      "auxiliary_loss_mlp": 0.01023763,
      "balance_loss_clip": 1.04337144,
      "balance_loss_mlp": 1.01650906,
      "epoch": 0.6241808453075212,
      "flos": 22018089450240.0,
      "grad_norm": 1.7651650770097989,
      "language_loss": 0.70727539,
      "learning_rate": 1.3073456791492192e-06,
      "loss": 0.72874933,
      "num_input_tokens_seen": 111816290,
      "step": 5191,
      "time_per_iteration": 2.7985713481903076
    },
    {
      "auxiliary_loss_clip": 0.01148586,
      "auxiliary_loss_mlp": 0.01027986,
      "balance_loss_clip": 1.0457468,
      "balance_loss_mlp": 1.02090526,
      "epoch": 0.6243010881981603,
      "flos": 21138708683520.0,
      "grad_norm": 2.010945957806756,
      "language_loss": 0.7859416,
      "learning_rate": 1.3066149676546801e-06,
      "loss": 0.80770731,
      "num_input_tokens_seen": 111834470,
      "step": 5192,
      "time_per_iteration": 2.6733171939849854
    },
    {
      "auxiliary_loss_clip": 0.01147444,
      "auxiliary_loss_mlp": 0.01025748,
      "balance_loss_clip": 1.05209756,
      "balance_loss_mlp": 1.01877761,
      "epoch": 0.6244213310887994,
      "flos": 22344948236160.0,
      "grad_norm": 2.2152246134327926,
      "language_loss": 0.66046417,
      "learning_rate": 1.3058843613460985e-06,
      "loss": 0.68219602,
      "num_input_tokens_seen": 111852410,
      "step": 5193,
      "time_per_iteration": 2.7289328575134277
    },
    {
      "auxiliary_loss_clip": 0.01141743,
      "auxiliary_loss_mlp": 0.01025657,
      "balance_loss_clip": 1.04586291,
      "balance_loss_mlp": 1.01801527,
      "epoch": 0.6245415739794384,
      "flos": 15231978524160.0,
      "grad_norm": 2.172803302578828,
      "language_loss": 0.74817193,
      "learning_rate": 1.3051538603343075e-06,
      "loss": 0.76984596,
      "num_input_tokens_seen": 111870340,
      "step": 5194,
      "time_per_iteration": 2.7779204845428467
    },
    {
      "auxiliary_loss_clip": 0.01163068,
      "auxiliary_loss_mlp": 0.01031473,
      "balance_loss_clip": 1.05117345,
      "balance_loss_mlp": 1.02423751,
      "epoch": 0.6246618168700776,
      "flos": 18879891960960.0,
      "grad_norm": 1.9060897301181912,
      "language_loss": 0.68065524,
      "learning_rate": 1.3044234647301235e-06,
      "loss": 0.70260066,
      "num_input_tokens_seen": 111888365,
      "step": 5195,
      "time_per_iteration": 2.59181809425354
    },
    {
      "auxiliary_loss_clip": 0.0115676,
      "auxiliary_loss_mlp": 0.01028203,
      "balance_loss_clip": 1.04832053,
      "balance_loss_mlp": 1.02154207,
      "epoch": 0.6247820597607167,
      "flos": 14319201087360.0,
      "grad_norm": 1.7474291677334377,
      "language_loss": 0.72415298,
      "learning_rate": 1.303693174644347e-06,
      "loss": 0.74600261,
      "num_input_tokens_seen": 111905840,
      "step": 5196,
      "time_per_iteration": 2.6396660804748535
    },
    {
      "auxiliary_loss_clip": 0.01141313,
      "auxiliary_loss_mlp": 0.01028211,
      "balance_loss_clip": 1.04483187,
      "balance_loss_mlp": 1.0203253,
      "epoch": 0.6249023026513557,
      "flos": 22637979388800.0,
      "grad_norm": 2.057619012703494,
      "language_loss": 0.80634171,
      "learning_rate": 1.3029629901877625e-06,
      "loss": 0.82803696,
      "num_input_tokens_seen": 111925215,
      "step": 5197,
      "time_per_iteration": 2.822543144226074
    },
    {
      "auxiliary_loss_clip": 0.01168712,
      "auxiliary_loss_mlp": 0.01026888,
      "balance_loss_clip": 1.05055368,
      "balance_loss_mlp": 1.01928854,
      "epoch": 0.6250225455419949,
      "flos": 20266690204800.0,
      "grad_norm": 4.077896614186699,
      "language_loss": 0.77503157,
      "learning_rate": 1.3022329114711376e-06,
      "loss": 0.79698765,
      "num_input_tokens_seen": 111943925,
      "step": 5198,
      "time_per_iteration": 2.6620635986328125
    },
    {
      "auxiliary_loss_clip": 0.01143509,
      "auxiliary_loss_mlp": 0.01029366,
      "balance_loss_clip": 1.04625988,
      "balance_loss_mlp": 1.02206469,
      "epoch": 0.6251427884326339,
      "flos": 23437853400960.0,
      "grad_norm": 1.7732169399525648,
      "language_loss": 0.69599688,
      "learning_rate": 1.3015029386052256e-06,
      "loss": 0.71772563,
      "num_input_tokens_seen": 111964095,
      "step": 5199,
      "time_per_iteration": 2.6389286518096924
    },
    {
      "auxiliary_loss_clip": 0.01145449,
      "auxiliary_loss_mlp": 0.01026667,
      "balance_loss_clip": 1.04775286,
      "balance_loss_mlp": 1.01921082,
      "epoch": 0.625263031323273,
      "flos": 31723055464320.0,
      "grad_norm": 2.2451072770929525,
      "language_loss": 0.72779465,
      "learning_rate": 1.3007730717007622e-06,
      "loss": 0.74951577,
      "num_input_tokens_seen": 111984910,
      "step": 5200,
      "time_per_iteration": 2.7686476707458496
    },
    {
      "auxiliary_loss_clip": 0.011808,
      "auxiliary_loss_mlp": 0.01025315,
      "balance_loss_clip": 1.05303085,
      "balance_loss_mlp": 1.01677048,
      "epoch": 0.6253832742139122,
      "flos": 24134341092480.0,
      "grad_norm": 1.8365324390972673,
      "language_loss": 0.75136322,
      "learning_rate": 1.3000433108684676e-06,
      "loss": 0.77342439,
      "num_input_tokens_seen": 112005410,
      "step": 5201,
      "time_per_iteration": 2.596453905105591
    },
    {
      "auxiliary_loss_clip": 0.01158517,
      "auxiliary_loss_mlp": 0.01026276,
      "balance_loss_clip": 1.04875398,
      "balance_loss_mlp": 1.01866746,
      "epoch": 0.6255035171045512,
      "flos": 27668812400640.0,
      "grad_norm": 2.4304844407984403,
      "language_loss": 0.80104089,
      "learning_rate": 1.2993136562190467e-06,
      "loss": 0.82288885,
      "num_input_tokens_seen": 112024530,
      "step": 5202,
      "time_per_iteration": 2.7395575046539307
    },
    {
      "auxiliary_loss_clip": 0.01153827,
      "auxiliary_loss_mlp": 0.01025926,
      "balance_loss_clip": 1.0493331,
      "balance_loss_mlp": 1.01827884,
      "epoch": 0.6256237599951903,
      "flos": 20227798753920.0,
      "grad_norm": 1.500772126087787,
      "language_loss": 0.70430833,
      "learning_rate": 1.2985841078631871e-06,
      "loss": 0.72610581,
      "num_input_tokens_seen": 112043850,
      "step": 5203,
      "time_per_iteration": 2.7329280376434326
    },
    {
      "auxiliary_loss_clip": 0.01101444,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.03908992,
      "balance_loss_mlp": 1.02232659,
      "epoch": 0.6257440028858293,
      "flos": 24170574936960.0,
      "grad_norm": 1.7638513433153822,
      "language_loss": 0.78384078,
      "learning_rate": 1.2978546659115608e-06,
      "loss": 0.80515766,
      "num_input_tokens_seen": 112061930,
      "step": 5204,
      "time_per_iteration": 2.819208860397339
    },
    {
      "auxiliary_loss_clip": 0.01153573,
      "auxiliary_loss_mlp": 0.01026053,
      "balance_loss_clip": 1.05015826,
      "balance_loss_mlp": 1.01839948,
      "epoch": 0.6258642457764685,
      "flos": 15851940289920.0,
      "grad_norm": 1.8078146928488974,
      "language_loss": 0.8520838,
      "learning_rate": 1.2971253304748228e-06,
      "loss": 0.87388003,
      "num_input_tokens_seen": 112079645,
      "step": 5205,
      "time_per_iteration": 2.7224647998809814
    },
    {
      "auxiliary_loss_clip": 0.01165357,
      "auxiliary_loss_mlp": 0.0103153,
      "balance_loss_clip": 1.05045915,
      "balance_loss_mlp": 1.02331066,
      "epoch": 0.6259844886671075,
      "flos": 11911354836480.0,
      "grad_norm": 1.7830712440868621,
      "language_loss": 0.75054866,
      "learning_rate": 1.296396101663614e-06,
      "loss": 0.77251756,
      "num_input_tokens_seen": 112096205,
      "step": 5206,
      "time_per_iteration": 2.543144702911377
    },
    {
      "auxiliary_loss_clip": 0.01165621,
      "auxiliary_loss_mlp": 0.01023112,
      "balance_loss_clip": 1.05072117,
      "balance_loss_mlp": 1.01579833,
      "epoch": 0.6261047315577466,
      "flos": 15887958652800.0,
      "grad_norm": 2.0749665865859446,
      "language_loss": 0.84210312,
      "learning_rate": 1.2956669795885565e-06,
      "loss": 0.86399043,
      "num_input_tokens_seen": 112112835,
      "step": 5207,
      "time_per_iteration": 2.603273630142212
    },
    {
      "auxiliary_loss_clip": 0.01125188,
      "auxiliary_loss_mlp": 0.01032843,
      "balance_loss_clip": 1.04599261,
      "balance_loss_mlp": 1.02442145,
      "epoch": 0.6262249744483858,
      "flos": 31248926916480.0,
      "grad_norm": 1.8166920335252112,
      "language_loss": 0.68329263,
      "learning_rate": 1.294937964360259e-06,
      "loss": 0.70487297,
      "num_input_tokens_seen": 112133105,
      "step": 5208,
      "time_per_iteration": 3.712355852127075
    },
    {
      "auxiliary_loss_clip": 0.01155005,
      "auxiliary_loss_mlp": 0.01025781,
      "balance_loss_clip": 1.04639602,
      "balance_loss_mlp": 1.01737678,
      "epoch": 0.6263452173390248,
      "flos": 27198598435200.0,
      "grad_norm": 2.281220222198203,
      "language_loss": 0.71568096,
      "learning_rate": 1.2942090560893108e-06,
      "loss": 0.73748875,
      "num_input_tokens_seen": 112152510,
      "step": 5209,
      "time_per_iteration": 2.7455599308013916
    },
    {
      "auxiliary_loss_clip": 0.01175449,
      "auxiliary_loss_mlp": 0.01028369,
      "balance_loss_clip": 1.05128515,
      "balance_loss_mlp": 1.02109146,
      "epoch": 0.6264654602296639,
      "flos": 37342069683840.0,
      "grad_norm": 1.7018628874425628,
      "language_loss": 0.60670936,
      "learning_rate": 1.2934802548862882e-06,
      "loss": 0.62874758,
      "num_input_tokens_seen": 112175295,
      "step": 5210,
      "time_per_iteration": 3.6763200759887695
    },
    {
      "auxiliary_loss_clip": 0.01144593,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.0454756,
      "balance_loss_mlp": 1.01853704,
      "epoch": 0.626585703120303,
      "flos": 14756952136320.0,
      "grad_norm": 2.3622471547907335,
      "language_loss": 0.82763648,
      "learning_rate": 1.292751560861749e-06,
      "loss": 0.84933943,
      "num_input_tokens_seen": 112190200,
      "step": 5211,
      "time_per_iteration": 3.6164357662200928
    },
    {
      "auxiliary_loss_clip": 0.01179718,
      "auxiliary_loss_mlp": 0.01031843,
      "balance_loss_clip": 1.05306745,
      "balance_loss_mlp": 1.02432692,
      "epoch": 0.6267059460109421,
      "flos": 22347318533760.0,
      "grad_norm": 1.8672179672466032,
      "language_loss": 0.79616684,
      "learning_rate": 1.2920229741262354e-06,
      "loss": 0.81828243,
      "num_input_tokens_seen": 112208205,
      "step": 5212,
      "time_per_iteration": 2.6227216720581055
    },
    {
      "auxiliary_loss_clip": 0.01150897,
      "auxiliary_loss_mlp": 0.01026817,
      "balance_loss_clip": 1.04796553,
      "balance_loss_mlp": 1.01946747,
      "epoch": 0.6268261889015811,
      "flos": 17748813617280.0,
      "grad_norm": 2.315121145441959,
      "language_loss": 0.75635123,
      "learning_rate": 1.2912944947902739e-06,
      "loss": 0.77812839,
      "num_input_tokens_seen": 112224690,
      "step": 5213,
      "time_per_iteration": 2.6239869594573975
    },
    {
      "auxiliary_loss_clip": 0.01156345,
      "auxiliary_loss_mlp": 0.01029407,
      "balance_loss_clip": 1.04920197,
      "balance_loss_mlp": 1.02136648,
      "epoch": 0.6269464317922203,
      "flos": 32846484211200.0,
      "grad_norm": 2.0096879998881616,
      "language_loss": 0.71103835,
      "learning_rate": 1.2905661229643742e-06,
      "loss": 0.73289585,
      "num_input_tokens_seen": 112244450,
      "step": 5214,
      "time_per_iteration": 3.7325282096862793
    },
    {
      "auxiliary_loss_clip": 0.01172677,
      "auxiliary_loss_mlp": 0.01029654,
      "balance_loss_clip": 1.04732442,
      "balance_loss_mlp": 1.02176869,
      "epoch": 0.6270666746828594,
      "flos": 17929192740480.0,
      "grad_norm": 3.32403892533118,
      "language_loss": 0.84107375,
      "learning_rate": 1.2898378587590299e-06,
      "loss": 0.86309701,
      "num_input_tokens_seen": 112261050,
      "step": 5215,
      "time_per_iteration": 2.598836660385132
    },
    {
      "auxiliary_loss_clip": 0.01158782,
      "auxiliary_loss_mlp": 0.01025534,
      "balance_loss_clip": 1.04887509,
      "balance_loss_mlp": 1.01795256,
      "epoch": 0.6271869175734984,
      "flos": 17457326749440.0,
      "grad_norm": 1.8791439894665605,
      "language_loss": 0.87720788,
      "learning_rate": 1.2891097022847173e-06,
      "loss": 0.89905107,
      "num_input_tokens_seen": 112278395,
      "step": 5216,
      "time_per_iteration": 2.6292872428894043
    },
    {
      "auxiliary_loss_clip": 0.01149583,
      "auxiliary_loss_mlp": 0.01027441,
      "balance_loss_clip": 1.0477066,
      "balance_loss_mlp": 1.01926947,
      "epoch": 0.6273071604641376,
      "flos": 26868615166080.0,
      "grad_norm": 6.044129248194314,
      "language_loss": 0.66519767,
      "learning_rate": 1.2883816536518978e-06,
      "loss": 0.68696797,
      "num_input_tokens_seen": 112299535,
      "step": 5217,
      "time_per_iteration": 2.7323081493377686
    },
    {
      "auxiliary_loss_clip": 0.01159115,
      "auxiliary_loss_mlp": 0.01028934,
      "balance_loss_clip": 1.04924166,
      "balance_loss_mlp": 1.02130413,
      "epoch": 0.6274274033547766,
      "flos": 26062384446720.0,
      "grad_norm": 2.8208762930993565,
      "language_loss": 0.81801426,
      "learning_rate": 1.2876537129710155e-06,
      "loss": 0.83989471,
      "num_input_tokens_seen": 112317265,
      "step": 5218,
      "time_per_iteration": 2.6719772815704346
    },
    {
      "auxiliary_loss_clip": 0.01149144,
      "auxiliary_loss_mlp": 0.01030074,
      "balance_loss_clip": 1.05345035,
      "balance_loss_mlp": 1.02173567,
      "epoch": 0.6275476462454157,
      "flos": 20266259241600.0,
      "grad_norm": 2.0773537865330267,
      "language_loss": 0.75413013,
      "learning_rate": 1.286925880352499e-06,
      "loss": 0.7759223,
      "num_input_tokens_seen": 112336125,
      "step": 5219,
      "time_per_iteration": 2.7685234546661377
    },
    {
      "auxiliary_loss_clip": 0.01146292,
      "auxiliary_loss_mlp": 0.01027125,
      "balance_loss_clip": 1.04904974,
      "balance_loss_mlp": 1.01919115,
      "epoch": 0.6276678891360549,
      "flos": 26320402817280.0,
      "grad_norm": 2.307175765344511,
      "language_loss": 0.71384776,
      "learning_rate": 1.2861981559067592e-06,
      "loss": 0.73558193,
      "num_input_tokens_seen": 112356730,
      "step": 5220,
      "time_per_iteration": 2.7136261463165283
    },
    {
      "auxiliary_loss_clip": 0.0110953,
      "auxiliary_loss_mlp": 0.01027795,
      "balance_loss_clip": 1.04138422,
      "balance_loss_mlp": 1.02002883,
      "epoch": 0.6277881320266939,
      "flos": 13912512324480.0,
      "grad_norm": 1.9620119408392362,
      "language_loss": 0.80192918,
      "learning_rate": 1.2854705397441917e-06,
      "loss": 0.82330245,
      "num_input_tokens_seen": 112372270,
      "step": 5221,
      "time_per_iteration": 2.751511812210083
    },
    {
      "auxiliary_loss_clip": 0.0112986,
      "auxiliary_loss_mlp": 0.01026121,
      "balance_loss_clip": 1.04419732,
      "balance_loss_mlp": 1.01902747,
      "epoch": 0.627908374917333,
      "flos": 27048922462080.0,
      "grad_norm": 2.37649303434929,
      "language_loss": 0.77555919,
      "learning_rate": 1.2847430319751747e-06,
      "loss": 0.79711902,
      "num_input_tokens_seen": 112390365,
      "step": 5222,
      "time_per_iteration": 2.7419698238372803
    },
    {
      "auxiliary_loss_clip": 0.01157314,
      "auxiliary_loss_mlp": 0.01030934,
      "balance_loss_clip": 1.05008256,
      "balance_loss_mlp": 1.02383494,
      "epoch": 0.6280286178079721,
      "flos": 23769201386880.0,
      "grad_norm": 2.458621915238166,
      "language_loss": 0.67458189,
      "learning_rate": 1.2840156327100712e-06,
      "loss": 0.69646436,
      "num_input_tokens_seen": 112407490,
      "step": 5223,
      "time_per_iteration": 2.592947006225586
    },
    {
      "auxiliary_loss_clip": 0.0117493,
      "auxiliary_loss_mlp": 0.01030571,
      "balance_loss_clip": 1.04985166,
      "balance_loss_mlp": 1.02289653,
      "epoch": 0.6281488606986112,
      "flos": 26359150613760.0,
      "grad_norm": 1.7826663993509597,
      "language_loss": 0.72112054,
      "learning_rate": 1.2832883420592272e-06,
      "loss": 0.74317551,
      "num_input_tokens_seen": 112426385,
      "step": 5224,
      "time_per_iteration": 2.6519529819488525
    },
    {
      "auxiliary_loss_clip": 0.0114247,
      "auxiliary_loss_mlp": 0.0103333,
      "balance_loss_clip": 1.04614985,
      "balance_loss_mlp": 1.02498829,
      "epoch": 0.6282691035892503,
      "flos": 36137194848000.0,
      "grad_norm": 2.554779050018085,
      "language_loss": 0.64855945,
      "learning_rate": 1.282561160132972e-06,
      "loss": 0.67031741,
      "num_input_tokens_seen": 112446905,
      "step": 5225,
      "time_per_iteration": 2.7584292888641357
    },
    {
      "auxiliary_loss_clip": 0.01153404,
      "auxiliary_loss_mlp": 0.01024563,
      "balance_loss_clip": 1.04640102,
      "balance_loss_mlp": 1.01682067,
      "epoch": 0.6283893464798894,
      "flos": 26537231266560.0,
      "grad_norm": 1.6691281558115578,
      "language_loss": 0.80983418,
      "learning_rate": 1.2818340870416186e-06,
      "loss": 0.83161384,
      "num_input_tokens_seen": 112468040,
      "step": 5226,
      "time_per_iteration": 2.750824451446533
    },
    {
      "auxiliary_loss_clip": 0.01141892,
      "auxiliary_loss_mlp": 0.01030663,
      "balance_loss_clip": 1.04439676,
      "balance_loss_mlp": 1.02253914,
      "epoch": 0.6285095893705285,
      "flos": 22237216369920.0,
      "grad_norm": 2.037206297775577,
      "language_loss": 0.76030457,
      "learning_rate": 1.2811071228954626e-06,
      "loss": 0.78203017,
      "num_input_tokens_seen": 112486675,
      "step": 5227,
      "time_per_iteration": 2.6922826766967773
    },
    {
      "auxiliary_loss_clip": 0.01147831,
      "auxiliary_loss_mlp": 0.01026966,
      "balance_loss_clip": 1.04923499,
      "balance_loss_mlp": 1.01957798,
      "epoch": 0.6286298322611675,
      "flos": 26542259170560.0,
      "grad_norm": 2.1808902380168718,
      "language_loss": 0.80902052,
      "learning_rate": 1.2803802678047846e-06,
      "loss": 0.83076847,
      "num_input_tokens_seen": 112506825,
      "step": 5228,
      "time_per_iteration": 2.759744882583618
    },
    {
      "auxiliary_loss_clip": 0.01152803,
      "auxiliary_loss_mlp": 0.01028514,
      "balance_loss_clip": 1.04938722,
      "balance_loss_mlp": 1.02030897,
      "epoch": 0.6287500751518067,
      "flos": 21795227516160.0,
      "grad_norm": 2.050991718841182,
      "language_loss": 0.74043691,
      "learning_rate": 1.279653521879848e-06,
      "loss": 0.76225007,
      "num_input_tokens_seen": 112526890,
      "step": 5229,
      "time_per_iteration": 2.684418201446533
    },
    {
      "auxiliary_loss_clip": 0.0108458,
      "auxiliary_loss_mlp": 0.01027877,
      "balance_loss_clip": 1.04181671,
      "balance_loss_mlp": 1.0203073,
      "epoch": 0.6288703180424458,
      "flos": 20009605587840.0,
      "grad_norm": 2.4005424501728703,
      "language_loss": 0.83924627,
      "learning_rate": 1.2789268852308997e-06,
      "loss": 0.86037081,
      "num_input_tokens_seen": 112542100,
      "step": 5230,
      "time_per_iteration": 2.9936957359313965
    },
    {
      "auxiliary_loss_clip": 0.01156493,
      "auxiliary_loss_mlp": 0.01026439,
      "balance_loss_clip": 1.04915404,
      "balance_loss_mlp": 1.01836205,
      "epoch": 0.6289905609330848,
      "flos": 22124923476480.0,
      "grad_norm": 2.189892330644738,
      "language_loss": 0.70461547,
      "learning_rate": 1.2782003579681688e-06,
      "loss": 0.72644484,
      "num_input_tokens_seen": 112561630,
      "step": 5231,
      "time_per_iteration": 2.779982089996338
    },
    {
      "auxiliary_loss_clip": 0.0117839,
      "auxiliary_loss_mlp": 0.01029455,
      "balance_loss_clip": 1.05192947,
      "balance_loss_mlp": 1.02189088,
      "epoch": 0.629110803823724,
      "flos": 25518481729920.0,
      "grad_norm": 1.6585234432446059,
      "language_loss": 0.74230003,
      "learning_rate": 1.2774739402018701e-06,
      "loss": 0.76437849,
      "num_input_tokens_seen": 112582465,
      "step": 5232,
      "time_per_iteration": 2.6644949913024902
    },
    {
      "auxiliary_loss_clip": 0.01163727,
      "auxiliary_loss_mlp": 0.01021944,
      "balance_loss_clip": 1.0535953,
      "balance_loss_mlp": 1.01415372,
      "epoch": 0.629231046714363,
      "flos": 20886616056960.0,
      "grad_norm": 1.798591138454584,
      "language_loss": 0.72998798,
      "learning_rate": 1.2767476320422002e-06,
      "loss": 0.75184464,
      "num_input_tokens_seen": 112602390,
      "step": 5233,
      "time_per_iteration": 2.781033992767334
    },
    {
      "auxiliary_loss_clip": 0.01045926,
      "auxiliary_loss_mlp": 0.0100241,
      "balance_loss_clip": 1.01368403,
      "balance_loss_mlp": 1.00098538,
      "epoch": 0.6293512896050021,
      "flos": 65050027908480.0,
      "grad_norm": 0.690127895417718,
      "language_loss": 0.57232964,
      "learning_rate": 1.2760214335993392e-06,
      "loss": 0.59281296,
      "num_input_tokens_seen": 112669035,
      "step": 5234,
      "time_per_iteration": 4.272091627120972
    },
    {
      "auxiliary_loss_clip": 0.01155224,
      "auxiliary_loss_mlp": 0.0103189,
      "balance_loss_clip": 1.04823685,
      "balance_loss_mlp": 1.02464843,
      "epoch": 0.6294715324956413,
      "flos": 34677857088000.0,
      "grad_norm": 2.3643171864412356,
      "language_loss": 0.58890367,
      "learning_rate": 1.2752953449834514e-06,
      "loss": 0.61077487,
      "num_input_tokens_seen": 112691485,
      "step": 5235,
      "time_per_iteration": 2.738816738128662
    },
    {
      "auxiliary_loss_clip": 0.0117726,
      "auxiliary_loss_mlp": 0.01025984,
      "balance_loss_clip": 1.05239344,
      "balance_loss_mlp": 1.01926661,
      "epoch": 0.6295917753862803,
      "flos": 22784207656320.0,
      "grad_norm": 1.6878301766201351,
      "language_loss": 0.80188656,
      "learning_rate": 1.2745693663046836e-06,
      "loss": 0.82391906,
      "num_input_tokens_seen": 112710555,
      "step": 5236,
      "time_per_iteration": 2.6195952892303467
    },
    {
      "auxiliary_loss_clip": 0.01156498,
      "auxiliary_loss_mlp": 0.01023607,
      "balance_loss_clip": 1.04722357,
      "balance_loss_mlp": 1.01639521,
      "epoch": 0.6297120182769194,
      "flos": 20850454039680.0,
      "grad_norm": 2.108978826833143,
      "language_loss": 0.80994105,
      "learning_rate": 1.2738434976731662e-06,
      "loss": 0.83174217,
      "num_input_tokens_seen": 112728740,
      "step": 5237,
      "time_per_iteration": 4.480233669281006
    },
    {
      "auxiliary_loss_clip": 0.01148959,
      "auxiliary_loss_mlp": 0.01028184,
      "balance_loss_clip": 1.0490222,
      "balance_loss_mlp": 1.01988137,
      "epoch": 0.6298322611675584,
      "flos": 19497662997120.0,
      "grad_norm": 1.508154106745284,
      "language_loss": 0.75040549,
      "learning_rate": 1.2731177391990125e-06,
      "loss": 0.77217692,
      "num_input_tokens_seen": 112748665,
      "step": 5238,
      "time_per_iteration": 2.7086920738220215
    },
    {
      "auxiliary_loss_clip": 0.01148212,
      "auxiliary_loss_mlp": 0.01022329,
      "balance_loss_clip": 1.04619277,
      "balance_loss_mlp": 1.0144608,
      "epoch": 0.6299525040581976,
      "flos": 12604466649600.0,
      "grad_norm": 2.108745013756051,
      "language_loss": 0.82074475,
      "learning_rate": 1.2723920909923203e-06,
      "loss": 0.84245014,
      "num_input_tokens_seen": 112764410,
      "step": 5239,
      "time_per_iteration": 2.717611789703369
    },
    {
      "auxiliary_loss_clip": 0.01072532,
      "auxiliary_loss_mlp": 0.01000497,
      "balance_loss_clip": 1.01363659,
      "balance_loss_mlp": 0.99911994,
      "epoch": 0.6300727469488366,
      "flos": 57725685636480.0,
      "grad_norm": 1.0493702367673794,
      "language_loss": 0.60468912,
      "learning_rate": 1.2716665531631688e-06,
      "loss": 0.62541938,
      "num_input_tokens_seen": 112818695,
      "step": 5240,
      "time_per_iteration": 4.189661741256714
    },
    {
      "auxiliary_loss_clip": 0.01166235,
      "auxiliary_loss_mlp": 0.01030145,
      "balance_loss_clip": 1.04791868,
      "balance_loss_mlp": 1.02272463,
      "epoch": 0.6301929898394757,
      "flos": 22527302607360.0,
      "grad_norm": 1.8343127121692446,
      "language_loss": 0.77563882,
      "learning_rate": 1.270941125821623e-06,
      "loss": 0.79760253,
      "num_input_tokens_seen": 112839120,
      "step": 5241,
      "time_per_iteration": 2.686007022857666
    },
    {
      "auxiliary_loss_clip": 0.01158262,
      "auxiliary_loss_mlp": 0.01024479,
      "balance_loss_clip": 1.047539,
      "balance_loss_mlp": 1.01692724,
      "epoch": 0.6303132327301149,
      "flos": 28293550675200.0,
      "grad_norm": 1.8888613216288033,
      "language_loss": 0.75380123,
      "learning_rate": 1.2702158090777278e-06,
      "loss": 0.77562863,
      "num_input_tokens_seen": 112860210,
      "step": 5242,
      "time_per_iteration": 2.679980516433716
    },
    {
      "auxiliary_loss_clip": 0.01127413,
      "auxiliary_loss_mlp": 0.0102773,
      "balance_loss_clip": 1.04370379,
      "balance_loss_mlp": 1.02006447,
      "epoch": 0.6304334756207539,
      "flos": 25264521596160.0,
      "grad_norm": 1.9347447441526566,
      "language_loss": 0.74932843,
      "learning_rate": 1.2694906030415148e-06,
      "loss": 0.77087986,
      "num_input_tokens_seen": 112877955,
      "step": 5243,
      "time_per_iteration": 2.742440700531006
    },
    {
      "auxiliary_loss_clip": 0.0115399,
      "auxiliary_loss_mlp": 0.01026938,
      "balance_loss_clip": 1.04743659,
      "balance_loss_mlp": 1.01895058,
      "epoch": 0.630553718511393,
      "flos": 18033548728320.0,
      "grad_norm": 3.322860633765409,
      "language_loss": 0.82389295,
      "learning_rate": 1.2687655078229958e-06,
      "loss": 0.84570217,
      "num_input_tokens_seen": 112892285,
      "step": 5244,
      "time_per_iteration": 2.61022686958313
    },
    {
      "auxiliary_loss_clip": 0.01147147,
      "auxiliary_loss_mlp": 0.01030941,
      "balance_loss_clip": 1.04939878,
      "balance_loss_mlp": 1.02288294,
      "epoch": 0.6306739614020321,
      "flos": 27304103658240.0,
      "grad_norm": 2.176585482787546,
      "language_loss": 0.69295555,
      "learning_rate": 1.2680405235321678e-06,
      "loss": 0.71473646,
      "num_input_tokens_seen": 112913620,
      "step": 5245,
      "time_per_iteration": 2.7868378162384033
    },
    {
      "auxiliary_loss_clip": 0.01150801,
      "auxiliary_loss_mlp": 0.00763275,
      "balance_loss_clip": 1.0500865,
      "balance_loss_mlp": 1.00050354,
      "epoch": 0.6307942042926712,
      "flos": 15341434243200.0,
      "grad_norm": 2.2064035664222206,
      "language_loss": 0.78862047,
      "learning_rate": 1.267315650279011e-06,
      "loss": 0.80776125,
      "num_input_tokens_seen": 112932090,
      "step": 5246,
      "time_per_iteration": 2.7240219116210938
    },
    {
      "auxiliary_loss_clip": 0.01130011,
      "auxiliary_loss_mlp": 0.01035133,
      "balance_loss_clip": 1.05047488,
      "balance_loss_mlp": 1.02696133,
      "epoch": 0.6309144471833102,
      "flos": 19606400444160.0,
      "grad_norm": 1.9053650983814128,
      "language_loss": 0.73962617,
      "learning_rate": 1.2665908881734874e-06,
      "loss": 0.76127762,
      "num_input_tokens_seen": 112950925,
      "step": 5247,
      "time_per_iteration": 2.735856056213379
    },
    {
      "auxiliary_loss_clip": 0.01162981,
      "auxiliary_loss_mlp": 0.01025973,
      "balance_loss_clip": 1.04940271,
      "balance_loss_mlp": 1.01863599,
      "epoch": 0.6310346900739494,
      "flos": 17493345112320.0,
      "grad_norm": 2.261371385351243,
      "language_loss": 0.84830868,
      "learning_rate": 1.2658662373255432e-06,
      "loss": 0.87019825,
      "num_input_tokens_seen": 112969315,
      "step": 5248,
      "time_per_iteration": 2.684905767440796
    },
    {
      "auxiliary_loss_clip": 0.0105003,
      "auxiliary_loss_mlp": 0.01005135,
      "balance_loss_clip": 1.01348805,
      "balance_loss_mlp": 1.00382996,
      "epoch": 0.6311549329645885,
      "flos": 55070164131840.0,
      "grad_norm": 0.7128106871617246,
      "language_loss": 0.52250278,
      "learning_rate": 1.2651416978451063e-06,
      "loss": 0.5430544,
      "num_input_tokens_seen": 113034700,
      "step": 5249,
      "time_per_iteration": 3.3005452156066895
    },
    {
      "auxiliary_loss_clip": 0.01179635,
      "auxiliary_loss_mlp": 0.01024327,
      "balance_loss_clip": 1.05264282,
      "balance_loss_mlp": 1.01595271,
      "epoch": 0.6312751758552275,
      "flos": 41902545075840.0,
      "grad_norm": 1.9224013915453904,
      "language_loss": 0.65360707,
      "learning_rate": 1.2644172698420903e-06,
      "loss": 0.67564666,
      "num_input_tokens_seen": 113056805,
      "step": 5250,
      "time_per_iteration": 2.886029005050659
    },
    {
      "auxiliary_loss_clip": 0.01136411,
      "auxiliary_loss_mlp": 0.01025545,
      "balance_loss_clip": 1.04705024,
      "balance_loss_mlp": 1.01734388,
      "epoch": 0.6313954187458667,
      "flos": 19646800266240.0,
      "grad_norm": 1.9284398155322893,
      "language_loss": 0.85078079,
      "learning_rate": 1.2636929534263892e-06,
      "loss": 0.8724004,
      "num_input_tokens_seen": 113075790,
      "step": 5251,
      "time_per_iteration": 2.803936004638672
    },
    {
      "auxiliary_loss_clip": 0.01134319,
      "auxiliary_loss_mlp": 0.01026733,
      "balance_loss_clip": 1.04151607,
      "balance_loss_mlp": 1.01918721,
      "epoch": 0.6315156616365057,
      "flos": 22894273906560.0,
      "grad_norm": 1.9408359147568504,
      "language_loss": 0.77854294,
      "learning_rate": 1.2629687487078821e-06,
      "loss": 0.80015349,
      "num_input_tokens_seen": 113094600,
      "step": 5252,
      "time_per_iteration": 2.7692813873291016
    },
    {
      "auxiliary_loss_clip": 0.01164661,
      "auxiliary_loss_mlp": 0.01019851,
      "balance_loss_clip": 1.04753828,
      "balance_loss_mlp": 1.01208496,
      "epoch": 0.6316359045271448,
      "flos": 23726251699200.0,
      "grad_norm": 2.098749712941411,
      "language_loss": 0.76709986,
      "learning_rate": 1.2622446557964293e-06,
      "loss": 0.78894496,
      "num_input_tokens_seen": 113112605,
      "step": 5253,
      "time_per_iteration": 2.636350631713867
    },
    {
      "auxiliary_loss_clip": 0.01144815,
      "auxiliary_loss_mlp": 0.01025104,
      "balance_loss_clip": 1.04222322,
      "balance_loss_mlp": 1.01767683,
      "epoch": 0.631756147417784,
      "flos": 33108417164160.0,
      "grad_norm": 1.7103046256413188,
      "language_loss": 0.71296573,
      "learning_rate": 1.261520674801876e-06,
      "loss": 0.73466492,
      "num_input_tokens_seen": 113133200,
      "step": 5254,
      "time_per_iteration": 2.6760008335113525
    },
    {
      "auxiliary_loss_clip": 0.01146138,
      "auxiliary_loss_mlp": 0.01025407,
      "balance_loss_clip": 1.04990578,
      "balance_loss_mlp": 1.01778996,
      "epoch": 0.631876390308423,
      "flos": 31248424126080.0,
      "grad_norm": 4.681023361472439,
      "language_loss": 0.72130442,
      "learning_rate": 1.2607968058340488e-06,
      "loss": 0.74301988,
      "num_input_tokens_seen": 113152895,
      "step": 5255,
      "time_per_iteration": 2.677609920501709
    },
    {
      "auxiliary_loss_clip": 0.01142429,
      "auxiliary_loss_mlp": 0.01028297,
      "balance_loss_clip": 1.04555833,
      "balance_loss_mlp": 1.02105463,
      "epoch": 0.6319966331990621,
      "flos": 24681152810880.0,
      "grad_norm": 1.7014455637467025,
      "language_loss": 0.73283833,
      "learning_rate": 1.2600730490027583e-06,
      "loss": 0.75454557,
      "num_input_tokens_seen": 113173135,
      "step": 5256,
      "time_per_iteration": 2.563051700592041
    },
    {
      "auxiliary_loss_clip": 0.01129929,
      "auxiliary_loss_mlp": 0.01029177,
      "balance_loss_clip": 1.04426682,
      "balance_loss_mlp": 1.0216012,
      "epoch": 0.6321168760897012,
      "flos": 17491764913920.0,
      "grad_norm": 1.8450099376041216,
      "language_loss": 0.80588508,
      "learning_rate": 1.2593494044177984e-06,
      "loss": 0.82747614,
      "num_input_tokens_seen": 113191440,
      "step": 5257,
      "time_per_iteration": 2.5518555641174316
    },
    {
      "auxiliary_loss_clip": 0.01178447,
      "auxiliary_loss_mlp": 0.01024552,
      "balance_loss_clip": 1.04919982,
      "balance_loss_mlp": 1.01673138,
      "epoch": 0.6322371189803403,
      "flos": 18295373940480.0,
      "grad_norm": 2.7308715307837548,
      "language_loss": 0.8097809,
      "learning_rate": 1.2586258721889448e-06,
      "loss": 0.83181089,
      "num_input_tokens_seen": 113208790,
      "step": 5258,
      "time_per_iteration": 2.4225099086761475
    },
    {
      "auxiliary_loss_clip": 0.01110858,
      "auxiliary_loss_mlp": 0.01029097,
      "balance_loss_clip": 1.04311728,
      "balance_loss_mlp": 1.02094269,
      "epoch": 0.6323573618709794,
      "flos": 20157270399360.0,
      "grad_norm": 1.9399142037700474,
      "language_loss": 0.81856251,
      "learning_rate": 1.2579024524259573e-06,
      "loss": 0.83996212,
      "num_input_tokens_seen": 113225050,
      "step": 5259,
      "time_per_iteration": 2.8477590084075928
    },
    {
      "auxiliary_loss_clip": 0.01141431,
      "auxiliary_loss_mlp": 0.01029612,
      "balance_loss_clip": 1.04236722,
      "balance_loss_mlp": 1.02234626,
      "epoch": 0.6324776047616185,
      "flos": 20042391726720.0,
      "grad_norm": 1.7595675280694352,
      "language_loss": 0.91455519,
      "learning_rate": 1.2571791452385768e-06,
      "loss": 0.93626565,
      "num_input_tokens_seen": 113242315,
      "step": 5260,
      "time_per_iteration": 3.546295166015625
    },
    {
      "auxiliary_loss_clip": 0.01144551,
      "auxiliary_loss_mlp": 0.01027154,
      "balance_loss_clip": 1.04651225,
      "balance_loss_mlp": 1.02019799,
      "epoch": 0.6325978476522576,
      "flos": 30848235724800.0,
      "grad_norm": 1.5980799780731958,
      "language_loss": 0.7708559,
      "learning_rate": 1.2564559507365301e-06,
      "loss": 0.79257298,
      "num_input_tokens_seen": 113264720,
      "step": 5261,
      "time_per_iteration": 2.7284038066864014
    },
    {
      "auxiliary_loss_clip": 0.01149378,
      "auxiliary_loss_mlp": 0.01031292,
      "balance_loss_clip": 1.04798865,
      "balance_loss_mlp": 1.0232811,
      "epoch": 0.6327180905428966,
      "flos": 24535104111360.0,
      "grad_norm": 2.479612934332986,
      "language_loss": 0.7888611,
      "learning_rate": 1.2557328690295244e-06,
      "loss": 0.81066775,
      "num_input_tokens_seen": 113282910,
      "step": 5262,
      "time_per_iteration": 3.6722707748413086
    },
    {
      "auxiliary_loss_clip": 0.01132748,
      "auxiliary_loss_mlp": 0.01025153,
      "balance_loss_clip": 1.0434922,
      "balance_loss_mlp": 1.01814926,
      "epoch": 0.6328383334335358,
      "flos": 21575274583680.0,
      "grad_norm": 1.7783285867579794,
      "language_loss": 0.76302314,
      "learning_rate": 1.255009900227251e-06,
      "loss": 0.78460217,
      "num_input_tokens_seen": 113301935,
      "step": 5263,
      "time_per_iteration": 3.60206937789917
    },
    {
      "auxiliary_loss_clip": 0.01173593,
      "auxiliary_loss_mlp": 0.01024664,
      "balance_loss_clip": 1.04965472,
      "balance_loss_mlp": 1.01757371,
      "epoch": 0.6329585763241748,
      "flos": 22929861306240.0,
      "grad_norm": 1.9708972384726533,
      "language_loss": 0.79302883,
      "learning_rate": 1.254287044439383e-06,
      "loss": 0.81501138,
      "num_input_tokens_seen": 113321540,
      "step": 5264,
      "time_per_iteration": 2.653999090194702
    },
    {
      "auxiliary_loss_clip": 0.01072333,
      "auxiliary_loss_mlp": 0.01000968,
      "balance_loss_clip": 1.01354587,
      "balance_loss_mlp": 0.99969286,
      "epoch": 0.6330788192148139,
      "flos": 70936897847040.0,
      "grad_norm": 0.7685258385029217,
      "language_loss": 0.54443771,
      "learning_rate": 1.2535643017755776e-06,
      "loss": 0.56517076,
      "num_input_tokens_seen": 113383730,
      "step": 5265,
      "time_per_iteration": 3.2321953773498535
    },
    {
      "auxiliary_loss_clip": 0.01131465,
      "auxiliary_loss_mlp": 0.01028915,
      "balance_loss_clip": 1.04325342,
      "balance_loss_mlp": 1.02184558,
      "epoch": 0.6331990621054531,
      "flos": 21244501215360.0,
      "grad_norm": 2.179287839555915,
      "language_loss": 0.72550017,
      "learning_rate": 1.2528416723454737e-06,
      "loss": 0.74710393,
      "num_input_tokens_seen": 113400400,
      "step": 5266,
      "time_per_iteration": 2.6771693229675293
    },
    {
      "auxiliary_loss_clip": 0.01172674,
      "auxiliary_loss_mlp": 0.01026536,
      "balance_loss_clip": 1.05031252,
      "balance_loss_mlp": 1.01929116,
      "epoch": 0.6333193049960921,
      "flos": 34459412526720.0,
      "grad_norm": 1.4101493332025945,
      "language_loss": 0.7122401,
      "learning_rate": 1.2521191562586945e-06,
      "loss": 0.73423219,
      "num_input_tokens_seen": 113424050,
      "step": 5267,
      "time_per_iteration": 3.644361734390259
    },
    {
      "auxiliary_loss_clip": 0.01173985,
      "auxiliary_loss_mlp": 0.00762369,
      "balance_loss_clip": 1.04910004,
      "balance_loss_mlp": 1.00046945,
      "epoch": 0.6334395478867312,
      "flos": 18329883932160.0,
      "grad_norm": 1.9412396535274732,
      "language_loss": 0.77111453,
      "learning_rate": 1.2513967536248445e-06,
      "loss": 0.79047805,
      "num_input_tokens_seen": 113440370,
      "step": 5268,
      "time_per_iteration": 2.5923566818237305
    },
    {
      "auxiliary_loss_clip": 0.01157413,
      "auxiliary_loss_mlp": 0.01030381,
      "balance_loss_clip": 1.04869974,
      "balance_loss_mlp": 1.02293646,
      "epoch": 0.6335597907773702,
      "flos": 23623152687360.0,
      "grad_norm": 1.9655436693651789,
      "language_loss": 0.81391728,
      "learning_rate": 1.2506744645535117e-06,
      "loss": 0.83579528,
      "num_input_tokens_seen": 113460800,
      "step": 5269,
      "time_per_iteration": 2.668367385864258
    },
    {
      "auxiliary_loss_clip": 0.01136196,
      "auxiliary_loss_mlp": 0.01022819,
      "balance_loss_clip": 1.03963041,
      "balance_loss_mlp": 1.01532686,
      "epoch": 0.6336800336680094,
      "flos": 22710913954560.0,
      "grad_norm": 1.9176739924810726,
      "language_loss": 0.60074353,
      "learning_rate": 1.249952289154267e-06,
      "loss": 0.62233377,
      "num_input_tokens_seen": 113480840,
      "step": 5270,
      "time_per_iteration": 2.6531097888946533
    },
    {
      "auxiliary_loss_clip": 0.01087948,
      "auxiliary_loss_mlp": 0.01028041,
      "balance_loss_clip": 1.03855634,
      "balance_loss_mlp": 1.02114725,
      "epoch": 0.6338002765586485,
      "flos": 23622757637760.0,
      "grad_norm": 1.9081654793991867,
      "language_loss": 0.76470029,
      "learning_rate": 1.2492302275366635e-06,
      "loss": 0.78586018,
      "num_input_tokens_seen": 113500515,
      "step": 5271,
      "time_per_iteration": 2.7984304428100586
    },
    {
      "auxiliary_loss_clip": 0.01154056,
      "auxiliary_loss_mlp": 0.01023897,
      "balance_loss_clip": 1.04560518,
      "balance_loss_mlp": 1.01595807,
      "epoch": 0.6339205194492875,
      "flos": 26505450708480.0,
      "grad_norm": 2.847661151612084,
      "language_loss": 0.64872015,
      "learning_rate": 1.2485082798102377e-06,
      "loss": 0.67049968,
      "num_input_tokens_seen": 113520930,
      "step": 5272,
      "time_per_iteration": 2.6851444244384766
    },
    {
      "auxiliary_loss_clip": 0.01136714,
      "auxiliary_loss_mlp": 0.01030417,
      "balance_loss_clip": 1.04423308,
      "balance_loss_mlp": 1.02225685,
      "epoch": 0.6340407623399267,
      "flos": 18544306170240.0,
      "grad_norm": 2.136818037044044,
      "language_loss": 0.68726349,
      "learning_rate": 1.2477864460845084e-06,
      "loss": 0.70893478,
      "num_input_tokens_seen": 113537330,
      "step": 5273,
      "time_per_iteration": 2.7614822387695312
    },
    {
      "auxiliary_loss_clip": 0.01145213,
      "auxiliary_loss_mlp": 0.01034647,
      "balance_loss_clip": 1.04568005,
      "balance_loss_mlp": 1.02652907,
      "epoch": 0.6341610052305657,
      "flos": 17712579772800.0,
      "grad_norm": 2.899265461942339,
      "language_loss": 0.73344773,
      "learning_rate": 1.2470647264689776e-06,
      "loss": 0.75524634,
      "num_input_tokens_seen": 113555810,
      "step": 5274,
      "time_per_iteration": 2.6677842140197754
    },
    {
      "auxiliary_loss_clip": 0.01107197,
      "auxiliary_loss_mlp": 0.01024417,
      "balance_loss_clip": 1.03973627,
      "balance_loss_mlp": 1.01625156,
      "epoch": 0.6342812481212048,
      "flos": 23587026583680.0,
      "grad_norm": 1.9968879115327172,
      "language_loss": 0.70827264,
      "learning_rate": 1.2463431210731282e-06,
      "loss": 0.72958875,
      "num_input_tokens_seen": 113575395,
      "step": 5275,
      "time_per_iteration": 2.7782480716705322
    },
    {
      "auxiliary_loss_clip": 0.01124148,
      "auxiliary_loss_mlp": 0.01028423,
      "balance_loss_clip": 1.0402149,
      "balance_loss_mlp": 1.02074647,
      "epoch": 0.634401491011844,
      "flos": 17821927751040.0,
      "grad_norm": 2.374643647540472,
      "language_loss": 0.76300782,
      "learning_rate": 1.2456216300064289e-06,
      "loss": 0.7845335,
      "num_input_tokens_seen": 113592945,
      "step": 5276,
      "time_per_iteration": 2.7298896312713623
    },
    {
      "auxiliary_loss_clip": 0.01140898,
      "auxiliary_loss_mlp": 0.01030407,
      "balance_loss_clip": 1.04537344,
      "balance_loss_mlp": 1.0228554,
      "epoch": 0.634521733902483,
      "flos": 21358158825600.0,
      "grad_norm": 1.6397897082300514,
      "language_loss": 0.78284878,
      "learning_rate": 1.244900253378328e-06,
      "loss": 0.80456185,
      "num_input_tokens_seen": 113613000,
      "step": 5277,
      "time_per_iteration": 2.6677749156951904
    },
    {
      "auxiliary_loss_clip": 0.01076397,
      "auxiliary_loss_mlp": 0.01030292,
      "balance_loss_clip": 1.04171216,
      "balance_loss_mlp": 1.02345276,
      "epoch": 0.6346419767931221,
      "flos": 16545052103040.0,
      "grad_norm": 1.9961195474176063,
      "language_loss": 0.69258094,
      "learning_rate": 1.2441789912982583e-06,
      "loss": 0.71364784,
      "num_input_tokens_seen": 113630085,
      "step": 5278,
      "time_per_iteration": 2.8389453887939453
    },
    {
      "auxiliary_loss_clip": 0.01163363,
      "auxiliary_loss_mlp": 0.01032075,
      "balance_loss_clip": 1.05043542,
      "balance_loss_mlp": 1.02368236,
      "epoch": 0.6347622196837612,
      "flos": 24350989973760.0,
      "grad_norm": 1.7637818613266685,
      "language_loss": 0.64792824,
      "learning_rate": 1.2434578438756346e-06,
      "loss": 0.6698826,
      "num_input_tokens_seen": 113650515,
      "step": 5279,
      "time_per_iteration": 2.6771962642669678
    },
    {
      "auxiliary_loss_clip": 0.01161409,
      "auxiliary_loss_mlp": 0.01026124,
      "balance_loss_clip": 1.04601741,
      "balance_loss_mlp": 1.01887941,
      "epoch": 0.6348824625744003,
      "flos": 64523178195840.0,
      "grad_norm": 2.3303269093291763,
      "language_loss": 0.78152448,
      "learning_rate": 1.242736811219855e-06,
      "loss": 0.8033998,
      "num_input_tokens_seen": 113676475,
      "step": 5280,
      "time_per_iteration": 2.9890549182891846
    },
    {
      "auxiliary_loss_clip": 0.01154543,
      "auxiliary_loss_mlp": 0.01024242,
      "balance_loss_clip": 1.04732609,
      "balance_loss_mlp": 1.01726866,
      "epoch": 0.6350027054650393,
      "flos": 28622133313920.0,
      "grad_norm": 1.675415837955028,
      "language_loss": 0.81703269,
      "learning_rate": 1.2420158934402988e-06,
      "loss": 0.83882052,
      "num_input_tokens_seen": 113697090,
      "step": 5281,
      "time_per_iteration": 2.7611465454101562
    },
    {
      "auxiliary_loss_clip": 0.01117522,
      "auxiliary_loss_mlp": 0.01024911,
      "balance_loss_clip": 1.04075909,
      "balance_loss_mlp": 1.01731431,
      "epoch": 0.6351229483556785,
      "flos": 23002544476800.0,
      "grad_norm": 2.1143692431512067,
      "language_loss": 0.85161644,
      "learning_rate": 1.2412950906463286e-06,
      "loss": 0.87304074,
      "num_input_tokens_seen": 113714395,
      "step": 5282,
      "time_per_iteration": 2.688275098800659
    },
    {
      "auxiliary_loss_clip": 0.01115977,
      "auxiliary_loss_mlp": 0.01030229,
      "balance_loss_clip": 1.04239368,
      "balance_loss_mlp": 1.02332079,
      "epoch": 0.6352431912463176,
      "flos": 21939300967680.0,
      "grad_norm": 2.0164715919051486,
      "language_loss": 0.89986688,
      "learning_rate": 1.2405744029472902e-06,
      "loss": 0.92132896,
      "num_input_tokens_seen": 113733880,
      "step": 5283,
      "time_per_iteration": 2.73823881149292
    },
    {
      "auxiliary_loss_clip": 0.01142669,
      "auxiliary_loss_mlp": 0.01025198,
      "balance_loss_clip": 1.04418111,
      "balance_loss_mlp": 1.01763439,
      "epoch": 0.6353634341369566,
      "flos": 13735257684480.0,
      "grad_norm": 1.8310045981907783,
      "language_loss": 0.76113969,
      "learning_rate": 1.2398538304525108e-06,
      "loss": 0.78281838,
      "num_input_tokens_seen": 113752505,
      "step": 5284,
      "time_per_iteration": 2.7103374004364014
    },
    {
      "auxiliary_loss_clip": 0.01126208,
      "auxiliary_loss_mlp": 0.01031939,
      "balance_loss_clip": 1.0457195,
      "balance_loss_mlp": 1.0232904,
      "epoch": 0.6354836770275958,
      "flos": 19316170552320.0,
      "grad_norm": 2.1330436067971297,
      "language_loss": 0.75468421,
      "learning_rate": 1.2391333732713016e-06,
      "loss": 0.77626568,
      "num_input_tokens_seen": 113770310,
      "step": 5285,
      "time_per_iteration": 2.7233896255493164
    },
    {
      "auxiliary_loss_clip": 0.0113029,
      "auxiliary_loss_mlp": 0.0103802,
      "balance_loss_clip": 1.04296708,
      "balance_loss_mlp": 1.02971148,
      "epoch": 0.6356039199182348,
      "flos": 21613375935360.0,
      "grad_norm": 2.0871409990848093,
      "language_loss": 0.79010153,
      "learning_rate": 1.2384130315129543e-06,
      "loss": 0.81178463,
      "num_input_tokens_seen": 113788635,
      "step": 5286,
      "time_per_iteration": 3.587177038192749
    },
    {
      "auxiliary_loss_clip": 0.01066996,
      "auxiliary_loss_mlp": 0.01027311,
      "balance_loss_clip": 1.03541219,
      "balance_loss_mlp": 1.01989055,
      "epoch": 0.6357241628088739,
      "flos": 18111978074880.0,
      "grad_norm": 2.120276243448093,
      "language_loss": 0.73289919,
      "learning_rate": 1.2376928052867447e-06,
      "loss": 0.75384223,
      "num_input_tokens_seen": 113807755,
      "step": 5287,
      "time_per_iteration": 3.038170576095581
    },
    {
      "auxiliary_loss_clip": 0.01148154,
      "auxiliary_loss_mlp": 0.01030738,
      "balance_loss_clip": 1.04968822,
      "balance_loss_mlp": 1.02346015,
      "epoch": 0.6358444056995131,
      "flos": 24935256599040.0,
      "grad_norm": 2.2837784572725206,
      "language_loss": 0.77486622,
      "learning_rate": 1.2369726947019299e-06,
      "loss": 0.79665518,
      "num_input_tokens_seen": 113828230,
      "step": 5288,
      "time_per_iteration": 4.634884595870972
    },
    {
      "auxiliary_loss_clip": 0.01156791,
      "auxiliary_loss_mlp": 0.01025436,
      "balance_loss_clip": 1.04478097,
      "balance_loss_mlp": 1.01846206,
      "epoch": 0.6359646485901521,
      "flos": 23293348986240.0,
      "grad_norm": 2.29941443290014,
      "language_loss": 0.67297822,
      "learning_rate": 1.2362526998677511e-06,
      "loss": 0.6948005,
      "num_input_tokens_seen": 113844595,
      "step": 5289,
      "time_per_iteration": 3.5291662216186523
    },
    {
      "auxiliary_loss_clip": 0.01148948,
      "auxiliary_loss_mlp": 0.01029778,
      "balance_loss_clip": 1.04633701,
      "balance_loss_mlp": 1.0230819,
      "epoch": 0.6360848914807912,
      "flos": 20887442069760.0,
      "grad_norm": 2.557268739667188,
      "language_loss": 0.84437758,
      "learning_rate": 1.2355328208934301e-06,
      "loss": 0.8661648,
      "num_input_tokens_seen": 113863470,
      "step": 5290,
      "time_per_iteration": 2.65129017829895
    },
    {
      "auxiliary_loss_clip": 0.01159182,
      "auxiliary_loss_mlp": 0.0076231,
      "balance_loss_clip": 1.04500699,
      "balance_loss_mlp": 1.00049663,
      "epoch": 0.6362051343714303,
      "flos": 18479775386880.0,
      "grad_norm": 1.77972467008367,
      "language_loss": 0.72071528,
      "learning_rate": 1.2348130578881728e-06,
      "loss": 0.73993021,
      "num_input_tokens_seen": 113881690,
      "step": 5291,
      "time_per_iteration": 2.618036985397339
    },
    {
      "auxiliary_loss_clip": 0.01177379,
      "auxiliary_loss_mlp": 0.01030144,
      "balance_loss_clip": 1.05040383,
      "balance_loss_mlp": 1.02203155,
      "epoch": 0.6363253772620694,
      "flos": 24389594115840.0,
      "grad_norm": 2.057771843033151,
      "language_loss": 0.75667763,
      "learning_rate": 1.2340934109611664e-06,
      "loss": 0.7787528,
      "num_input_tokens_seen": 113902450,
      "step": 5292,
      "time_per_iteration": 3.573744773864746
    },
    {
      "auxiliary_loss_clip": 0.01152606,
      "auxiliary_loss_mlp": 0.01027618,
      "balance_loss_clip": 1.04839814,
      "balance_loss_mlp": 1.01967573,
      "epoch": 0.6364456201527084,
      "flos": 25958243940480.0,
      "grad_norm": 2.181494360196889,
      "language_loss": 0.68380755,
      "learning_rate": 1.2333738802215798e-06,
      "loss": 0.7056098,
      "num_input_tokens_seen": 113922670,
      "step": 5293,
      "time_per_iteration": 2.7434213161468506
    },
    {
      "auxiliary_loss_clip": 0.01110469,
      "auxiliary_loss_mlp": 0.01022387,
      "balance_loss_clip": 1.039469,
      "balance_loss_mlp": 1.0154345,
      "epoch": 0.6365658630433476,
      "flos": 20740711011840.0,
      "grad_norm": 2.2689876735538337,
      "language_loss": 0.81064498,
      "learning_rate": 1.2326544657785668e-06,
      "loss": 0.83197355,
      "num_input_tokens_seen": 113942360,
      "step": 5294,
      "time_per_iteration": 2.8029656410217285
    },
    {
      "auxiliary_loss_clip": 0.0112284,
      "auxiliary_loss_mlp": 0.01035096,
      "balance_loss_clip": 1.04234719,
      "balance_loss_mlp": 1.02670979,
      "epoch": 0.6366861059339867,
      "flos": 21434146047360.0,
      "grad_norm": 2.5436829478612086,
      "language_loss": 0.74066776,
      "learning_rate": 1.2319351677412608e-06,
      "loss": 0.76224715,
      "num_input_tokens_seen": 113959405,
      "step": 5295,
      "time_per_iteration": 2.727306365966797
    },
    {
      "auxiliary_loss_clip": 0.01142147,
      "auxiliary_loss_mlp": 0.01025837,
      "balance_loss_clip": 1.04753351,
      "balance_loss_mlp": 1.01822233,
      "epoch": 0.6368063488246257,
      "flos": 22267093507200.0,
      "grad_norm": 1.8415800188896279,
      "language_loss": 0.74934649,
      "learning_rate": 1.2312159862187796e-06,
      "loss": 0.77102637,
      "num_input_tokens_seen": 113977815,
      "step": 5296,
      "time_per_iteration": 2.7386586666107178
    },
    {
      "auxiliary_loss_clip": 0.01180077,
      "auxiliary_loss_mlp": 0.01028374,
      "balance_loss_clip": 1.05303597,
      "balance_loss_mlp": 1.02061319,
      "epoch": 0.6369265917152649,
      "flos": 22420719976320.0,
      "grad_norm": 1.5819765176360219,
      "language_loss": 0.76343715,
      "learning_rate": 1.2304969213202217e-06,
      "loss": 0.78552169,
      "num_input_tokens_seen": 113999075,
      "step": 5297,
      "time_per_iteration": 2.6726162433624268
    },
    {
      "auxiliary_loss_clip": 0.01141602,
      "auxiliary_loss_mlp": 0.01029542,
      "balance_loss_clip": 1.04410934,
      "balance_loss_mlp": 1.0221715,
      "epoch": 0.6370468346059039,
      "flos": 24718176754560.0,
      "grad_norm": 2.4925276795690294,
      "language_loss": 0.79390979,
      "learning_rate": 1.2297779731546692e-06,
      "loss": 0.81562126,
      "num_input_tokens_seen": 114018170,
      "step": 5298,
      "time_per_iteration": 2.7346549034118652
    },
    {
      "auxiliary_loss_clip": 0.01143444,
      "auxiliary_loss_mlp": 0.01029182,
      "balance_loss_clip": 1.04820967,
      "balance_loss_mlp": 1.02209508,
      "epoch": 0.637167077496543,
      "flos": 25296589463040.0,
      "grad_norm": 2.067276268599477,
      "language_loss": 0.77789748,
      "learning_rate": 1.2290591418311853e-06,
      "loss": 0.79962373,
      "num_input_tokens_seen": 114035565,
      "step": 5299,
      "time_per_iteration": 2.7106876373291016
    },
    {
      "auxiliary_loss_clip": 0.01161208,
      "auxiliary_loss_mlp": 0.01027847,
      "balance_loss_clip": 1.05019212,
      "balance_loss_mlp": 1.02036107,
      "epoch": 0.637287320387182,
      "flos": 27671110871040.0,
      "grad_norm": 1.5693646145175868,
      "language_loss": 0.72306943,
      "learning_rate": 1.2283404274588172e-06,
      "loss": 0.74495995,
      "num_input_tokens_seen": 114054510,
      "step": 5300,
      "time_per_iteration": 2.7469778060913086
    },
    {
      "auxiliary_loss_clip": 0.00994892,
      "auxiliary_loss_mlp": 0.01000833,
      "balance_loss_clip": 1.00952315,
      "balance_loss_mlp": 0.9996528,
      "epoch": 0.6374075632778212,
      "flos": 63173406873600.0,
      "grad_norm": 0.7415873848415078,
      "language_loss": 0.52883971,
      "learning_rate": 1.227621830146592e-06,
      "loss": 0.54879701,
      "num_input_tokens_seen": 114109875,
      "step": 5301,
      "time_per_iteration": 3.404991388320923
    },
    {
      "auxiliary_loss_clip": 0.01136921,
      "auxiliary_loss_mlp": 0.0102631,
      "balance_loss_clip": 1.04707646,
      "balance_loss_mlp": 1.01862669,
      "epoch": 0.6375278061684603,
      "flos": 25558127366400.0,
      "grad_norm": 1.9585651414638527,
      "language_loss": 0.79148632,
      "learning_rate": 1.2269033500035217e-06,
      "loss": 0.81311864,
      "num_input_tokens_seen": 114130010,
      "step": 5302,
      "time_per_iteration": 3.368680000305176
    },
    {
      "auxiliary_loss_clip": 0.01131449,
      "auxiliary_loss_mlp": 0.01025915,
      "balance_loss_clip": 1.04551506,
      "balance_loss_mlp": 1.01802659,
      "epoch": 0.6376480490590993,
      "flos": 25666362023040.0,
      "grad_norm": 2.0675034852357927,
      "language_loss": 0.73615062,
      "learning_rate": 1.2261849871385988e-06,
      "loss": 0.75772429,
      "num_input_tokens_seen": 114151115,
      "step": 5303,
      "time_per_iteration": 2.760967493057251
    },
    {
      "auxiliary_loss_clip": 0.01176325,
      "auxiliary_loss_mlp": 0.0102894,
      "balance_loss_clip": 1.0501461,
      "balance_loss_mlp": 1.02156949,
      "epoch": 0.6377682919497385,
      "flos": 31537684350720.0,
      "grad_norm": 2.246283143761021,
      "language_loss": 0.62398255,
      "learning_rate": 1.2254667416607972e-06,
      "loss": 0.64603519,
      "num_input_tokens_seen": 114172715,
      "step": 5304,
      "time_per_iteration": 2.687709093093872
    },
    {
      "auxiliary_loss_clip": 0.01161412,
      "auxiliary_loss_mlp": 0.0102548,
      "balance_loss_clip": 1.05033386,
      "balance_loss_mlp": 1.01786828,
      "epoch": 0.6378885348403776,
      "flos": 23039209284480.0,
      "grad_norm": 1.925133563341257,
      "language_loss": 0.82811928,
      "learning_rate": 1.2247486136790756e-06,
      "loss": 0.84998822,
      "num_input_tokens_seen": 114192195,
      "step": 5305,
      "time_per_iteration": 2.841505527496338
    },
    {
      "auxiliary_loss_clip": 0.01162759,
      "auxiliary_loss_mlp": 0.01029764,
      "balance_loss_clip": 1.04940081,
      "balance_loss_mlp": 1.0223788,
      "epoch": 0.6380087777310166,
      "flos": 18697070712960.0,
      "grad_norm": 2.1775179487553835,
      "language_loss": 0.80852914,
      "learning_rate": 1.2240306033023726e-06,
      "loss": 0.83045435,
      "num_input_tokens_seen": 114210020,
      "step": 5306,
      "time_per_iteration": 2.594621181488037
    },
    {
      "auxiliary_loss_clip": 0.01131688,
      "auxiliary_loss_mlp": 0.01026027,
      "balance_loss_clip": 1.04118454,
      "balance_loss_mlp": 1.01871324,
      "epoch": 0.6381290206216558,
      "flos": 23331558078720.0,
      "grad_norm": 1.7488616917789837,
      "language_loss": 0.72020495,
      "learning_rate": 1.223312710639611e-06,
      "loss": 0.74178207,
      "num_input_tokens_seen": 114228740,
      "step": 5307,
      "time_per_iteration": 2.68925142288208
    },
    {
      "auxiliary_loss_clip": 0.01142872,
      "auxiliary_loss_mlp": 0.01029347,
      "balance_loss_clip": 1.04559803,
      "balance_loss_mlp": 1.02173531,
      "epoch": 0.6382492635122948,
      "flos": 18880466578560.0,
      "grad_norm": 2.1993259722054153,
      "language_loss": 0.86659992,
      "learning_rate": 1.2225949357996928e-06,
      "loss": 0.88832211,
      "num_input_tokens_seen": 114246865,
      "step": 5308,
      "time_per_iteration": 2.6075196266174316
    },
    {
      "auxiliary_loss_clip": 0.01156911,
      "auxiliary_loss_mlp": 0.01032299,
      "balance_loss_clip": 1.04963899,
      "balance_loss_mlp": 1.02519393,
      "epoch": 0.6383695064029339,
      "flos": 27819134818560.0,
      "grad_norm": 1.4832887252028344,
      "language_loss": 0.80379075,
      "learning_rate": 1.221877278891505e-06,
      "loss": 0.82568276,
      "num_input_tokens_seen": 114266120,
      "step": 5309,
      "time_per_iteration": 2.715855598449707
    },
    {
      "auxiliary_loss_clip": 0.01165209,
      "auxiliary_loss_mlp": 0.01033828,
      "balance_loss_clip": 1.04855645,
      "balance_loss_mlp": 1.02567101,
      "epoch": 0.638489749293573,
      "flos": 26395635853440.0,
      "grad_norm": 2.018405432019353,
      "language_loss": 0.71561974,
      "learning_rate": 1.221159740023915e-06,
      "loss": 0.7376101,
      "num_input_tokens_seen": 114285950,
      "step": 5310,
      "time_per_iteration": 2.655057907104492
    },
    {
      "auxiliary_loss_clip": 0.01141862,
      "auxiliary_loss_mlp": 0.00762883,
      "balance_loss_clip": 1.04819918,
      "balance_loss_mlp": 1.00053453,
      "epoch": 0.6386099921842121,
      "flos": 23988328306560.0,
      "grad_norm": 6.035588048634325,
      "language_loss": 0.7234509,
      "learning_rate": 1.2204423193057735e-06,
      "loss": 0.74249834,
      "num_input_tokens_seen": 114304780,
      "step": 5311,
      "time_per_iteration": 2.8233413696289062
    },
    {
      "auxiliary_loss_clip": 0.01053852,
      "auxiliary_loss_mlp": 0.01001003,
      "balance_loss_clip": 1.01542854,
      "balance_loss_mlp": 0.99974567,
      "epoch": 0.6387302350748512,
      "flos": 71731169337600.0,
      "grad_norm": 0.8452672690425748,
      "language_loss": 0.63308752,
      "learning_rate": 1.2197250168459122e-06,
      "loss": 0.65363598,
      "num_input_tokens_seen": 114361180,
      "step": 5312,
      "time_per_iteration": 4.199880838394165
    },
    {
      "auxiliary_loss_clip": 0.01163399,
      "auxiliary_loss_mlp": 0.01027068,
      "balance_loss_clip": 1.04858959,
      "balance_loss_mlp": 1.02010036,
      "epoch": 0.6388504779654903,
      "flos": 14535778141440.0,
      "grad_norm": 1.691616405903377,
      "language_loss": 0.7419101,
      "learning_rate": 1.2190078327531454e-06,
      "loss": 0.76381481,
      "num_input_tokens_seen": 114377425,
      "step": 5313,
      "time_per_iteration": 2.603571653366089
    },
    {
      "auxiliary_loss_clip": 0.01161222,
      "auxiliary_loss_mlp": 0.01025379,
      "balance_loss_clip": 1.04836512,
      "balance_loss_mlp": 1.01854503,
      "epoch": 0.6389707208561294,
      "flos": 22346133384960.0,
      "grad_norm": 1.4173063366601204,
      "language_loss": 0.7275964,
      "learning_rate": 1.2182907671362697e-06,
      "loss": 0.74946243,
      "num_input_tokens_seen": 114398120,
      "step": 5314,
      "time_per_iteration": 3.656221628189087
    },
    {
      "auxiliary_loss_clip": 0.01161402,
      "auxiliary_loss_mlp": 0.01028432,
      "balance_loss_clip": 1.05037165,
      "balance_loss_mlp": 1.0211457,
      "epoch": 0.6390909637467684,
      "flos": 19426883247360.0,
      "grad_norm": 2.7112823466930536,
      "language_loss": 0.78783345,
      "learning_rate": 1.2175738201040626e-06,
      "loss": 0.80973178,
      "num_input_tokens_seen": 114415160,
      "step": 5315,
      "time_per_iteration": 3.570990800857544
    },
    {
      "auxiliary_loss_clip": 0.01160563,
      "auxiliary_loss_mlp": 0.01027233,
      "balance_loss_clip": 1.04875159,
      "balance_loss_mlp": 1.01973462,
      "epoch": 0.6392112066374076,
      "flos": 24090852700800.0,
      "grad_norm": 2.4512719865386874,
      "language_loss": 0.78191501,
      "learning_rate": 1.2168569917652855e-06,
      "loss": 0.80379295,
      "num_input_tokens_seen": 114435015,
      "step": 5316,
      "time_per_iteration": 2.711521863937378
    },
    {
      "auxiliary_loss_clip": 0.01158919,
      "auxiliary_loss_mlp": 0.01025774,
      "balance_loss_clip": 1.04927981,
      "balance_loss_mlp": 1.01870775,
      "epoch": 0.6393314495280467,
      "flos": 26795141896320.0,
      "grad_norm": 1.5665602098075946,
      "language_loss": 0.64286017,
      "learning_rate": 1.2161402822286797e-06,
      "loss": 0.66470706,
      "num_input_tokens_seen": 114455700,
      "step": 5317,
      "time_per_iteration": 2.690436601638794
    },
    {
      "auxiliary_loss_clip": 0.01129294,
      "auxiliary_loss_mlp": 0.01023429,
      "balance_loss_clip": 1.04329109,
      "balance_loss_mlp": 1.0158056,
      "epoch": 0.6394516924186857,
      "flos": 20260692633600.0,
      "grad_norm": 1.9781297407167873,
      "language_loss": 0.79439914,
      "learning_rate": 1.2154236916029703e-06,
      "loss": 0.81592643,
      "num_input_tokens_seen": 114473675,
      "step": 5318,
      "time_per_iteration": 3.6103367805480957
    },
    {
      "auxiliary_loss_clip": 0.01114,
      "auxiliary_loss_mlp": 0.01024496,
      "balance_loss_clip": 1.03892767,
      "balance_loss_mlp": 1.01703084,
      "epoch": 0.6395719353093249,
      "flos": 18368847210240.0,
      "grad_norm": 2.2051832447757045,
      "language_loss": 0.73096967,
      "learning_rate": 1.2147072199968627e-06,
      "loss": 0.75235468,
      "num_input_tokens_seen": 114492310,
      "step": 5319,
      "time_per_iteration": 2.7295048236846924
    },
    {
      "auxiliary_loss_clip": 0.01155768,
      "auxiliary_loss_mlp": 0.01021281,
      "balance_loss_clip": 1.04699922,
      "balance_loss_mlp": 1.01420879,
      "epoch": 0.6396921781999639,
      "flos": 17566315591680.0,
      "grad_norm": 1.8858108512048277,
      "language_loss": 0.71759582,
      "learning_rate": 1.2139908675190454e-06,
      "loss": 0.73936629,
      "num_input_tokens_seen": 114511520,
      "step": 5320,
      "time_per_iteration": 2.731945753097534
    },
    {
      "auxiliary_loss_clip": 0.01090251,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.03541028,
      "balance_loss_mlp": 1.01953852,
      "epoch": 0.639812421090603,
      "flos": 21251252972160.0,
      "grad_norm": 1.9917184137324324,
      "language_loss": 0.74906778,
      "learning_rate": 1.2132746342781883e-06,
      "loss": 0.77023935,
      "num_input_tokens_seen": 114532680,
      "step": 5321,
      "time_per_iteration": 2.7771174907684326
    },
    {
      "auxiliary_loss_clip": 0.01174724,
      "auxiliary_loss_mlp": 0.01029606,
      "balance_loss_clip": 1.04901111,
      "balance_loss_mlp": 1.02220654,
      "epoch": 0.6399326639812422,
      "flos": 11180967684480.0,
      "grad_norm": 2.5922901615688914,
      "language_loss": 0.79438376,
      "learning_rate": 1.2125585203829442e-06,
      "loss": 0.81642711,
      "num_input_tokens_seen": 114548320,
      "step": 5322,
      "time_per_iteration": 2.5808699131011963
    },
    {
      "auxiliary_loss_clip": 0.01119064,
      "auxiliary_loss_mlp": 0.01028691,
      "balance_loss_clip": 1.04365969,
      "balance_loss_mlp": 1.02037048,
      "epoch": 0.6400529068718812,
      "flos": 23911048195200.0,
      "grad_norm": 2.04856855257734,
      "language_loss": 0.74466324,
      "learning_rate": 1.211842525941946e-06,
      "loss": 0.76614082,
      "num_input_tokens_seen": 114568115,
      "step": 5323,
      "time_per_iteration": 2.723644971847534
    },
    {
      "auxiliary_loss_clip": 0.01114824,
      "auxiliary_loss_mlp": 0.0102218,
      "balance_loss_clip": 1.04454517,
      "balance_loss_mlp": 1.01557875,
      "epoch": 0.6401731497625203,
      "flos": 44018724890880.0,
      "grad_norm": 2.113475347509435,
      "language_loss": 0.78656989,
      "learning_rate": 1.2111266510638105e-06,
      "loss": 0.80794001,
      "num_input_tokens_seen": 114591040,
      "step": 5324,
      "time_per_iteration": 2.922257661819458
    },
    {
      "auxiliary_loss_clip": 0.01096338,
      "auxiliary_loss_mlp": 0.0103017,
      "balance_loss_clip": 1.0416286,
      "balance_loss_mlp": 1.02291,
      "epoch": 0.6402933926531594,
      "flos": 20662209838080.0,
      "grad_norm": 1.8445528298622493,
      "language_loss": 0.80207598,
      "learning_rate": 1.2104108958571346e-06,
      "loss": 0.82334113,
      "num_input_tokens_seen": 114609310,
      "step": 5325,
      "time_per_iteration": 2.795203685760498
    },
    {
      "auxiliary_loss_clip": 0.01157155,
      "auxiliary_loss_mlp": 0.01026778,
      "balance_loss_clip": 1.04815531,
      "balance_loss_mlp": 1.01999474,
      "epoch": 0.6404136355437985,
      "flos": 24863327614080.0,
      "grad_norm": 1.4590001563982964,
      "language_loss": 0.75836217,
      "learning_rate": 1.2096952604304975e-06,
      "loss": 0.78020144,
      "num_input_tokens_seen": 114629740,
      "step": 5326,
      "time_per_iteration": 2.6378660202026367
    },
    {
      "auxiliary_loss_clip": 0.0115992,
      "auxiliary_loss_mlp": 0.01028004,
      "balance_loss_clip": 1.04558539,
      "balance_loss_mlp": 1.02039886,
      "epoch": 0.6405338784344375,
      "flos": 40479548901120.0,
      "grad_norm": 1.938829284381412,
      "language_loss": 0.70430952,
      "learning_rate": 1.2089797448924616e-06,
      "loss": 0.72618878,
      "num_input_tokens_seen": 114653615,
      "step": 5327,
      "time_per_iteration": 2.9087090492248535
    },
    {
      "auxiliary_loss_clip": 0.01120455,
      "auxiliary_loss_mlp": 0.01028842,
      "balance_loss_clip": 1.04040921,
      "balance_loss_mlp": 1.02136779,
      "epoch": 0.6406541213250767,
      "flos": 20886041439360.0,
      "grad_norm": 2.656040116407189,
      "language_loss": 0.65939748,
      "learning_rate": 1.2082643493515692e-06,
      "loss": 0.68089044,
      "num_input_tokens_seen": 114671935,
      "step": 5328,
      "time_per_iteration": 2.7894093990325928
    },
    {
      "auxiliary_loss_clip": 0.01159272,
      "auxiliary_loss_mlp": 0.01025569,
      "balance_loss_clip": 1.04792619,
      "balance_loss_mlp": 1.01824045,
      "epoch": 0.6407743642157158,
      "flos": 23295970679040.0,
      "grad_norm": 2.059808614781452,
      "language_loss": 0.8161419,
      "learning_rate": 1.207549073916346e-06,
      "loss": 0.83799028,
      "num_input_tokens_seen": 114692870,
      "step": 5329,
      "time_per_iteration": 2.7692503929138184
    },
    {
      "auxiliary_loss_clip": 0.01136793,
      "auxiliary_loss_mlp": 0.01023226,
      "balance_loss_clip": 1.04543293,
      "balance_loss_mlp": 1.01607907,
      "epoch": 0.6408946071063548,
      "flos": 15012636122880.0,
      "grad_norm": 2.5858881972164154,
      "language_loss": 0.77797782,
      "learning_rate": 1.2068339186952976e-06,
      "loss": 0.79957807,
      "num_input_tokens_seen": 114710410,
      "step": 5330,
      "time_per_iteration": 2.664731502532959
    },
    {
      "auxiliary_loss_clip": 0.01164942,
      "auxiliary_loss_mlp": 0.01025934,
      "balance_loss_clip": 1.04909468,
      "balance_loss_mlp": 1.01810765,
      "epoch": 0.6410148499969939,
      "flos": 22528595496960.0,
      "grad_norm": 2.022033454527087,
      "language_loss": 0.73591536,
      "learning_rate": 1.2061188837969136e-06,
      "loss": 0.75782406,
      "num_input_tokens_seen": 114730020,
      "step": 5331,
      "time_per_iteration": 2.6846861839294434
    },
    {
      "auxiliary_loss_clip": 0.01126116,
      "auxiliary_loss_mlp": 0.01025968,
      "balance_loss_clip": 1.04050219,
      "balance_loss_mlp": 1.01835036,
      "epoch": 0.641135092887633,
      "flos": 12422004537600.0,
      "grad_norm": 2.342157508519514,
      "language_loss": 0.84378564,
      "learning_rate": 1.2054039693296631e-06,
      "loss": 0.86530644,
      "num_input_tokens_seen": 114748015,
      "step": 5332,
      "time_per_iteration": 2.660792350769043
    },
    {
      "auxiliary_loss_clip": 0.0112665,
      "auxiliary_loss_mlp": 0.01024171,
      "balance_loss_clip": 1.04358208,
      "balance_loss_mlp": 1.01672626,
      "epoch": 0.6412553357782721,
      "flos": 22127329687680.0,
      "grad_norm": 1.639156255112616,
      "language_loss": 0.81725526,
      "learning_rate": 1.2046891754019992e-06,
      "loss": 0.83876348,
      "num_input_tokens_seen": 114768625,
      "step": 5333,
      "time_per_iteration": 2.844221353530884
    },
    {
      "auxiliary_loss_clip": 0.01159995,
      "auxiliary_loss_mlp": 0.01028302,
      "balance_loss_clip": 1.04634595,
      "balance_loss_mlp": 1.0206461,
      "epoch": 0.6413755786689112,
      "flos": 15888605097600.0,
      "grad_norm": 2.40033997367305,
      "language_loss": 0.82681113,
      "learning_rate": 1.2039745021223548e-06,
      "loss": 0.84869409,
      "num_input_tokens_seen": 114786045,
      "step": 5334,
      "time_per_iteration": 2.5942797660827637
    },
    {
      "auxiliary_loss_clip": 0.0102952,
      "auxiliary_loss_mlp": 0.01002306,
      "balance_loss_clip": 1.0182972,
      "balance_loss_mlp": 1.00113201,
      "epoch": 0.6414958215595503,
      "flos": 68039159955840.0,
      "grad_norm": 0.788256646346243,
      "language_loss": 0.57024866,
      "learning_rate": 1.2032599495991456e-06,
      "loss": 0.59056693,
      "num_input_tokens_seen": 114850785,
      "step": 5335,
      "time_per_iteration": 3.3964450359344482
    },
    {
      "auxiliary_loss_clip": 0.011604,
      "auxiliary_loss_mlp": 0.01022576,
      "balance_loss_clip": 1.04789495,
      "balance_loss_mlp": 1.01513433,
      "epoch": 0.6416160644501894,
      "flos": 44091300320640.0,
      "grad_norm": 1.76425795870613,
      "language_loss": 0.69726133,
      "learning_rate": 1.2025455179407685e-06,
      "loss": 0.71909112,
      "num_input_tokens_seen": 114871945,
      "step": 5336,
      "time_per_iteration": 2.86095929145813
    },
    {
      "auxiliary_loss_clip": 0.01155821,
      "auxiliary_loss_mlp": 0.00762644,
      "balance_loss_clip": 1.0478338,
      "balance_loss_mlp": 1.00054431,
      "epoch": 0.6417363073408284,
      "flos": 20959837931520.0,
      "grad_norm": 2.063697684325237,
      "language_loss": 0.73905039,
      "learning_rate": 1.2018312072556022e-06,
      "loss": 0.75823504,
      "num_input_tokens_seen": 114890445,
      "step": 5337,
      "time_per_iteration": 3.5729243755340576
    },
    {
      "auxiliary_loss_clip": 0.01170587,
      "auxiliary_loss_mlp": 0.00762284,
      "balance_loss_clip": 1.04721975,
      "balance_loss_mlp": 1.00055599,
      "epoch": 0.6418565502314676,
      "flos": 22455122227200.0,
      "grad_norm": 2.1030914912776986,
      "language_loss": 0.74635285,
      "learning_rate": 1.2011170176520077e-06,
      "loss": 0.76568151,
      "num_input_tokens_seen": 114911360,
      "step": 5338,
      "time_per_iteration": 2.669750690460205
    },
    {
      "auxiliary_loss_clip": 0.01085651,
      "auxiliary_loss_mlp": 0.01027062,
      "balance_loss_clip": 1.03789282,
      "balance_loss_mlp": 1.01955819,
      "epoch": 0.6419767931221066,
      "flos": 25045502417280.0,
      "grad_norm": 1.9703900969457737,
      "language_loss": 0.81288373,
      "learning_rate": 1.2004029492383256e-06,
      "loss": 0.8340109,
      "num_input_tokens_seen": 114932700,
      "step": 5339,
      "time_per_iteration": 2.810675859451294
    },
    {
      "auxiliary_loss_clip": 0.01156756,
      "auxiliary_loss_mlp": 0.01030442,
      "balance_loss_clip": 1.04779112,
      "balance_loss_mlp": 1.02319729,
      "epoch": 0.6420970360127457,
      "flos": 19463691709440.0,
      "grad_norm": 2.170590560732831,
      "language_loss": 0.73677218,
      "learning_rate": 1.1996890021228814e-06,
      "loss": 0.75864422,
      "num_input_tokens_seen": 114949475,
      "step": 5340,
      "time_per_iteration": 4.567188739776611
    },
    {
      "auxiliary_loss_clip": 0.0113898,
      "auxiliary_loss_mlp": 0.01026927,
      "balance_loss_clip": 1.04254699,
      "balance_loss_mlp": 1.01958919,
      "epoch": 0.6422172789033849,
      "flos": 40406147458560.0,
      "grad_norm": 1.5906500217097792,
      "language_loss": 0.69832337,
      "learning_rate": 1.1989751764139785e-06,
      "loss": 0.71998245,
      "num_input_tokens_seen": 114973125,
      "step": 5341,
      "time_per_iteration": 2.8053700923919678
    },
    {
      "auxiliary_loss_clip": 0.01112559,
      "auxiliary_loss_mlp": 0.0102459,
      "balance_loss_clip": 1.03769433,
      "balance_loss_mlp": 1.01711881,
      "epoch": 0.6423375217940239,
      "flos": 27672870637440.0,
      "grad_norm": 1.7448467086093649,
      "language_loss": 0.83006179,
      "learning_rate": 1.1982614722199044e-06,
      "loss": 0.85143328,
      "num_input_tokens_seen": 114994300,
      "step": 5342,
      "time_per_iteration": 2.772854804992676
    },
    {
      "auxiliary_loss_clip": 0.01149223,
      "auxiliary_loss_mlp": 0.01025057,
      "balance_loss_clip": 1.04457331,
      "balance_loss_mlp": 1.01786304,
      "epoch": 0.642457764684663,
      "flos": 18369242259840.0,
      "grad_norm": 2.269767469478735,
      "language_loss": 0.7776804,
      "learning_rate": 1.1975478896489276e-06,
      "loss": 0.79942322,
      "num_input_tokens_seen": 115012135,
      "step": 5343,
      "time_per_iteration": 2.7478439807891846
    },
    {
      "auxiliary_loss_clip": 0.01170525,
      "auxiliary_loss_mlp": 0.01029527,
      "balance_loss_clip": 1.04794669,
      "balance_loss_mlp": 1.02266681,
      "epoch": 0.6425780075753021,
      "flos": 19750509809280.0,
      "grad_norm": 2.042513485887519,
      "language_loss": 0.76595056,
      "learning_rate": 1.1968344288092981e-06,
      "loss": 0.78795111,
      "num_input_tokens_seen": 115028715,
      "step": 5344,
      "time_per_iteration": 3.436119318008423
    },
    {
      "auxiliary_loss_clip": 0.0115733,
      "auxiliary_loss_mlp": 0.00762572,
      "balance_loss_clip": 1.04792917,
      "balance_loss_mlp": 1.00066936,
      "epoch": 0.6426982504659412,
      "flos": 20558536208640.0,
      "grad_norm": 1.7391774110332807,
      "language_loss": 0.64619517,
      "learning_rate": 1.1961210898092468e-06,
      "loss": 0.66539419,
      "num_input_tokens_seen": 115047665,
      "step": 5345,
      "time_per_iteration": 2.6262311935424805
    },
    {
      "auxiliary_loss_clip": 0.01150191,
      "auxiliary_loss_mlp": 0.01026694,
      "balance_loss_clip": 1.04644859,
      "balance_loss_mlp": 1.01834619,
      "epoch": 0.6428184933565803,
      "flos": 17851984456320.0,
      "grad_norm": 2.676841641594589,
      "language_loss": 0.7957865,
      "learning_rate": 1.1954078727569874e-06,
      "loss": 0.81755531,
      "num_input_tokens_seen": 115064965,
      "step": 5346,
      "time_per_iteration": 2.70442533493042
    },
    {
      "auxiliary_loss_clip": 0.01137285,
      "auxiliary_loss_mlp": 0.00762464,
      "balance_loss_clip": 1.04552233,
      "balance_loss_mlp": 1.00056553,
      "epoch": 0.6429387362472194,
      "flos": 22456953820800.0,
      "grad_norm": 1.5705360765861687,
      "language_loss": 0.78144526,
      "learning_rate": 1.1946947777607141e-06,
      "loss": 0.80044276,
      "num_input_tokens_seen": 115086100,
      "step": 5347,
      "time_per_iteration": 2.7072107791900635
    },
    {
      "auxiliary_loss_clip": 0.01108173,
      "auxiliary_loss_mlp": 0.01025181,
      "balance_loss_clip": 1.03998256,
      "balance_loss_mlp": 1.01758099,
      "epoch": 0.6430589791378585,
      "flos": 24752579005440.0,
      "grad_norm": 2.0372635490273057,
      "language_loss": 0.80356467,
      "learning_rate": 1.1939818049286024e-06,
      "loss": 0.82489812,
      "num_input_tokens_seen": 115104260,
      "step": 5348,
      "time_per_iteration": 2.7729461193084717
    },
    {
      "auxiliary_loss_clip": 0.01088804,
      "auxiliary_loss_mlp": 0.01025379,
      "balance_loss_clip": 1.03917503,
      "balance_loss_mlp": 1.01836908,
      "epoch": 0.6431792220284975,
      "flos": 24901249397760.0,
      "grad_norm": 1.563965092637992,
      "language_loss": 0.75819218,
      "learning_rate": 1.1932689543688101e-06,
      "loss": 0.77933395,
      "num_input_tokens_seen": 115125365,
      "step": 5349,
      "time_per_iteration": 2.7750625610351562
    },
    {
      "auxiliary_loss_clip": 0.01145894,
      "auxiliary_loss_mlp": 0.01023075,
      "balance_loss_clip": 1.04834282,
      "balance_loss_mlp": 1.0155623,
      "epoch": 0.6432994649191367,
      "flos": 21032305620480.0,
      "grad_norm": 2.0100338099152957,
      "language_loss": 0.72439885,
      "learning_rate": 1.1925562261894756e-06,
      "loss": 0.7460885,
      "num_input_tokens_seen": 115144445,
      "step": 5350,
      "time_per_iteration": 2.7003867626190186
    },
    {
      "auxiliary_loss_clip": 0.01138003,
      "auxiliary_loss_mlp": 0.01026935,
      "balance_loss_clip": 1.04225433,
      "balance_loss_mlp": 1.020015,
      "epoch": 0.6434197078097758,
      "flos": 30884433655680.0,
      "grad_norm": 1.8184370871729474,
      "language_loss": 0.77702719,
      "learning_rate": 1.1918436204987207e-06,
      "loss": 0.79867655,
      "num_input_tokens_seen": 115166305,
      "step": 5351,
      "time_per_iteration": 2.7806246280670166
    },
    {
      "auxiliary_loss_clip": 0.01154874,
      "auxiliary_loss_mlp": 0.01025233,
      "balance_loss_clip": 1.04820895,
      "balance_loss_mlp": 1.01785064,
      "epoch": 0.6435399507004148,
      "flos": 15012492468480.0,
      "grad_norm": 2.056899739662769,
      "language_loss": 0.81682646,
      "learning_rate": 1.191131137404645e-06,
      "loss": 0.83862752,
      "num_input_tokens_seen": 115183045,
      "step": 5352,
      "time_per_iteration": 2.649717092514038
    },
    {
      "auxiliary_loss_clip": 0.01118949,
      "auxiliary_loss_mlp": 0.01027643,
      "balance_loss_clip": 1.04208505,
      "balance_loss_mlp": 1.02077365,
      "epoch": 0.643660193591054,
      "flos": 19901981462400.0,
      "grad_norm": 1.78680452105505,
      "language_loss": 0.7746706,
      "learning_rate": 1.190418777015333e-06,
      "loss": 0.79613656,
      "num_input_tokens_seen": 115201955,
      "step": 5353,
      "time_per_iteration": 2.694457530975342
    },
    {
      "auxiliary_loss_clip": 0.01144403,
      "auxiliary_loss_mlp": 0.01021693,
      "balance_loss_clip": 1.04655051,
      "balance_loss_mlp": 1.01486814,
      "epoch": 0.643780436481693,
      "flos": 24133622820480.0,
      "grad_norm": 1.4943752766375038,
      "language_loss": 0.73403805,
      "learning_rate": 1.1897065394388487e-06,
      "loss": 0.75569898,
      "num_input_tokens_seen": 115222395,
      "step": 5354,
      "time_per_iteration": 2.702378511428833
    },
    {
      "auxiliary_loss_clip": 0.01148265,
      "auxiliary_loss_mlp": 0.01028956,
      "balance_loss_clip": 1.05039942,
      "balance_loss_mlp": 1.02148151,
      "epoch": 0.6439006793723321,
      "flos": 23148808657920.0,
      "grad_norm": 1.4912080869991633,
      "language_loss": 0.76575661,
      "learning_rate": 1.1889944247832385e-06,
      "loss": 0.78752887,
      "num_input_tokens_seen": 115242635,
      "step": 5355,
      "time_per_iteration": 2.7087044715881348
    },
    {
      "auxiliary_loss_clip": 0.01161436,
      "auxiliary_loss_mlp": 0.01027851,
      "balance_loss_clip": 1.04634571,
      "balance_loss_mlp": 1.02054954,
      "epoch": 0.6440209222629713,
      "flos": 23617909301760.0,
      "grad_norm": 2.0959938159232556,
      "language_loss": 0.70790398,
      "learning_rate": 1.1882824331565283e-06,
      "loss": 0.72979689,
      "num_input_tokens_seen": 115262095,
      "step": 5356,
      "time_per_iteration": 2.683272361755371
    },
    {
      "auxiliary_loss_clip": 0.01124464,
      "auxiliary_loss_mlp": 0.01024784,
      "balance_loss_clip": 1.04017138,
      "balance_loss_mlp": 1.01733327,
      "epoch": 0.6441411651536103,
      "flos": 16544872535040.0,
      "grad_norm": 2.062048874571363,
      "language_loss": 0.89189792,
      "learning_rate": 1.1875705646667287e-06,
      "loss": 0.9133904,
      "num_input_tokens_seen": 115279985,
      "step": 5357,
      "time_per_iteration": 2.7670819759368896
    },
    {
      "auxiliary_loss_clip": 0.01153797,
      "auxiliary_loss_mlp": 0.01025196,
      "balance_loss_clip": 1.04364038,
      "balance_loss_mlp": 1.01761711,
      "epoch": 0.6442614080442494,
      "flos": 25410965345280.0,
      "grad_norm": 3.400425634250684,
      "language_loss": 0.75669146,
      "learning_rate": 1.1868588194218282e-06,
      "loss": 0.77848148,
      "num_input_tokens_seen": 115300365,
      "step": 5358,
      "time_per_iteration": 2.672905445098877
    },
    {
      "auxiliary_loss_clip": 0.01149685,
      "auxiliary_loss_mlp": 0.01031557,
      "balance_loss_clip": 1.04421461,
      "balance_loss_mlp": 1.02336156,
      "epoch": 0.6443816509348885,
      "flos": 28294017552000.0,
      "grad_norm": 1.4901082172100037,
      "language_loss": 0.73926723,
      "learning_rate": 1.1861471975297979e-06,
      "loss": 0.76107961,
      "num_input_tokens_seen": 115322060,
      "step": 5359,
      "time_per_iteration": 2.826045274734497
    },
    {
      "auxiliary_loss_clip": 0.01127298,
      "auxiliary_loss_mlp": 0.01026704,
      "balance_loss_clip": 1.04403734,
      "balance_loss_mlp": 1.01946807,
      "epoch": 0.6445018938255276,
      "flos": 36690075964800.0,
      "grad_norm": 2.3990169971496846,
      "language_loss": 0.71077365,
      "learning_rate": 1.185435699098591e-06,
      "loss": 0.73231363,
      "num_input_tokens_seen": 115348255,
      "step": 5360,
      "time_per_iteration": 2.884958267211914
    },
    {
      "auxiliary_loss_clip": 0.01147157,
      "auxiliary_loss_mlp": 0.01023726,
      "balance_loss_clip": 1.04547536,
      "balance_loss_mlp": 1.01559031,
      "epoch": 0.6446221367161666,
      "flos": 14501411804160.0,
      "grad_norm": 2.2496068432067675,
      "language_loss": 0.78449708,
      "learning_rate": 1.1847243242361403e-06,
      "loss": 0.80620587,
      "num_input_tokens_seen": 115366845,
      "step": 5361,
      "time_per_iteration": 2.6301448345184326
    },
    {
      "auxiliary_loss_clip": 0.01145422,
      "auxiliary_loss_mlp": 0.01021795,
      "balance_loss_clip": 1.04507196,
      "balance_loss_mlp": 1.01356339,
      "epoch": 0.6447423796068057,
      "flos": 24609367480320.0,
      "grad_norm": 1.5649113824567857,
      "language_loss": 0.78285044,
      "learning_rate": 1.1840130730503624e-06,
      "loss": 0.80452263,
      "num_input_tokens_seen": 115388125,
      "step": 5362,
      "time_per_iteration": 2.726356267929077
    },
    {
      "auxiliary_loss_clip": 0.01171805,
      "auxiliary_loss_mlp": 0.01022534,
      "balance_loss_clip": 1.04729223,
      "balance_loss_mlp": 1.01476693,
      "epoch": 0.6448626224974449,
      "flos": 25047298097280.0,
      "grad_norm": 1.9315023608156254,
      "language_loss": 0.75241446,
      "learning_rate": 1.1833019456491518e-06,
      "loss": 0.7743578,
      "num_input_tokens_seen": 115409655,
      "step": 5363,
      "time_per_iteration": 3.5432562828063965
    },
    {
      "auxiliary_loss_clip": 0.0115861,
      "auxiliary_loss_mlp": 0.01028924,
      "balance_loss_clip": 1.04617751,
      "balance_loss_mlp": 1.02152693,
      "epoch": 0.6449828653880839,
      "flos": 22530355263360.0,
      "grad_norm": 2.5329935574136164,
      "language_loss": 0.79068565,
      "learning_rate": 1.1825909421403871e-06,
      "loss": 0.81256104,
      "num_input_tokens_seen": 115428750,
      "step": 5364,
      "time_per_iteration": 2.677765130996704
    },
    {
      "auxiliary_loss_clip": 0.01160076,
      "auxiliary_loss_mlp": 0.01023817,
      "balance_loss_clip": 1.04730439,
      "balance_loss_mlp": 1.01676846,
      "epoch": 0.645103108278723,
      "flos": 25695736369920.0,
      "grad_norm": 1.7365938867513604,
      "language_loss": 0.76530206,
      "learning_rate": 1.181880062631926e-06,
      "loss": 0.78714097,
      "num_input_tokens_seen": 115448085,
      "step": 5365,
      "time_per_iteration": 2.6400818824768066
    },
    {
      "auxiliary_loss_clip": 0.01137887,
      "auxiliary_loss_mlp": 0.01031518,
      "balance_loss_clip": 1.04567862,
      "balance_loss_mlp": 1.02371609,
      "epoch": 0.6452233511693621,
      "flos": 27450331925760.0,
      "grad_norm": 2.0865732231088625,
      "language_loss": 0.84499347,
      "learning_rate": 1.1811693072316093e-06,
      "loss": 0.86668754,
      "num_input_tokens_seen": 115465765,
      "step": 5366,
      "time_per_iteration": 3.6165668964385986
    },
    {
      "auxiliary_loss_clip": 0.01172568,
      "auxiliary_loss_mlp": 0.00762807,
      "balance_loss_clip": 1.04796934,
      "balance_loss_mlp": 1.0004921,
      "epoch": 0.6453435940600012,
      "flos": 19208618254080.0,
      "grad_norm": 2.5388547274568136,
      "language_loss": 0.84377205,
      "learning_rate": 1.1804586760472574e-06,
      "loss": 0.86312586,
      "num_input_tokens_seen": 115482230,
      "step": 5367,
      "time_per_iteration": 2.614454984664917
    },
    {
      "auxiliary_loss_clip": 0.01127642,
      "auxiliary_loss_mlp": 0.0102078,
      "balance_loss_clip": 1.04290581,
      "balance_loss_mlp": 1.01383281,
      "epoch": 0.6454638369506402,
      "flos": 25737680476800.0,
      "grad_norm": 2.199759009682726,
      "language_loss": 0.80076301,
      "learning_rate": 1.1797481691866736e-06,
      "loss": 0.82224733,
      "num_input_tokens_seen": 115499455,
      "step": 5368,
      "time_per_iteration": 2.783536434173584
    },
    {
      "auxiliary_loss_clip": 0.01135192,
      "auxiliary_loss_mlp": 0.01025577,
      "balance_loss_clip": 1.04511857,
      "balance_loss_mlp": 1.0185411,
      "epoch": 0.6455840798412794,
      "flos": 20989176364800.0,
      "grad_norm": 1.9244119840657354,
      "language_loss": 0.83499479,
      "learning_rate": 1.1790377867576393e-06,
      "loss": 0.85660255,
      "num_input_tokens_seen": 115517205,
      "step": 5369,
      "time_per_iteration": 2.707237958908081
    },
    {
      "auxiliary_loss_clip": 0.01145322,
      "auxiliary_loss_mlp": 0.01024524,
      "balance_loss_clip": 1.04304314,
      "balance_loss_mlp": 1.01680529,
      "epoch": 0.6457043227319185,
      "flos": 26067556005120.0,
      "grad_norm": 2.1214243686687397,
      "language_loss": 0.76618493,
      "learning_rate": 1.1783275288679203e-06,
      "loss": 0.78788334,
      "num_input_tokens_seen": 115534370,
      "step": 5370,
      "time_per_iteration": 3.6010429859161377
    },
    {
      "auxiliary_loss_clip": 0.01065607,
      "auxiliary_loss_mlp": 0.01000759,
      "balance_loss_clip": 1.01591372,
      "balance_loss_mlp": 0.99952519,
      "epoch": 0.6458245656225575,
      "flos": 60370831088640.0,
      "grad_norm": 0.8434385549710722,
      "language_loss": 0.57112616,
      "learning_rate": 1.177617395625262e-06,
      "loss": 0.59178984,
      "num_input_tokens_seen": 115592345,
      "step": 5371,
      "time_per_iteration": 3.1792173385620117
    },
    {
      "auxiliary_loss_clip": 0.01157239,
      "auxiliary_loss_mlp": 0.0102442,
      "balance_loss_clip": 1.04744673,
      "balance_loss_mlp": 1.01751781,
      "epoch": 0.6459448085131967,
      "flos": 23076771932160.0,
      "grad_norm": 1.7551207272230125,
      "language_loss": 0.75131011,
      "learning_rate": 1.1769073871373908e-06,
      "loss": 0.77312672,
      "num_input_tokens_seen": 115612550,
      "step": 5372,
      "time_per_iteration": 2.7512664794921875
    },
    {
      "auxiliary_loss_clip": 0.01126117,
      "auxiliary_loss_mlp": 0.01026109,
      "balance_loss_clip": 1.04085863,
      "balance_loss_mlp": 1.01887631,
      "epoch": 0.6460650514038357,
      "flos": 22598190097920.0,
      "grad_norm": 1.6600787171452684,
      "language_loss": 0.8398149,
      "learning_rate": 1.176197503512015e-06,
      "loss": 0.86133718,
      "num_input_tokens_seen": 115632265,
      "step": 5373,
      "time_per_iteration": 2.6576781272888184
    },
    {
      "auxiliary_loss_clip": 0.01140494,
      "auxiliary_loss_mlp": 0.0102508,
      "balance_loss_clip": 1.04493272,
      "balance_loss_mlp": 1.0184257,
      "epoch": 0.6461852942944748,
      "flos": 20266726118400.0,
      "grad_norm": 2.1542696555323886,
      "language_loss": 0.82502747,
      "learning_rate": 1.1754877448568223e-06,
      "loss": 0.8466832,
      "num_input_tokens_seen": 115651720,
      "step": 5374,
      "time_per_iteration": 2.673281192779541
    },
    {
      "auxiliary_loss_clip": 0.01145753,
      "auxiliary_loss_mlp": 0.01023188,
      "balance_loss_clip": 1.04633498,
      "balance_loss_mlp": 1.01609468,
      "epoch": 0.646305537185114,
      "flos": 23367109564800.0,
      "grad_norm": 2.0571233315915074,
      "language_loss": 0.89899641,
      "learning_rate": 1.1747781112794837e-06,
      "loss": 0.92068577,
      "num_input_tokens_seen": 115668215,
      "step": 5375,
      "time_per_iteration": 2.6856021881103516
    },
    {
      "auxiliary_loss_clip": 0.0112873,
      "auxiliary_loss_mlp": 0.01029758,
      "balance_loss_clip": 1.04537749,
      "balance_loss_mlp": 1.02239394,
      "epoch": 0.646425780075753,
      "flos": 24277480790400.0,
      "grad_norm": 1.6753893931179573,
      "language_loss": 0.83064592,
      "learning_rate": 1.1740686028876487e-06,
      "loss": 0.85223079,
      "num_input_tokens_seen": 115687080,
      "step": 5376,
      "time_per_iteration": 2.7336153984069824
    },
    {
      "auxiliary_loss_clip": 0.01153482,
      "auxiliary_loss_mlp": 0.01023194,
      "balance_loss_clip": 1.04701662,
      "balance_loss_mlp": 1.01630116,
      "epoch": 0.6465460229663921,
      "flos": 20813968800000.0,
      "grad_norm": 2.5946284431008664,
      "language_loss": 0.75111806,
      "learning_rate": 1.1733592197889507e-06,
      "loss": 0.77288485,
      "num_input_tokens_seen": 115703990,
      "step": 5377,
      "time_per_iteration": 2.651505708694458
    },
    {
      "auxiliary_loss_clip": 0.01152227,
      "auxiliary_loss_mlp": 0.01030671,
      "balance_loss_clip": 1.04684377,
      "balance_loss_mlp": 1.02424288,
      "epoch": 0.6466662658570312,
      "flos": 22853299466880.0,
      "grad_norm": 1.8094205101710146,
      "language_loss": 0.72773337,
      "learning_rate": 1.1726499620910014e-06,
      "loss": 0.74956238,
      "num_input_tokens_seen": 115724270,
      "step": 5378,
      "time_per_iteration": 2.6381213665008545
    },
    {
      "auxiliary_loss_clip": 0.01155724,
      "auxiliary_loss_mlp": 0.01024297,
      "balance_loss_clip": 1.04727411,
      "balance_loss_mlp": 1.01688802,
      "epoch": 0.6467865087476703,
      "flos": 15304553953920.0,
      "grad_norm": 2.0303361256069183,
      "language_loss": 0.777053,
      "learning_rate": 1.1719408299013955e-06,
      "loss": 0.79885322,
      "num_input_tokens_seen": 115742995,
      "step": 5379,
      "time_per_iteration": 2.5652832984924316
    },
    {
      "auxiliary_loss_clip": 0.0117156,
      "auxiliary_loss_mlp": 0.01023965,
      "balance_loss_clip": 1.04996824,
      "balance_loss_mlp": 1.01718473,
      "epoch": 0.6469067516383094,
      "flos": 19573650218880.0,
      "grad_norm": 2.762577765048556,
      "language_loss": 0.75946212,
      "learning_rate": 1.1712318233277067e-06,
      "loss": 0.78141737,
      "num_input_tokens_seen": 115762015,
      "step": 5380,
      "time_per_iteration": 2.6073129177093506
    },
    {
      "auxiliary_loss_clip": 0.01064556,
      "auxiliary_loss_mlp": 0.010011,
      "balance_loss_clip": 1.01540661,
      "balance_loss_mlp": 0.99994355,
      "epoch": 0.6470269945289485,
      "flos": 65098002522240.0,
      "grad_norm": 0.7589278556937867,
      "language_loss": 0.57853854,
      "learning_rate": 1.1705229424774916e-06,
      "loss": 0.59919512,
      "num_input_tokens_seen": 115816285,
      "step": 5381,
      "time_per_iteration": 3.082566738128662
    },
    {
      "auxiliary_loss_clip": 0.01139932,
      "auxiliary_loss_mlp": 0.01026484,
      "balance_loss_clip": 1.04369426,
      "balance_loss_mlp": 1.01919484,
      "epoch": 0.6471472374195876,
      "flos": 30696943639680.0,
      "grad_norm": 1.7766920374217674,
      "language_loss": 0.64240742,
      "learning_rate": 1.1698141874582867e-06,
      "loss": 0.66407156,
      "num_input_tokens_seen": 115837330,
      "step": 5382,
      "time_per_iteration": 2.7807562351226807
    },
    {
      "auxiliary_loss_clip": 0.01168423,
      "auxiliary_loss_mlp": 0.01029899,
      "balance_loss_clip": 1.04828072,
      "balance_loss_mlp": 1.02307475,
      "epoch": 0.6472674803102266,
      "flos": 20521835487360.0,
      "grad_norm": 2.622466362224973,
      "language_loss": 0.72253215,
      "learning_rate": 1.169105558377609e-06,
      "loss": 0.74451542,
      "num_input_tokens_seen": 115857420,
      "step": 5383,
      "time_per_iteration": 2.6052772998809814
    },
    {
      "auxiliary_loss_clip": 0.01114848,
      "auxiliary_loss_mlp": 0.00762527,
      "balance_loss_clip": 1.04684377,
      "balance_loss_mlp": 1.00050974,
      "epoch": 0.6473877232008658,
      "flos": 24715447320960.0,
      "grad_norm": 1.745143412095823,
      "language_loss": 0.78389323,
      "learning_rate": 1.1683970553429587e-06,
      "loss": 0.8026669,
      "num_input_tokens_seen": 115878875,
      "step": 5384,
      "time_per_iteration": 2.7730660438537598
    },
    {
      "auxiliary_loss_clip": 0.01132576,
      "auxiliary_loss_mlp": 0.01031975,
      "balance_loss_clip": 1.04480267,
      "balance_loss_mlp": 1.02465534,
      "epoch": 0.6475079660915048,
      "flos": 15885552441600.0,
      "grad_norm": 1.7826656202278064,
      "language_loss": 0.81935829,
      "learning_rate": 1.1676886784618128e-06,
      "loss": 0.84100378,
      "num_input_tokens_seen": 115895540,
      "step": 5385,
      "time_per_iteration": 2.6778366565704346
    },
    {
      "auxiliary_loss_clip": 0.01156847,
      "auxiliary_loss_mlp": 0.01022849,
      "balance_loss_clip": 1.04699314,
      "balance_loss_mlp": 1.0151124,
      "epoch": 0.6476282089821439,
      "flos": 17381590922880.0,
      "grad_norm": 2.785706232897928,
      "language_loss": 0.84173679,
      "learning_rate": 1.1669804278416332e-06,
      "loss": 0.86353374,
      "num_input_tokens_seen": 115910265,
      "step": 5386,
      "time_per_iteration": 2.603102207183838
    },
    {
      "auxiliary_loss_clip": 0.01149008,
      "auxiliary_loss_mlp": 0.01027174,
      "balance_loss_clip": 1.04705393,
      "balance_loss_mlp": 1.0190084,
      "epoch": 0.6477484518727831,
      "flos": 20194078861440.0,
      "grad_norm": 1.8791672616888675,
      "language_loss": 0.71299517,
      "learning_rate": 1.1662723035898602e-06,
      "loss": 0.73475701,
      "num_input_tokens_seen": 115930025,
      "step": 5387,
      "time_per_iteration": 2.6967480182647705
    },
    {
      "auxiliary_loss_clip": 0.01158085,
      "auxiliary_loss_mlp": 0.01026651,
      "balance_loss_clip": 1.04805946,
      "balance_loss_mlp": 1.01870561,
      "epoch": 0.6478686947634221,
      "flos": 25410426641280.0,
      "grad_norm": 1.9906512472541673,
      "language_loss": 0.82071292,
      "learning_rate": 1.165564305813915e-06,
      "loss": 0.84256029,
      "num_input_tokens_seen": 115949025,
      "step": 5388,
      "time_per_iteration": 2.6486003398895264
    },
    {
      "auxiliary_loss_clip": 0.01154544,
      "auxiliary_loss_mlp": 0.01023304,
      "balance_loss_clip": 1.0463239,
      "balance_loss_mlp": 1.01638651,
      "epoch": 0.6479889376540612,
      "flos": 20083581648000.0,
      "grad_norm": 1.724494940142618,
      "language_loss": 0.81310701,
      "learning_rate": 1.1648564346212019e-06,
      "loss": 0.83488548,
      "num_input_tokens_seen": 115968145,
      "step": 5389,
      "time_per_iteration": 3.5773353576660156
    },
    {
      "auxiliary_loss_clip": 0.01155133,
      "auxiliary_loss_mlp": 0.01023854,
      "balance_loss_clip": 1.04952407,
      "balance_loss_mlp": 1.01712155,
      "epoch": 0.6481091805447003,
      "flos": 26758082039040.0,
      "grad_norm": 2.0638957706907353,
      "language_loss": 0.76039481,
      "learning_rate": 1.164148690119104e-06,
      "loss": 0.78218472,
      "num_input_tokens_seen": 115989425,
      "step": 5390,
      "time_per_iteration": 2.6782243251800537
    },
    {
      "auxiliary_loss_clip": 0.01170245,
      "auxiliary_loss_mlp": 0.01027856,
      "balance_loss_clip": 1.0484283,
      "balance_loss_mlp": 1.02068591,
      "epoch": 0.6482294234353394,
      "flos": 23952094462080.0,
      "grad_norm": 2.50375965284962,
      "language_loss": 0.7417559,
      "learning_rate": 1.163441072414985e-06,
      "loss": 0.76373696,
      "num_input_tokens_seen": 116009630,
      "step": 5391,
      "time_per_iteration": 2.625136137008667
    },
    {
      "auxiliary_loss_clip": 0.01158311,
      "auxiliary_loss_mlp": 0.01027435,
      "balance_loss_clip": 1.04985511,
      "balance_loss_mlp": 1.02055645,
      "epoch": 0.6483496663259785,
      "flos": 26209833776640.0,
      "grad_norm": 1.8415004086431124,
      "language_loss": 0.69824743,
      "learning_rate": 1.16273358161619e-06,
      "loss": 0.72010493,
      "num_input_tokens_seen": 116029965,
      "step": 5392,
      "time_per_iteration": 3.6492624282836914
    },
    {
      "auxiliary_loss_clip": 0.01150966,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 1.04654264,
      "balance_loss_mlp": 1.01960969,
      "epoch": 0.6484699092166175,
      "flos": 20922239370240.0,
      "grad_norm": 3.1760407728827507,
      "language_loss": 0.83272099,
      "learning_rate": 1.1620262178300446e-06,
      "loss": 0.85450161,
      "num_input_tokens_seen": 116048580,
      "step": 5393,
      "time_per_iteration": 3.5717647075653076
    },
    {
      "auxiliary_loss_clip": 0.01129393,
      "auxiliary_loss_mlp": 0.01024892,
      "balance_loss_clip": 1.04260683,
      "balance_loss_mlp": 1.01779866,
      "epoch": 0.6485901521072567,
      "flos": 33072865678080.0,
      "grad_norm": 1.6846624131637096,
      "language_loss": 0.75879163,
      "learning_rate": 1.1613189811638563e-06,
      "loss": 0.78033447,
      "num_input_tokens_seen": 116070305,
      "step": 5394,
      "time_per_iteration": 2.795661449432373
    },
    {
      "auxiliary_loss_clip": 0.01160925,
      "auxiliary_loss_mlp": 0.01020659,
      "balance_loss_clip": 1.04954934,
      "balance_loss_mlp": 1.01436806,
      "epoch": 0.6487103949978957,
      "flos": 22274060745600.0,
      "grad_norm": 1.5979339611823231,
      "language_loss": 0.78150076,
      "learning_rate": 1.1606118717249117e-06,
      "loss": 0.80331659,
      "num_input_tokens_seen": 116090405,
      "step": 5395,
      "time_per_iteration": 2.6128180027008057
    },
    {
      "auxiliary_loss_clip": 0.01173326,
      "auxiliary_loss_mlp": 0.01025245,
      "balance_loss_clip": 1.04685473,
      "balance_loss_mlp": 1.01746702,
      "epoch": 0.6488306378885348,
      "flos": 22930400010240.0,
      "grad_norm": 1.905316553607197,
      "language_loss": 0.67845178,
      "learning_rate": 1.1599048896204787e-06,
      "loss": 0.70043749,
      "num_input_tokens_seen": 116110285,
      "step": 5396,
      "time_per_iteration": 2.5943233966827393
    },
    {
      "auxiliary_loss_clip": 0.0113266,
      "auxiliary_loss_mlp": 0.01029409,
      "balance_loss_clip": 1.04477215,
      "balance_loss_mlp": 1.02202415,
      "epoch": 0.648950880779174,
      "flos": 20376110010240.0,
      "grad_norm": 1.7319305926143758,
      "language_loss": 0.80941182,
      "learning_rate": 1.1591980349578061e-06,
      "loss": 0.83103251,
      "num_input_tokens_seen": 116128955,
      "step": 5397,
      "time_per_iteration": 3.5727624893188477
    },
    {
      "auxiliary_loss_clip": 0.01038084,
      "auxiliary_loss_mlp": 0.0100091,
      "balance_loss_clip": 1.01132655,
      "balance_loss_mlp": 0.99975318,
      "epoch": 0.649071123669813,
      "flos": 59930889310080.0,
      "grad_norm": 0.735480428262205,
      "language_loss": 0.54254055,
      "learning_rate": 1.158491307844123e-06,
      "loss": 0.56293058,
      "num_input_tokens_seen": 116188875,
      "step": 5398,
      "time_per_iteration": 3.2304584980010986
    },
    {
      "auxiliary_loss_clip": 0.01143708,
      "auxiliary_loss_mlp": 0.01027668,
      "balance_loss_clip": 1.04708064,
      "balance_loss_mlp": 1.02061391,
      "epoch": 0.6491913665604521,
      "flos": 20446566537600.0,
      "grad_norm": 1.6317134464993608,
      "language_loss": 0.83845091,
      "learning_rate": 1.1577847083866387e-06,
      "loss": 0.8601647,
      "num_input_tokens_seen": 116207910,
      "step": 5399,
      "time_per_iteration": 2.709592580795288
    },
    {
      "auxiliary_loss_clip": 0.01132397,
      "auxiliary_loss_mlp": 0.01020527,
      "balance_loss_clip": 1.04224432,
      "balance_loss_mlp": 1.01310015,
      "epoch": 0.6493116094510912,
      "flos": 16946820702720.0,
      "grad_norm": 1.7623246428376285,
      "language_loss": 0.7204386,
      "learning_rate": 1.1570782366925453e-06,
      "loss": 0.74196786,
      "num_input_tokens_seen": 116226425,
      "step": 5400,
      "time_per_iteration": 2.642521858215332
    },
    {
      "auxiliary_loss_clip": 0.0114257,
      "auxiliary_loss_mlp": 0.01023367,
      "balance_loss_clip": 1.04100692,
      "balance_loss_mlp": 1.01579142,
      "epoch": 0.6494318523417303,
      "flos": 18802935072000.0,
      "grad_norm": 1.715697018064535,
      "language_loss": 0.75334656,
      "learning_rate": 1.1563718928690132e-06,
      "loss": 0.77500594,
      "num_input_tokens_seen": 116243860,
      "step": 5401,
      "time_per_iteration": 2.6233150959014893
    },
    {
      "auxiliary_loss_clip": 0.01128021,
      "auxiliary_loss_mlp": 0.01028948,
      "balance_loss_clip": 1.04357743,
      "balance_loss_mlp": 1.02189422,
      "epoch": 0.6495520952323693,
      "flos": 18982847318400.0,
      "grad_norm": 2.011850520844283,
      "language_loss": 0.71412146,
      "learning_rate": 1.1556656770231942e-06,
      "loss": 0.73569107,
      "num_input_tokens_seen": 116260055,
      "step": 5402,
      "time_per_iteration": 2.649190902709961
    },
    {
      "auxiliary_loss_clip": 0.01157813,
      "auxiliary_loss_mlp": 0.01024573,
      "balance_loss_clip": 1.04456699,
      "balance_loss_mlp": 1.01775169,
      "epoch": 0.6496723381230085,
      "flos": 22745388032640.0,
      "grad_norm": 2.3851315034547227,
      "language_loss": 0.76308608,
      "learning_rate": 1.1549595892622207e-06,
      "loss": 0.78490996,
      "num_input_tokens_seen": 116278825,
      "step": 5403,
      "time_per_iteration": 2.653712034225464
    },
    {
      "auxiliary_loss_clip": 0.01025039,
      "auxiliary_loss_mlp": 0.01004934,
      "balance_loss_clip": 1.01533735,
      "balance_loss_mlp": 1.00368822,
      "epoch": 0.6497925810136476,
      "flos": 62145283887360.0,
      "grad_norm": 0.872415412008942,
      "language_loss": 0.59044087,
      "learning_rate": 1.1542536296932047e-06,
      "loss": 0.61074066,
      "num_input_tokens_seen": 116342360,
      "step": 5404,
      "time_per_iteration": 3.2951741218566895
    },
    {
      "auxiliary_loss_clip": 0.01134161,
      "auxiliary_loss_mlp": 0.01032397,
      "balance_loss_clip": 1.04115021,
      "balance_loss_mlp": 1.02464259,
      "epoch": 0.6499128239042866,
      "flos": 20156731695360.0,
      "grad_norm": 1.7964754462553298,
      "language_loss": 0.70241433,
      "learning_rate": 1.1535477984232414e-06,
      "loss": 0.72407991,
      "num_input_tokens_seen": 116362235,
      "step": 5405,
      "time_per_iteration": 2.6729917526245117
    },
    {
      "auxiliary_loss_clip": 0.01115545,
      "auxiliary_loss_mlp": 0.01025985,
      "balance_loss_clip": 1.03858578,
      "balance_loss_mlp": 1.01922286,
      "epoch": 0.6500330667949258,
      "flos": 24462420940800.0,
      "grad_norm": 1.9955711547701482,
      "language_loss": 0.76900685,
      "learning_rate": 1.152842095559404e-06,
      "loss": 0.7904222,
      "num_input_tokens_seen": 116382895,
      "step": 5406,
      "time_per_iteration": 2.769484043121338
    },
    {
      "auxiliary_loss_clip": 0.01145621,
      "auxiliary_loss_mlp": 0.01023956,
      "balance_loss_clip": 1.04306126,
      "balance_loss_mlp": 1.01708984,
      "epoch": 0.6501533096855648,
      "flos": 25477399549440.0,
      "grad_norm": 1.726778454656137,
      "language_loss": 0.76670277,
      "learning_rate": 1.1521365212087474e-06,
      "loss": 0.7883985,
      "num_input_tokens_seen": 116402880,
      "step": 5407,
      "time_per_iteration": 2.660686492919922
    },
    {
      "auxiliary_loss_clip": 0.01158551,
      "auxiliary_loss_mlp": 0.01019927,
      "balance_loss_clip": 1.04721785,
      "balance_loss_mlp": 1.01285779,
      "epoch": 0.6502735525762039,
      "flos": 44819245347840.0,
      "grad_norm": 1.563196949250548,
      "language_loss": 0.70448095,
      "learning_rate": 1.1514310754783062e-06,
      "loss": 0.72626573,
      "num_input_tokens_seen": 116425830,
      "step": 5408,
      "time_per_iteration": 2.8571908473968506
    },
    {
      "auxiliary_loss_clip": 0.01147057,
      "auxiliary_loss_mlp": 0.01026382,
      "balance_loss_clip": 1.04704273,
      "balance_loss_mlp": 1.01957548,
      "epoch": 0.6503937954668431,
      "flos": 28658546726400.0,
      "grad_norm": 4.01134766643033,
      "language_loss": 0.7306217,
      "learning_rate": 1.1507257584750964e-06,
      "loss": 0.75235611,
      "num_input_tokens_seen": 116446010,
      "step": 5409,
      "time_per_iteration": 2.6980931758880615
    },
    {
      "auxiliary_loss_clip": 0.01172516,
      "auxiliary_loss_mlp": 0.01028999,
      "balance_loss_clip": 1.04956305,
      "balance_loss_mlp": 1.02189422,
      "epoch": 0.6505140383574821,
      "flos": 20922562592640.0,
      "grad_norm": 1.8660198188386592,
      "language_loss": 0.77320361,
      "learning_rate": 1.150020570306113e-06,
      "loss": 0.79521871,
      "num_input_tokens_seen": 116465150,
      "step": 5410,
      "time_per_iteration": 2.6570024490356445
    },
    {
      "auxiliary_loss_clip": 0.01135957,
      "auxiliary_loss_mlp": 0.01026596,
      "balance_loss_clip": 1.04110074,
      "balance_loss_mlp": 1.01956892,
      "epoch": 0.6506342812481212,
      "flos": 20595236929920.0,
      "grad_norm": 2.4755844118668,
      "language_loss": 0.74568045,
      "learning_rate": 1.1493155110783338e-06,
      "loss": 0.76730597,
      "num_input_tokens_seen": 116483675,
      "step": 5411,
      "time_per_iteration": 2.6703391075134277
    },
    {
      "auxiliary_loss_clip": 0.01158435,
      "auxiliary_loss_mlp": 0.01024718,
      "balance_loss_clip": 1.04702044,
      "balance_loss_mlp": 1.01734543,
      "epoch": 0.6507545241387603,
      "flos": 30226478279040.0,
      "grad_norm": 2.4042505292161764,
      "language_loss": 0.70207345,
      "learning_rate": 1.1486105808987155e-06,
      "loss": 0.72390497,
      "num_input_tokens_seen": 116505165,
      "step": 5412,
      "time_per_iteration": 2.636512517929077
    },
    {
      "auxiliary_loss_clip": 0.01160066,
      "auxiliary_loss_mlp": 0.01027762,
      "balance_loss_clip": 1.04919744,
      "balance_loss_mlp": 1.01984644,
      "epoch": 0.6508747670293994,
      "flos": 17128241320320.0,
      "grad_norm": 1.9826676412316855,
      "language_loss": 0.81387025,
      "learning_rate": 1.1479057798741947e-06,
      "loss": 0.83574855,
      "num_input_tokens_seen": 116523220,
      "step": 5413,
      "time_per_iteration": 2.604240894317627
    },
    {
      "auxiliary_loss_clip": 0.01057215,
      "auxiliary_loss_mlp": 0.01001098,
      "balance_loss_clip": 1.02328157,
      "balance_loss_mlp": 0.99967939,
      "epoch": 0.6509950099200384,
      "flos": 68559826573440.0,
      "grad_norm": 0.8052093363843152,
      "language_loss": 0.53339481,
      "learning_rate": 1.14720110811169e-06,
      "loss": 0.55397785,
      "num_input_tokens_seen": 116580450,
      "step": 5414,
      "time_per_iteration": 3.2446889877319336
    },
    {
      "auxiliary_loss_clip": 0.0116139,
      "auxiliary_loss_mlp": 0.01027527,
      "balance_loss_clip": 1.04973483,
      "balance_loss_mlp": 1.01975131,
      "epoch": 0.6511152528106776,
      "flos": 22347462188160.0,
      "grad_norm": 2.204017208741017,
      "language_loss": 0.76576138,
      "learning_rate": 1.146496565718098e-06,
      "loss": 0.78765059,
      "num_input_tokens_seen": 116601020,
      "step": 5415,
      "time_per_iteration": 3.6857950687408447
    },
    {
      "auxiliary_loss_clip": 0.01145992,
      "auxiliary_loss_mlp": 0.01023834,
      "balance_loss_clip": 1.04879451,
      "balance_loss_mlp": 1.01604915,
      "epoch": 0.6512354957013167,
      "flos": 20522158709760.0,
      "grad_norm": 1.8332592368788272,
      "language_loss": 0.75680029,
      "learning_rate": 1.1457921528002996e-06,
      "loss": 0.77849853,
      "num_input_tokens_seen": 116619455,
      "step": 5416,
      "time_per_iteration": 2.700971841812134
    },
    {
      "auxiliary_loss_clip": 0.01171104,
      "auxiliary_loss_mlp": 0.00762073,
      "balance_loss_clip": 1.04754615,
      "balance_loss_mlp": 1.00050199,
      "epoch": 0.6513557385919557,
      "flos": 32337342881280.0,
      "grad_norm": 2.6279831354721908,
      "language_loss": 0.72310966,
      "learning_rate": 1.1450878694651522e-06,
      "loss": 0.74244142,
      "num_input_tokens_seen": 116640020,
      "step": 5417,
      "time_per_iteration": 2.7004945278167725
    },
    {
      "auxiliary_loss_clip": 0.01115631,
      "auxiliary_loss_mlp": 0.01023467,
      "balance_loss_clip": 1.04102659,
      "balance_loss_mlp": 1.01631427,
      "epoch": 0.6514759814825949,
      "flos": 12093206417280.0,
      "grad_norm": 2.328681465638885,
      "language_loss": 0.63656169,
      "learning_rate": 1.1443837158194954e-06,
      "loss": 0.65795267,
      "num_input_tokens_seen": 116655165,
      "step": 5418,
      "time_per_iteration": 3.667316198348999
    },
    {
      "auxiliary_loss_clip": 0.0113202,
      "auxiliary_loss_mlp": 0.01027569,
      "balance_loss_clip": 1.04764938,
      "balance_loss_mlp": 1.02004123,
      "epoch": 0.651596224373234,
      "flos": 22526907557760.0,
      "grad_norm": 1.598869245555437,
      "language_loss": 0.74887311,
      "learning_rate": 1.1436796919701484e-06,
      "loss": 0.77046895,
      "num_input_tokens_seen": 116673880,
      "step": 5419,
      "time_per_iteration": 2.7325007915496826
    },
    {
      "auxiliary_loss_clip": 0.01144146,
      "auxiliary_loss_mlp": 0.01023724,
      "balance_loss_clip": 1.04707742,
      "balance_loss_mlp": 1.01643133,
      "epoch": 0.651716467263873,
      "flos": 27818955250560.0,
      "grad_norm": 2.1714434054568383,
      "language_loss": 0.61942703,
      "learning_rate": 1.1429757980239115e-06,
      "loss": 0.64110571,
      "num_input_tokens_seen": 116694305,
      "step": 5420,
      "time_per_iteration": 2.7146098613739014
    },
    {
      "auxiliary_loss_clip": 0.01175955,
      "auxiliary_loss_mlp": 0.0102877,
      "balance_loss_clip": 1.05051994,
      "balance_loss_mlp": 1.0206821,
      "epoch": 0.6518367101545122,
      "flos": 24316300414080.0,
      "grad_norm": 3.130492621938327,
      "language_loss": 0.82160842,
      "learning_rate": 1.1422720340875636e-06,
      "loss": 0.84365565,
      "num_input_tokens_seen": 116713055,
      "step": 5421,
      "time_per_iteration": 2.6725056171417236
    },
    {
      "auxiliary_loss_clip": 0.0116417,
      "auxiliary_loss_mlp": 0.01031291,
      "balance_loss_clip": 1.04841805,
      "balance_loss_mlp": 1.02323806,
      "epoch": 0.6519569530451512,
      "flos": 20011939971840.0,
      "grad_norm": 2.1480051088584795,
      "language_loss": 0.78879976,
      "learning_rate": 1.1415684002678671e-06,
      "loss": 0.81075442,
      "num_input_tokens_seen": 116731815,
      "step": 5422,
      "time_per_iteration": 3.566349983215332
    },
    {
      "auxiliary_loss_clip": 0.01146094,
      "auxiliary_loss_mlp": 0.01024068,
      "balance_loss_clip": 1.04419553,
      "balance_loss_mlp": 1.01646852,
      "epoch": 0.6520771959357903,
      "flos": 21576064682880.0,
      "grad_norm": 2.973947916144434,
      "language_loss": 0.77853322,
      "learning_rate": 1.1408648966715617e-06,
      "loss": 0.80023479,
      "num_input_tokens_seen": 116749335,
      "step": 5423,
      "time_per_iteration": 2.665269613265991
    },
    {
      "auxiliary_loss_clip": 0.01140427,
      "auxiliary_loss_mlp": 0.01030264,
      "balance_loss_clip": 1.04054153,
      "balance_loss_mlp": 1.0223124,
      "epoch": 0.6521974388264293,
      "flos": 22711021695360.0,
      "grad_norm": 2.0724395507587934,
      "language_loss": 0.72666824,
      "learning_rate": 1.1401615234053683e-06,
      "loss": 0.74837506,
      "num_input_tokens_seen": 116768155,
      "step": 5424,
      "time_per_iteration": 2.69214129447937
    },
    {
      "auxiliary_loss_clip": 0.01144465,
      "auxiliary_loss_mlp": 0.01024124,
      "balance_loss_clip": 1.04468095,
      "balance_loss_mlp": 1.0165422,
      "epoch": 0.6523176817170685,
      "flos": 23002939526400.0,
      "grad_norm": 2.256307187130822,
      "language_loss": 0.75869381,
      "learning_rate": 1.1394582805759885e-06,
      "loss": 0.78037965,
      "num_input_tokens_seen": 116787435,
      "step": 5425,
      "time_per_iteration": 2.7005515098571777
    },
    {
      "auxiliary_loss_clip": 0.01158407,
      "auxiliary_loss_mlp": 0.0102455,
      "balance_loss_clip": 1.04829323,
      "balance_loss_mlp": 1.01689935,
      "epoch": 0.6524379246077076,
      "flos": 21688249835520.0,
      "grad_norm": 14.192135172050383,
      "language_loss": 0.75780183,
      "learning_rate": 1.1387551682901022e-06,
      "loss": 0.77963144,
      "num_input_tokens_seen": 116808040,
      "step": 5426,
      "time_per_iteration": 2.585571765899658
    },
    {
      "auxiliary_loss_clip": 0.01128893,
      "auxiliary_loss_mlp": 0.01024659,
      "balance_loss_clip": 1.04469848,
      "balance_loss_mlp": 1.01715469,
      "epoch": 0.6525581674983466,
      "flos": 19390936711680.0,
      "grad_norm": 1.7720465234051483,
      "language_loss": 0.70572406,
      "learning_rate": 1.138052186654373e-06,
      "loss": 0.72725958,
      "num_input_tokens_seen": 116825510,
      "step": 5427,
      "time_per_iteration": 2.826972246170044
    },
    {
      "auxiliary_loss_clip": 0.01144647,
      "auxiliary_loss_mlp": 0.01029423,
      "balance_loss_clip": 1.04493654,
      "balance_loss_mlp": 1.02131665,
      "epoch": 0.6526784103889858,
      "flos": 17165444832000.0,
      "grad_norm": 2.048287840030225,
      "language_loss": 0.88153243,
      "learning_rate": 1.1373493357754417e-06,
      "loss": 0.90327311,
      "num_input_tokens_seen": 116844415,
      "step": 5428,
      "time_per_iteration": 2.656383991241455
    },
    {
      "auxiliary_loss_clip": 0.01171794,
      "auxiliary_loss_mlp": 0.01023824,
      "balance_loss_clip": 1.04709828,
      "balance_loss_mlp": 1.01684463,
      "epoch": 0.6527986532796248,
      "flos": 18989168112000.0,
      "grad_norm": 3.12947405051866,
      "language_loss": 0.7733174,
      "learning_rate": 1.1366466157599303e-06,
      "loss": 0.7952736,
      "num_input_tokens_seen": 116863690,
      "step": 5429,
      "time_per_iteration": 2.6151983737945557
    },
    {
      "auxiliary_loss_clip": 0.01110877,
      "auxiliary_loss_mlp": 0.00762361,
      "balance_loss_clip": 1.04040897,
      "balance_loss_mlp": 1.00063884,
      "epoch": 0.6529188961702639,
      "flos": 14238581011200.0,
      "grad_norm": 1.9820861923337434,
      "language_loss": 0.76243013,
      "learning_rate": 1.1359440267144412e-06,
      "loss": 0.7811625,
      "num_input_tokens_seen": 116881145,
      "step": 5430,
      "time_per_iteration": 2.710097312927246
    },
    {
      "auxiliary_loss_clip": 0.01159032,
      "auxiliary_loss_mlp": 0.01029199,
      "balance_loss_clip": 1.04650855,
      "balance_loss_mlp": 1.02196002,
      "epoch": 0.653039139060903,
      "flos": 36682929158400.0,
      "grad_norm": 1.9852106600436288,
      "language_loss": 0.74370182,
      "learning_rate": 1.1352415687455556e-06,
      "loss": 0.76558411,
      "num_input_tokens_seen": 116902405,
      "step": 5431,
      "time_per_iteration": 2.7089178562164307
    },
    {
      "auxiliary_loss_clip": 0.01158992,
      "auxiliary_loss_mlp": 0.01030857,
      "balance_loss_clip": 1.04902005,
      "balance_loss_mlp": 1.0231142,
      "epoch": 0.6531593819515421,
      "flos": 25376275785600.0,
      "grad_norm": 1.9541790218140125,
      "language_loss": 0.62939543,
      "learning_rate": 1.1345392419598362e-06,
      "loss": 0.65129387,
      "num_input_tokens_seen": 116921285,
      "step": 5432,
      "time_per_iteration": 2.644500732421875
    },
    {
      "auxiliary_loss_clip": 0.01152708,
      "auxiliary_loss_mlp": 0.01025972,
      "balance_loss_clip": 1.04670835,
      "balance_loss_mlp": 1.01852751,
      "epoch": 0.6532796248421812,
      "flos": 21178533888000.0,
      "grad_norm": 2.4586799907585264,
      "language_loss": 0.71873152,
      "learning_rate": 1.1338370464638263e-06,
      "loss": 0.74051833,
      "num_input_tokens_seen": 116940685,
      "step": 5433,
      "time_per_iteration": 2.6236720085144043
    },
    {
      "auxiliary_loss_clip": 0.01170687,
      "auxiliary_loss_mlp": 0.01024599,
      "balance_loss_clip": 1.04627526,
      "balance_loss_mlp": 1.01722264,
      "epoch": 0.6533998677328203,
      "flos": 17675950878720.0,
      "grad_norm": 2.3490334832223594,
      "language_loss": 0.63760912,
      "learning_rate": 1.1331349823640474e-06,
      "loss": 0.65956199,
      "num_input_tokens_seen": 116958115,
      "step": 5434,
      "time_per_iteration": 2.526662588119507
    },
    {
      "auxiliary_loss_clip": 0.01161081,
      "auxiliary_loss_mlp": 0.00761872,
      "balance_loss_clip": 1.04820263,
      "balance_loss_mlp": 1.00054562,
      "epoch": 0.6535201106234594,
      "flos": 28400384701440.0,
      "grad_norm": 2.0875325874195707,
      "language_loss": 0.77917004,
      "learning_rate": 1.132433049767003e-06,
      "loss": 0.79839957,
      "num_input_tokens_seen": 116976030,
      "step": 5435,
      "time_per_iteration": 2.6820363998413086
    },
    {
      "auxiliary_loss_clip": 0.01139829,
      "auxiliary_loss_mlp": 0.01021041,
      "balance_loss_clip": 1.04427826,
      "balance_loss_mlp": 1.01446319,
      "epoch": 0.6536403535140984,
      "flos": 23586667447680.0,
      "grad_norm": 2.4718569205573404,
      "language_loss": 0.81296974,
      "learning_rate": 1.1317312487791748e-06,
      "loss": 0.83457845,
      "num_input_tokens_seen": 116997680,
      "step": 5436,
      "time_per_iteration": 2.684680700302124
    },
    {
      "auxiliary_loss_clip": 0.01150623,
      "auxiliary_loss_mlp": 0.01024661,
      "balance_loss_clip": 1.04379344,
      "balance_loss_mlp": 1.01702011,
      "epoch": 0.6537605964047376,
      "flos": 21579476474880.0,
      "grad_norm": 2.1886455924998667,
      "language_loss": 0.72962093,
      "learning_rate": 1.1310295795070253e-06,
      "loss": 0.75137377,
      "num_input_tokens_seen": 117017620,
      "step": 5437,
      "time_per_iteration": 2.611844062805176
    },
    {
      "auxiliary_loss_clip": 0.01118673,
      "auxiliary_loss_mlp": 0.01023476,
      "balance_loss_clip": 1.04174495,
      "balance_loss_mlp": 1.01654983,
      "epoch": 0.6538808392953767,
      "flos": 26833997433600.0,
      "grad_norm": 1.8512680484387822,
      "language_loss": 0.80941689,
      "learning_rate": 1.1303280420569982e-06,
      "loss": 0.83083838,
      "num_input_tokens_seen": 117039505,
      "step": 5438,
      "time_per_iteration": 2.762617826461792
    },
    {
      "auxiliary_loss_clip": 0.01149272,
      "auxiliary_loss_mlp": 0.01023871,
      "balance_loss_clip": 1.04278326,
      "balance_loss_mlp": 1.01659274,
      "epoch": 0.6540010821860157,
      "flos": 30738241301760.0,
      "grad_norm": 1.6508634136155322,
      "language_loss": 0.77544975,
      "learning_rate": 1.1296266365355158e-06,
      "loss": 0.79718113,
      "num_input_tokens_seen": 117062890,
      "step": 5439,
      "time_per_iteration": 2.7710511684417725
    },
    {
      "auxiliary_loss_clip": 0.01133261,
      "auxiliary_loss_mlp": 0.01027653,
      "balance_loss_clip": 1.0453527,
      "balance_loss_mlp": 1.0199579,
      "epoch": 0.6541213250766549,
      "flos": 26907147480960.0,
      "grad_norm": 1.9747510129372408,
      "language_loss": 0.73850209,
      "learning_rate": 1.1289253630489806e-06,
      "loss": 0.76011121,
      "num_input_tokens_seen": 117083940,
      "step": 5440,
      "time_per_iteration": 2.748257637023926
    },
    {
      "auxiliary_loss_clip": 0.01164017,
      "auxiliary_loss_mlp": 0.01027785,
      "balance_loss_clip": 1.0468595,
      "balance_loss_mlp": 1.0198338,
      "epoch": 0.6542415679672939,
      "flos": 19172384409600.0,
      "grad_norm": 7.235445362496793,
      "language_loss": 0.72368771,
      "learning_rate": 1.1282242217037753e-06,
      "loss": 0.74560571,
      "num_input_tokens_seen": 117101440,
      "step": 5441,
      "time_per_iteration": 3.5673255920410156
    },
    {
      "auxiliary_loss_clip": 0.01106425,
      "auxiliary_loss_mlp": 0.0103019,
      "balance_loss_clip": 1.03682756,
      "balance_loss_mlp": 1.02189291,
      "epoch": 0.654361810857933,
      "flos": 48173517100800.0,
      "grad_norm": 2.039002463636506,
      "language_loss": 0.61901879,
      "learning_rate": 1.127523212606262e-06,
      "loss": 0.64038491,
      "num_input_tokens_seen": 117124265,
      "step": 5442,
      "time_per_iteration": 3.021934986114502
    },
    {
      "auxiliary_loss_clip": 0.01155248,
      "auxiliary_loss_mlp": 0.01023198,
      "balance_loss_clip": 1.0454073,
      "balance_loss_mlp": 1.01627851,
      "epoch": 0.6544820537485722,
      "flos": 26943165843840.0,
      "grad_norm": 1.562868577499215,
      "language_loss": 0.73021519,
      "learning_rate": 1.1268223358627835e-06,
      "loss": 0.75199974,
      "num_input_tokens_seen": 117146755,
      "step": 5443,
      "time_per_iteration": 2.7337329387664795
    },
    {
      "auxiliary_loss_clip": 0.01173182,
      "auxiliary_loss_mlp": 0.01024453,
      "balance_loss_clip": 1.04843426,
      "balance_loss_mlp": 1.01687121,
      "epoch": 0.6546022966392112,
      "flos": 20886328748160.0,
      "grad_norm": 1.92902295729455,
      "language_loss": 0.71977317,
      "learning_rate": 1.126121591579663e-06,
      "loss": 0.74174953,
      "num_input_tokens_seen": 117165960,
      "step": 5444,
      "time_per_iteration": 3.609884262084961
    },
    {
      "auxiliary_loss_clip": 0.01156524,
      "auxiliary_loss_mlp": 0.01028184,
      "balance_loss_clip": 1.04848206,
      "balance_loss_mlp": 1.02129102,
      "epoch": 0.6547225395298503,
      "flos": 24936693143040.0,
      "grad_norm": 1.8844845001010355,
      "language_loss": 0.69023132,
      "learning_rate": 1.1254209798632018e-06,
      "loss": 0.71207845,
      "num_input_tokens_seen": 117186980,
      "step": 5445,
      "time_per_iteration": 3.633784532546997
    },
    {
      "auxiliary_loss_clip": 0.01091987,
      "auxiliary_loss_mlp": 0.01024385,
      "balance_loss_clip": 1.03946352,
      "balance_loss_mlp": 1.01717305,
      "epoch": 0.6548427824204894,
      "flos": 22565942663040.0,
      "grad_norm": 1.8656957040198157,
      "language_loss": 0.84995329,
      "learning_rate": 1.124720500819683e-06,
      "loss": 0.87111706,
      "num_input_tokens_seen": 117205135,
      "step": 5446,
      "time_per_iteration": 2.8088340759277344
    },
    {
      "auxiliary_loss_clip": 0.01171817,
      "auxiliary_loss_mlp": 0.01028724,
      "balance_loss_clip": 1.04843223,
      "balance_loss_mlp": 1.02078414,
      "epoch": 0.6549630253111285,
      "flos": 18442500048000.0,
      "grad_norm": 1.979403183696688,
      "language_loss": 0.81863779,
      "learning_rate": 1.1240201545553682e-06,
      "loss": 0.84064317,
      "num_input_tokens_seen": 117222935,
      "step": 5447,
      "time_per_iteration": 2.6181750297546387
    },
    {
      "auxiliary_loss_clip": 0.01127831,
      "auxiliary_loss_mlp": 0.01023802,
      "balance_loss_clip": 1.04406905,
      "balance_loss_mlp": 1.01694751,
      "epoch": 0.6550832682017675,
      "flos": 25187313312000.0,
      "grad_norm": 1.8741119284952055,
      "language_loss": 0.73022419,
      "learning_rate": 1.1233199411764987e-06,
      "loss": 0.75174052,
      "num_input_tokens_seen": 117242370,
      "step": 5448,
      "time_per_iteration": 3.6163389682769775
    },
    {
      "auxiliary_loss_clip": 0.01114659,
      "auxiliary_loss_mlp": 0.01024741,
      "balance_loss_clip": 1.03919482,
      "balance_loss_mlp": 1.01771379,
      "epoch": 0.6552035110924067,
      "flos": 22748153379840.0,
      "grad_norm": 1.87866204952456,
      "language_loss": 0.69066441,
      "learning_rate": 1.1226198607892978e-06,
      "loss": 0.71205837,
      "num_input_tokens_seen": 117262930,
      "step": 5449,
      "time_per_iteration": 2.655757188796997
    },
    {
      "auxiliary_loss_clip": 0.01116352,
      "auxiliary_loss_mlp": 0.0102607,
      "balance_loss_clip": 1.04289889,
      "balance_loss_mlp": 1.01896548,
      "epoch": 0.6553237539830458,
      "flos": 21799178012160.0,
      "grad_norm": 1.8271886080493103,
      "language_loss": 0.79738712,
      "learning_rate": 1.1219199134999664e-06,
      "loss": 0.81881136,
      "num_input_tokens_seen": 117281430,
      "step": 5450,
      "time_per_iteration": 2.7710142135620117
    },
    {
      "auxiliary_loss_clip": 0.01146221,
      "auxiliary_loss_mlp": 0.01030534,
      "balance_loss_clip": 1.0451349,
      "balance_loss_mlp": 1.02238011,
      "epoch": 0.6554439968736848,
      "flos": 20887226588160.0,
      "grad_norm": 2.643118109770891,
      "language_loss": 0.78683543,
      "learning_rate": 1.1212200994146863e-06,
      "loss": 0.80860293,
      "num_input_tokens_seen": 117299185,
      "step": 5451,
      "time_per_iteration": 2.628506660461426
    },
    {
      "auxiliary_loss_clip": 0.01123432,
      "auxiliary_loss_mlp": 0.01023022,
      "balance_loss_clip": 1.03740525,
      "balance_loss_mlp": 1.01577997,
      "epoch": 0.655564239764324,
      "flos": 16139045698560.0,
      "grad_norm": 1.8898299904941038,
      "language_loss": 0.75610709,
      "learning_rate": 1.120520418639618e-06,
      "loss": 0.77757162,
      "num_input_tokens_seen": 117317720,
      "step": 5452,
      "time_per_iteration": 2.7033801078796387
    },
    {
      "auxiliary_loss_clip": 0.01159936,
      "auxiliary_loss_mlp": 0.01028286,
      "balance_loss_clip": 1.04898763,
      "balance_loss_mlp": 1.02123833,
      "epoch": 0.655684482654963,
      "flos": 29570354496000.0,
      "grad_norm": 2.0669862359186904,
      "language_loss": 0.83199477,
      "learning_rate": 1.119820871280903e-06,
      "loss": 0.85387707,
      "num_input_tokens_seen": 117338795,
      "step": 5453,
      "time_per_iteration": 2.7102773189544678
    },
    {
      "auxiliary_loss_clip": 0.01156932,
      "auxiliary_loss_mlp": 0.01028254,
      "balance_loss_clip": 1.04686296,
      "balance_loss_mlp": 1.02087843,
      "epoch": 0.6558047255456021,
      "flos": 29789409588480.0,
      "grad_norm": 1.9247218060188158,
      "language_loss": 0.73231924,
      "learning_rate": 1.1191214574446614e-06,
      "loss": 0.75417113,
      "num_input_tokens_seen": 117359040,
      "step": 5454,
      "time_per_iteration": 2.6666131019592285
    },
    {
      "auxiliary_loss_clip": 0.01137471,
      "auxiliary_loss_mlp": 0.01028082,
      "balance_loss_clip": 1.04276526,
      "balance_loss_mlp": 1.02043438,
      "epoch": 0.6559249684362413,
      "flos": 29059166090880.0,
      "grad_norm": 1.4907246106809866,
      "language_loss": 0.80146497,
      "learning_rate": 1.118422177236995e-06,
      "loss": 0.82312047,
      "num_input_tokens_seen": 117380865,
      "step": 5455,
      "time_per_iteration": 2.7533791065216064
    },
    {
      "auxiliary_loss_clip": 0.01144411,
      "auxiliary_loss_mlp": 0.01029918,
      "balance_loss_clip": 1.04467463,
      "balance_loss_mlp": 1.02224672,
      "epoch": 0.6560452113268803,
      "flos": 20225464369920.0,
      "grad_norm": 2.2121812929439137,
      "language_loss": 0.85582626,
      "learning_rate": 1.1177230307639835e-06,
      "loss": 0.87756956,
      "num_input_tokens_seen": 117398405,
      "step": 5456,
      "time_per_iteration": 2.679054021835327
    },
    {
      "auxiliary_loss_clip": 0.0112437,
      "auxiliary_loss_mlp": 0.01029358,
      "balance_loss_clip": 1.04262471,
      "balance_loss_mlp": 1.02138281,
      "epoch": 0.6561654542175194,
      "flos": 25045538330880.0,
      "grad_norm": 1.6528735227835665,
      "language_loss": 0.79051501,
      "learning_rate": 1.1170240181316865e-06,
      "loss": 0.81205225,
      "num_input_tokens_seen": 117419850,
      "step": 5457,
      "time_per_iteration": 2.8566107749938965
    },
    {
      "auxiliary_loss_clip": 0.01123401,
      "auxiliary_loss_mlp": 0.01026631,
      "balance_loss_clip": 1.04012227,
      "balance_loss_mlp": 1.01878715,
      "epoch": 0.6562856971081584,
      "flos": 22856711258880.0,
      "grad_norm": 2.1949371381820093,
      "language_loss": 0.79118311,
      "learning_rate": 1.1163251394461442e-06,
      "loss": 0.81268346,
      "num_input_tokens_seen": 117438330,
      "step": 5458,
      "time_per_iteration": 2.6920738220214844
    },
    {
      "auxiliary_loss_clip": 0.01156233,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.04736137,
      "balance_loss_mlp": 1.01796389,
      "epoch": 0.6564059399987976,
      "flos": 18872565586560.0,
      "grad_norm": 1.9913528696587777,
      "language_loss": 0.82567966,
      "learning_rate": 1.1156263948133746e-06,
      "loss": 0.84749907,
      "num_input_tokens_seen": 117454985,
      "step": 5459,
      "time_per_iteration": 2.632329225540161
    },
    {
      "auxiliary_loss_clip": 0.01106778,
      "auxiliary_loss_mlp": 0.00762426,
      "balance_loss_clip": 1.0408628,
      "balance_loss_mlp": 1.00059128,
      "epoch": 0.6565261828894366,
      "flos": 25484187219840.0,
      "grad_norm": 1.78046461095855,
      "language_loss": 0.77948117,
      "learning_rate": 1.1149277843393787e-06,
      "loss": 0.79817313,
      "num_input_tokens_seen": 117476145,
      "step": 5460,
      "time_per_iteration": 2.750566244125366
    },
    {
      "auxiliary_loss_clip": 0.01096152,
      "auxiliary_loss_mlp": 0.00763103,
      "balance_loss_clip": 1.03509784,
      "balance_loss_mlp": 1.0006268,
      "epoch": 0.6566464257800757,
      "flos": 19683500987520.0,
      "grad_norm": 2.592483128713835,
      "language_loss": 0.63285899,
      "learning_rate": 1.1142293081301342e-06,
      "loss": 0.65145159,
      "num_input_tokens_seen": 117494025,
      "step": 5461,
      "time_per_iteration": 2.8487167358398438
    },
    {
      "auxiliary_loss_clip": 0.01140657,
      "auxiliary_loss_mlp": 0.0102043,
      "balance_loss_clip": 1.04515111,
      "balance_loss_mlp": 1.01409757,
      "epoch": 0.6567666686707149,
      "flos": 23514127931520.0,
      "grad_norm": 1.5406863366007868,
      "language_loss": 0.67606211,
      "learning_rate": 1.1135309662915995e-06,
      "loss": 0.69767296,
      "num_input_tokens_seen": 117514190,
      "step": 5462,
      "time_per_iteration": 2.682324171066284
    },
    {
      "auxiliary_loss_clip": 0.01118302,
      "auxiliary_loss_mlp": 0.01020956,
      "balance_loss_clip": 1.03891563,
      "balance_loss_mlp": 1.0138036,
      "epoch": 0.6568869115613539,
      "flos": 32781342896640.0,
      "grad_norm": 2.137361897536097,
      "language_loss": 0.60419106,
      "learning_rate": 1.112832758929712e-06,
      "loss": 0.62558365,
      "num_input_tokens_seen": 117536800,
      "step": 5463,
      "time_per_iteration": 2.9119882583618164
    },
    {
      "auxiliary_loss_clip": 0.01155892,
      "auxiliary_loss_mlp": 0.01026229,
      "balance_loss_clip": 1.04612386,
      "balance_loss_mlp": 1.0185281,
      "epoch": 0.657007154451993,
      "flos": 18442428220800.0,
      "grad_norm": 1.7851865878587936,
      "language_loss": 0.74959987,
      "learning_rate": 1.11213468615039e-06,
      "loss": 0.77142107,
      "num_input_tokens_seen": 117556230,
      "step": 5464,
      "time_per_iteration": 2.647881031036377
    },
    {
      "auxiliary_loss_clip": 0.01098081,
      "auxiliary_loss_mlp": 0.01026833,
      "balance_loss_clip": 1.03952932,
      "balance_loss_mlp": 1.01974642,
      "epoch": 0.6571273973426321,
      "flos": 25156717902720.0,
      "grad_norm": 1.6952970761632329,
      "language_loss": 0.75329107,
      "learning_rate": 1.1114367480595292e-06,
      "loss": 0.77454019,
      "num_input_tokens_seen": 117577310,
      "step": 5465,
      "time_per_iteration": 2.7637722492218018
    },
    {
      "auxiliary_loss_clip": 0.01101648,
      "auxiliary_loss_mlp": 0.01029454,
      "balance_loss_clip": 1.04095185,
      "balance_loss_mlp": 1.02160454,
      "epoch": 0.6572476402332712,
      "flos": 17529830352000.0,
      "grad_norm": 2.2240586006309724,
      "language_loss": 0.81227696,
      "learning_rate": 1.1107389447630086e-06,
      "loss": 0.833588,
      "num_input_tokens_seen": 117596010,
      "step": 5466,
      "time_per_iteration": 2.7606360912323
    },
    {
      "auxiliary_loss_clip": 0.01138469,
      "auxiliary_loss_mlp": 0.00762392,
      "balance_loss_clip": 1.04174888,
      "balance_loss_mlp": 1.00056565,
      "epoch": 0.6573678831239103,
      "flos": 17014260487680.0,
      "grad_norm": 2.1024014960869915,
      "language_loss": 0.78359342,
      "learning_rate": 1.1100412763666818e-06,
      "loss": 0.80260205,
      "num_input_tokens_seen": 117611270,
      "step": 5467,
      "time_per_iteration": 3.6081461906433105
    },
    {
      "auxiliary_loss_clip": 0.011447,
      "auxiliary_loss_mlp": 0.01029111,
      "balance_loss_clip": 1.04576194,
      "balance_loss_mlp": 1.02168417,
      "epoch": 0.6574881260145494,
      "flos": 23910078528000.0,
      "grad_norm": 1.500033296809999,
      "language_loss": 0.80108434,
      "learning_rate": 1.1093437429763865e-06,
      "loss": 0.82282245,
      "num_input_tokens_seen": 117631535,
      "step": 5468,
      "time_per_iteration": 2.907742500305176
    },
    {
      "auxiliary_loss_clip": 0.0115705,
      "auxiliary_loss_mlp": 0.0102122,
      "balance_loss_clip": 1.04665971,
      "balance_loss_mlp": 1.013623,
      "epoch": 0.6576083689051885,
      "flos": 11218458504960.0,
      "grad_norm": 1.985508983726818,
      "language_loss": 0.73289478,
      "learning_rate": 1.1086463446979361e-06,
      "loss": 0.75467747,
      "num_input_tokens_seen": 117649885,
      "step": 5469,
      "time_per_iteration": 3.612353801727295
    },
    {
      "auxiliary_loss_clip": 0.01161031,
      "auxiliary_loss_mlp": 0.01025187,
      "balance_loss_clip": 1.04951799,
      "balance_loss_mlp": 1.01738501,
      "epoch": 0.6577286117958275,
      "flos": 22455553190400.0,
      "grad_norm": 1.8283972849280057,
      "language_loss": 0.77487117,
      "learning_rate": 1.1079490816371277e-06,
      "loss": 0.79673332,
      "num_input_tokens_seen": 117669650,
      "step": 5470,
      "time_per_iteration": 3.594592571258545
    },
    {
      "auxiliary_loss_clip": 0.01158167,
      "auxiliary_loss_mlp": 0.00762435,
      "balance_loss_clip": 1.04612708,
      "balance_loss_mlp": 1.0005641,
      "epoch": 0.6578488546864667,
      "flos": 21872184405120.0,
      "grad_norm": 1.9841601941796894,
      "language_loss": 0.7479279,
      "learning_rate": 1.1072519538997352e-06,
      "loss": 0.76713395,
      "num_input_tokens_seen": 117688790,
      "step": 5471,
      "time_per_iteration": 2.711440324783325
    },
    {
      "auxiliary_loss_clip": 0.01145636,
      "auxiliary_loss_mlp": 0.01021532,
      "balance_loss_clip": 1.04342043,
      "balance_loss_mlp": 1.01401544,
      "epoch": 0.6579690975771058,
      "flos": 23543753673600.0,
      "grad_norm": 2.8462165881849573,
      "language_loss": 0.82188445,
      "learning_rate": 1.1065549615915095e-06,
      "loss": 0.84355617,
      "num_input_tokens_seen": 117708620,
      "step": 5472,
      "time_per_iteration": 2.760585069656372
    },
    {
      "auxiliary_loss_clip": 0.01160081,
      "auxiliary_loss_mlp": 0.01028413,
      "balance_loss_clip": 1.0508858,
      "balance_loss_mlp": 1.02101862,
      "epoch": 0.6580893404677448,
      "flos": 32743995730560.0,
      "grad_norm": 2.6766810270575276,
      "language_loss": 0.78588343,
      "learning_rate": 1.105858104818187e-06,
      "loss": 0.80776846,
      "num_input_tokens_seen": 117729775,
      "step": 5473,
      "time_per_iteration": 2.7282819747924805
    },
    {
      "auxiliary_loss_clip": 0.01161712,
      "auxiliary_loss_mlp": 0.01030131,
      "balance_loss_clip": 1.04681206,
      "balance_loss_mlp": 1.02187562,
      "epoch": 0.658209583358384,
      "flos": 15888138220800.0,
      "grad_norm": 2.5388165388070534,
      "language_loss": 0.74683529,
      "learning_rate": 1.105161383685478e-06,
      "loss": 0.76875371,
      "num_input_tokens_seen": 117746160,
      "step": 5474,
      "time_per_iteration": 3.528578758239746
    },
    {
      "auxiliary_loss_clip": 0.01040328,
      "auxiliary_loss_mlp": 0.01004542,
      "balance_loss_clip": 1.01474738,
      "balance_loss_mlp": 1.00326681,
      "epoch": 0.658329826249023,
      "flos": 62695902447360.0,
      "grad_norm": 0.7225109128915901,
      "language_loss": 0.56300116,
      "learning_rate": 1.1044647982990771e-06,
      "loss": 0.58344984,
      "num_input_tokens_seen": 117808045,
      "step": 5475,
      "time_per_iteration": 3.2037813663482666
    },
    {
      "auxiliary_loss_clip": 0.01144841,
      "auxiliary_loss_mlp": 0.01026915,
      "balance_loss_clip": 1.04602408,
      "balance_loss_mlp": 1.01889825,
      "epoch": 0.6584500691396621,
      "flos": 31722624501120.0,
      "grad_norm": 2.4479948901781903,
      "language_loss": 0.64614105,
      "learning_rate": 1.1037683487646536e-06,
      "loss": 0.6678586,
      "num_input_tokens_seen": 117828330,
      "step": 5476,
      "time_per_iteration": 2.7692580223083496
    },
    {
      "auxiliary_loss_clip": 0.01141904,
      "auxiliary_loss_mlp": 0.00762235,
      "balance_loss_clip": 1.04712749,
      "balance_loss_mlp": 1.00052738,
      "epoch": 0.6585703120303013,
      "flos": 18406086635520.0,
      "grad_norm": 2.126267399094209,
      "language_loss": 0.77087557,
      "learning_rate": 1.1030720351878583e-06,
      "loss": 0.78991699,
      "num_input_tokens_seen": 117846450,
      "step": 5477,
      "time_per_iteration": 2.664565324783325
    },
    {
      "auxiliary_loss_clip": 0.01056041,
      "auxiliary_loss_mlp": 0.01004171,
      "balance_loss_clip": 1.01599836,
      "balance_loss_mlp": 1.00284219,
      "epoch": 0.6586905549209403,
      "flos": 58309880434560.0,
      "grad_norm": 0.811914981546454,
      "language_loss": 0.57625175,
      "learning_rate": 1.102375857674323e-06,
      "loss": 0.59685385,
      "num_input_tokens_seen": 117908365,
      "step": 5478,
      "time_per_iteration": 3.239750862121582
    },
    {
      "auxiliary_loss_clip": 0.0114321,
      "auxiliary_loss_mlp": 0.01023213,
      "balance_loss_clip": 1.04405212,
      "balance_loss_mlp": 1.01624179,
      "epoch": 0.6588107978115794,
      "flos": 22782627457920.0,
      "grad_norm": 1.781733151578761,
      "language_loss": 0.90316474,
      "learning_rate": 1.1016798163296561e-06,
      "loss": 0.92482889,
      "num_input_tokens_seen": 117927565,
      "step": 5479,
      "time_per_iteration": 2.743032455444336
    },
    {
      "auxiliary_loss_clip": 0.01159236,
      "auxiliary_loss_mlp": 0.01024034,
      "balance_loss_clip": 1.04555297,
      "balance_loss_mlp": 1.01663709,
      "epoch": 0.6589310407022185,
      "flos": 20667525050880.0,
      "grad_norm": 1.9220757351270759,
      "language_loss": 0.66131699,
      "learning_rate": 1.1009839112594471e-06,
      "loss": 0.68314964,
      "num_input_tokens_seen": 117945590,
      "step": 5480,
      "time_per_iteration": 2.5985333919525146
    },
    {
      "auxiliary_loss_clip": 0.01161761,
      "auxiliary_loss_mlp": 0.01021811,
      "balance_loss_clip": 1.04692984,
      "balance_loss_mlp": 1.01464379,
      "epoch": 0.6590512835928576,
      "flos": 25630595055360.0,
      "grad_norm": 2.07095634426907,
      "language_loss": 0.71949542,
      "learning_rate": 1.1002881425692638e-06,
      "loss": 0.74133122,
      "num_input_tokens_seen": 117966020,
      "step": 5481,
      "time_per_iteration": 2.7071070671081543
    },
    {
      "auxiliary_loss_clip": 0.01152304,
      "auxiliary_loss_mlp": 0.01026735,
      "balance_loss_clip": 1.04462111,
      "balance_loss_mlp": 1.01943922,
      "epoch": 0.6591715264834966,
      "flos": 23726108044800.0,
      "grad_norm": 1.848814859607197,
      "language_loss": 0.75193357,
      "learning_rate": 1.0995925103646532e-06,
      "loss": 0.77372402,
      "num_input_tokens_seen": 117984620,
      "step": 5482,
      "time_per_iteration": 2.6249442100524902
    },
    {
      "auxiliary_loss_clip": 0.01125295,
      "auxiliary_loss_mlp": 0.01023902,
      "balance_loss_clip": 1.0446291,
      "balance_loss_mlp": 1.01704466,
      "epoch": 0.6592917693741358,
      "flos": 35773850822400.0,
      "grad_norm": 1.6201716615860045,
      "language_loss": 0.67212021,
      "learning_rate": 1.0988970147511437e-06,
      "loss": 0.69361216,
      "num_input_tokens_seen": 118006500,
      "step": 5483,
      "time_per_iteration": 2.976287603378296
    },
    {
      "auxiliary_loss_clip": 0.01144515,
      "auxiliary_loss_mlp": 0.01023693,
      "balance_loss_clip": 1.04769921,
      "balance_loss_mlp": 1.01637316,
      "epoch": 0.6594120122647749,
      "flos": 21396834794880.0,
      "grad_norm": 2.121134265993353,
      "language_loss": 0.80464113,
      "learning_rate": 1.0982016558342405e-06,
      "loss": 0.82632327,
      "num_input_tokens_seen": 118025470,
      "step": 5484,
      "time_per_iteration": 2.6980535984039307
    },
    {
      "auxiliary_loss_clip": 0.01174448,
      "auxiliary_loss_mlp": 0.01022196,
      "balance_loss_clip": 1.05054033,
      "balance_loss_mlp": 1.01434946,
      "epoch": 0.6595322551554139,
      "flos": 19351829779200.0,
      "grad_norm": 2.2219558952327687,
      "language_loss": 0.71317214,
      "learning_rate": 1.0975064337194291e-06,
      "loss": 0.73513854,
      "num_input_tokens_seen": 118043515,
      "step": 5485,
      "time_per_iteration": 2.656280994415283
    },
    {
      "auxiliary_loss_clip": 0.01122892,
      "auxiliary_loss_mlp": 0.01024897,
      "balance_loss_clip": 1.04209244,
      "balance_loss_mlp": 1.01781559,
      "epoch": 0.6596524980460531,
      "flos": 16837113588480.0,
      "grad_norm": 1.576516265155251,
      "language_loss": 0.70254576,
      "learning_rate": 1.0968113485121743e-06,
      "loss": 0.7240237,
      "num_input_tokens_seen": 118063105,
      "step": 5486,
      "time_per_iteration": 2.7931225299835205
    },
    {
      "auxiliary_loss_clip": 0.01160024,
      "auxiliary_loss_mlp": 0.00763121,
      "balance_loss_clip": 1.04546154,
      "balance_loss_mlp": 1.00047016,
      "epoch": 0.6597727409366921,
      "flos": 21798567480960.0,
      "grad_norm": 2.432777151933775,
      "language_loss": 0.79819953,
      "learning_rate": 1.0961164003179185e-06,
      "loss": 0.81743097,
      "num_input_tokens_seen": 118081615,
      "step": 5487,
      "time_per_iteration": 2.6206037998199463
    },
    {
      "auxiliary_loss_clip": 0.01129307,
      "auxiliary_loss_mlp": 0.01028684,
      "balance_loss_clip": 1.04364777,
      "balance_loss_mlp": 1.02076876,
      "epoch": 0.6598929838273312,
      "flos": 23730704985600.0,
      "grad_norm": 1.8526709003714423,
      "language_loss": 0.84030318,
      "learning_rate": 1.0954215892420884e-06,
      "loss": 0.86188304,
      "num_input_tokens_seen": 118102315,
      "step": 5488,
      "time_per_iteration": 2.714679718017578
    },
    {
      "auxiliary_loss_clip": 0.01136802,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.04749644,
      "balance_loss_mlp": 1.01908588,
      "epoch": 0.6600132267179702,
      "flos": 19974520978560.0,
      "grad_norm": 1.591750087384913,
      "language_loss": 0.70661318,
      "learning_rate": 1.094726915390082e-06,
      "loss": 0.72825027,
      "num_input_tokens_seen": 118120650,
      "step": 5489,
      "time_per_iteration": 2.6814777851104736
    },
    {
      "auxiliary_loss_clip": 0.01161526,
      "auxiliary_loss_mlp": 0.01029809,
      "balance_loss_clip": 1.04956603,
      "balance_loss_mlp": 1.02231634,
      "epoch": 0.6601334696086094,
      "flos": 22342649765760.0,
      "grad_norm": 2.188090733610378,
      "language_loss": 0.6966483,
      "learning_rate": 1.0940323788672836e-06,
      "loss": 0.71856165,
      "num_input_tokens_seen": 118139825,
      "step": 5490,
      "time_per_iteration": 2.6518373489379883
    },
    {
      "auxiliary_loss_clip": 0.01155279,
      "auxiliary_loss_mlp": 0.01026078,
      "balance_loss_clip": 1.04761505,
      "balance_loss_mlp": 1.01876736,
      "epoch": 0.6602537124992485,
      "flos": 25703098657920.0,
      "grad_norm": 1.7186666304886211,
      "language_loss": 0.73776901,
      "learning_rate": 1.0933379797790522e-06,
      "loss": 0.75958258,
      "num_input_tokens_seen": 118159240,
      "step": 5491,
      "time_per_iteration": 2.6417956352233887
    },
    {
      "auxiliary_loss_clip": 0.01174063,
      "auxiliary_loss_mlp": 0.01033365,
      "balance_loss_clip": 1.05069685,
      "balance_loss_mlp": 1.02617669,
      "epoch": 0.6603739553898875,
      "flos": 25848572739840.0,
      "grad_norm": 2.4435977981069645,
      "language_loss": 0.71681023,
      "learning_rate": 1.0926437182307293e-06,
      "loss": 0.73888451,
      "num_input_tokens_seen": 118178050,
      "step": 5492,
      "time_per_iteration": 2.71757173538208
    },
    {
      "auxiliary_loss_clip": 0.01148584,
      "auxiliary_loss_mlp": 0.0102771,
      "balance_loss_clip": 1.0444417,
      "balance_loss_mlp": 1.02022111,
      "epoch": 0.6604941982805267,
      "flos": 24570296461440.0,
      "grad_norm": 2.0339032195782836,
      "language_loss": 0.78528816,
      "learning_rate": 1.0919495943276338e-06,
      "loss": 0.80705106,
      "num_input_tokens_seen": 118199070,
      "step": 5493,
      "time_per_iteration": 3.571887969970703
    },
    {
      "auxiliary_loss_clip": 0.01131805,
      "auxiliary_loss_mlp": 0.01025348,
      "balance_loss_clip": 1.04094172,
      "balance_loss_mlp": 1.01678252,
      "epoch": 0.6606144411711657,
      "flos": 13261775581440.0,
      "grad_norm": 2.7133279835832464,
      "language_loss": 0.76680374,
      "learning_rate": 1.0912556081750611e-06,
      "loss": 0.78837526,
      "num_input_tokens_seen": 118217000,
      "step": 5494,
      "time_per_iteration": 2.747368574142456
    },
    {
      "auxiliary_loss_clip": 0.01139339,
      "auxiliary_loss_mlp": 0.01026373,
      "balance_loss_clip": 1.04573631,
      "balance_loss_mlp": 1.01910686,
      "epoch": 0.6607346840618048,
      "flos": 25155281358720.0,
      "grad_norm": 2.065491172661617,
      "language_loss": 0.76419115,
      "learning_rate": 1.0905617598782909e-06,
      "loss": 0.78584832,
      "num_input_tokens_seen": 118237205,
      "step": 5495,
      "time_per_iteration": 3.659820318222046
    },
    {
      "auxiliary_loss_clip": 0.01109052,
      "auxiliary_loss_mlp": 0.01029537,
      "balance_loss_clip": 1.04073954,
      "balance_loss_mlp": 1.02217257,
      "epoch": 0.660854926952444,
      "flos": 17638029095040.0,
      "grad_norm": 1.9292010475395294,
      "language_loss": 0.81465316,
      "learning_rate": 1.0898680495425775e-06,
      "loss": 0.83603907,
      "num_input_tokens_seen": 118255495,
      "step": 5496,
      "time_per_iteration": 3.6012468338012695
    },
    {
      "auxiliary_loss_clip": 0.01148603,
      "auxiliary_loss_mlp": 0.01028611,
      "balance_loss_clip": 1.04728031,
      "balance_loss_mlp": 1.0205406,
      "epoch": 0.660975169843083,
      "flos": 16836000266880.0,
      "grad_norm": 1.6510132486357605,
      "language_loss": 0.80179226,
      "learning_rate": 1.0891744772731594e-06,
      "loss": 0.82356435,
      "num_input_tokens_seen": 118273310,
      "step": 5497,
      "time_per_iteration": 2.724155902862549
    },
    {
      "auxiliary_loss_clip": 0.01160206,
      "auxiliary_loss_mlp": 0.01025262,
      "balance_loss_clip": 1.04783177,
      "balance_loss_mlp": 1.01803732,
      "epoch": 0.6610954127337221,
      "flos": 26870410846080.0,
      "grad_norm": 1.598582658853339,
      "language_loss": 0.66152996,
      "learning_rate": 1.088481043175248e-06,
      "loss": 0.68338466,
      "num_input_tokens_seen": 118293880,
      "step": 5498,
      "time_per_iteration": 2.690173625946045
    },
    {
      "auxiliary_loss_clip": 0.01134624,
      "auxiliary_loss_mlp": 0.01032053,
      "balance_loss_clip": 1.04189885,
      "balance_loss_mlp": 1.02458167,
      "epoch": 0.6612156556243612,
      "flos": 26465697331200.0,
      "grad_norm": 1.7988312282958177,
      "language_loss": 0.75674796,
      "learning_rate": 1.0877877473540368e-06,
      "loss": 0.77841473,
      "num_input_tokens_seen": 118314465,
      "step": 5499,
      "time_per_iteration": 3.664267063140869
    },
    {
      "auxiliary_loss_clip": 0.01174087,
      "auxiliary_loss_mlp": 0.01028747,
      "balance_loss_clip": 1.04904902,
      "balance_loss_mlp": 1.02175224,
      "epoch": 0.6613358985150003,
      "flos": 19791915212160.0,
      "grad_norm": 2.102514570742598,
      "language_loss": 0.7224583,
      "learning_rate": 1.0870945899147002e-06,
      "loss": 0.74448669,
      "num_input_tokens_seen": 118331110,
      "step": 5500,
      "time_per_iteration": 2.6241533756256104
    },
    {
      "auxiliary_loss_clip": 0.01154793,
      "auxiliary_loss_mlp": 0.01027506,
      "balance_loss_clip": 1.04684925,
      "balance_loss_mlp": 1.02025795,
      "epoch": 0.6614561414056394,
      "flos": 26831627136000.0,
      "grad_norm": 1.8155396236889962,
      "language_loss": 0.76010931,
      "learning_rate": 1.0864015709623879e-06,
      "loss": 0.78193223,
      "num_input_tokens_seen": 118351980,
      "step": 5501,
      "time_per_iteration": 2.6770269870758057
    },
    {
      "auxiliary_loss_clip": 0.01161782,
      "auxiliary_loss_mlp": 0.01031645,
      "balance_loss_clip": 1.0465275,
      "balance_loss_mlp": 1.02427208,
      "epoch": 0.6615763842962785,
      "flos": 22894597128960.0,
      "grad_norm": 2.552213806209215,
      "language_loss": 0.80229872,
      "learning_rate": 1.0857086906022313e-06,
      "loss": 0.82423306,
      "num_input_tokens_seen": 118370315,
      "step": 5502,
      "time_per_iteration": 2.662813663482666
    },
    {
      "auxiliary_loss_clip": 0.0109075,
      "auxiliary_loss_mlp": 0.01027485,
      "balance_loss_clip": 1.04247904,
      "balance_loss_mlp": 1.02021933,
      "epoch": 0.6616966271869176,
      "flos": 24790321221120.0,
      "grad_norm": 2.0650104342377698,
      "language_loss": 0.73240244,
      "learning_rate": 1.0850159489393388e-06,
      "loss": 0.75358474,
      "num_input_tokens_seen": 118389575,
      "step": 5503,
      "time_per_iteration": 2.834150791168213
    },
    {
      "auxiliary_loss_clip": 0.01121909,
      "auxiliary_loss_mlp": 0.01033111,
      "balance_loss_clip": 1.03857088,
      "balance_loss_mlp": 1.02554107,
      "epoch": 0.6618168700775566,
      "flos": 17202109639680.0,
      "grad_norm": 1.8106058426215674,
      "language_loss": 0.82583642,
      "learning_rate": 1.0843233460787992e-06,
      "loss": 0.8473866,
      "num_input_tokens_seen": 118406790,
      "step": 5504,
      "time_per_iteration": 2.795630693435669
    },
    {
      "auxiliary_loss_clip": 0.01119394,
      "auxiliary_loss_mlp": 0.01025791,
      "balance_loss_clip": 1.04410267,
      "balance_loss_mlp": 1.01866293,
      "epoch": 0.6619371129681958,
      "flos": 25447091448960.0,
      "grad_norm": 1.8394838767047852,
      "language_loss": 0.78213453,
      "learning_rate": 1.0836308821256805e-06,
      "loss": 0.80358636,
      "num_input_tokens_seen": 118427590,
      "step": 5505,
      "time_per_iteration": 2.77901029586792
    },
    {
      "auxiliary_loss_clip": 0.01154781,
      "auxiliary_loss_mlp": 0.01023426,
      "balance_loss_clip": 1.04603505,
      "balance_loss_mlp": 1.01646411,
      "epoch": 0.6620573558588349,
      "flos": 18040444139520.0,
      "grad_norm": 2.077687820804473,
      "language_loss": 0.77653837,
      "learning_rate": 1.0829385571850282e-06,
      "loss": 0.79832047,
      "num_input_tokens_seen": 118444570,
      "step": 5506,
      "time_per_iteration": 2.66082501411438
    },
    {
      "auxiliary_loss_clip": 0.01178372,
      "auxiliary_loss_mlp": 0.01025744,
      "balance_loss_clip": 1.04963076,
      "balance_loss_mlp": 1.01711333,
      "epoch": 0.6621775987494739,
      "flos": 17785586165760.0,
      "grad_norm": 2.3946796795695886,
      "language_loss": 0.83441937,
      "learning_rate": 1.0822463713618679e-06,
      "loss": 0.85646051,
      "num_input_tokens_seen": 118461425,
      "step": 5507,
      "time_per_iteration": 2.6162307262420654
    },
    {
      "auxiliary_loss_clip": 0.01132775,
      "auxiliary_loss_mlp": 0.01027862,
      "balance_loss_clip": 1.04476309,
      "balance_loss_mlp": 1.0202328,
      "epoch": 0.6622978416401131,
      "flos": 17492590926720.0,
      "grad_norm": 2.0632533790232346,
      "language_loss": 0.84682941,
      "learning_rate": 1.0815543247612034e-06,
      "loss": 0.86843574,
      "num_input_tokens_seen": 118478495,
      "step": 5508,
      "time_per_iteration": 2.679636001586914
    },
    {
      "auxiliary_loss_clip": 0.01141056,
      "auxiliary_loss_mlp": 0.01025997,
      "balance_loss_clip": 1.04006851,
      "balance_loss_mlp": 1.01830816,
      "epoch": 0.6624180845307521,
      "flos": 21648352803840.0,
      "grad_norm": 2.2324963777055684,
      "language_loss": 0.83122355,
      "learning_rate": 1.0808624174880168e-06,
      "loss": 0.85289407,
      "num_input_tokens_seen": 118499145,
      "step": 5509,
      "time_per_iteration": 2.708400011062622
    },
    {
      "auxiliary_loss_clip": 0.01168854,
      "auxiliary_loss_mlp": 0.01027746,
      "balance_loss_clip": 1.0480504,
      "balance_loss_mlp": 1.02022934,
      "epoch": 0.6625383274213912,
      "flos": 23805902108160.0,
      "grad_norm": 1.6469696707950738,
      "language_loss": 0.80131209,
      "learning_rate": 1.080170649647272e-06,
      "loss": 0.82327807,
      "num_input_tokens_seen": 118518950,
      "step": 5510,
      "time_per_iteration": 2.6337897777557373
    },
    {
      "auxiliary_loss_clip": 0.01171859,
      "auxiliary_loss_mlp": 0.01023753,
      "balance_loss_clip": 1.0484221,
      "balance_loss_mlp": 1.01577163,
      "epoch": 0.6626585703120303,
      "flos": 33262941473280.0,
      "grad_norm": 1.6770824545589964,
      "language_loss": 0.67115468,
      "learning_rate": 1.0794790213439068e-06,
      "loss": 0.69311076,
      "num_input_tokens_seen": 118545850,
      "step": 5511,
      "time_per_iteration": 2.7604665756225586
    },
    {
      "auxiliary_loss_clip": 0.01114377,
      "auxiliary_loss_mlp": 0.01027744,
      "balance_loss_clip": 1.04134619,
      "balance_loss_mlp": 1.02014494,
      "epoch": 0.6627788132026694,
      "flos": 22085780630400.0,
      "grad_norm": 2.034582447812792,
      "language_loss": 0.78409696,
      "learning_rate": 1.078787532682843e-06,
      "loss": 0.80551815,
      "num_input_tokens_seen": 118563325,
      "step": 5512,
      "time_per_iteration": 2.817513942718506
    },
    {
      "auxiliary_loss_clip": 0.01157195,
      "auxiliary_loss_mlp": 0.0102428,
      "balance_loss_clip": 1.04712284,
      "balance_loss_mlp": 1.01704705,
      "epoch": 0.6628990560933085,
      "flos": 36173608260480.0,
      "grad_norm": 2.299569890728429,
      "language_loss": 0.7573086,
      "learning_rate": 1.0780961837689773e-06,
      "loss": 0.77912343,
      "num_input_tokens_seen": 118582835,
      "step": 5513,
      "time_per_iteration": 2.8635032176971436
    },
    {
      "auxiliary_loss_clip": 0.01136543,
      "auxiliary_loss_mlp": 0.01029725,
      "balance_loss_clip": 1.04424274,
      "balance_loss_mlp": 1.0220207,
      "epoch": 0.6630192989839476,
      "flos": 18513567106560.0,
      "grad_norm": 1.6382171626285609,
      "language_loss": 0.70061421,
      "learning_rate": 1.0774049747071883e-06,
      "loss": 0.72227687,
      "num_input_tokens_seen": 118600715,
      "step": 5514,
      "time_per_iteration": 2.7097952365875244
    },
    {
      "auxiliary_loss_clip": 0.01114872,
      "auxiliary_loss_mlp": 0.01028289,
      "balance_loss_clip": 1.04515576,
      "balance_loss_mlp": 1.02088642,
      "epoch": 0.6631395418745867,
      "flos": 35809510049280.0,
      "grad_norm": 2.9863664959167218,
      "language_loss": 0.68248427,
      "learning_rate": 1.076713905602332e-06,
      "loss": 0.70391583,
      "num_input_tokens_seen": 118621290,
      "step": 5515,
      "time_per_iteration": 2.9045188426971436
    },
    {
      "auxiliary_loss_clip": 0.01162276,
      "auxiliary_loss_mlp": 0.01028723,
      "balance_loss_clip": 1.04989243,
      "balance_loss_mlp": 1.02160668,
      "epoch": 0.6632597847652257,
      "flos": 20047742853120.0,
      "grad_norm": 1.71671176800019,
      "language_loss": 0.81306565,
      "learning_rate": 1.07602297655924e-06,
      "loss": 0.8349756,
      "num_input_tokens_seen": 118639610,
      "step": 5516,
      "time_per_iteration": 2.7744195461273193
    },
    {
      "auxiliary_loss_clip": 0.01175741,
      "auxiliary_loss_mlp": 0.01028792,
      "balance_loss_clip": 1.05223,
      "balance_loss_mlp": 1.02153802,
      "epoch": 0.6633800276558649,
      "flos": 21214480423680.0,
      "grad_norm": 1.7762758474308264,
      "language_loss": 0.81181604,
      "learning_rate": 1.0753321876827292e-06,
      "loss": 0.83386135,
      "num_input_tokens_seen": 118658895,
      "step": 5517,
      "time_per_iteration": 2.631793975830078
    },
    {
      "auxiliary_loss_clip": 0.01171993,
      "auxiliary_loss_mlp": 0.01024337,
      "balance_loss_clip": 1.04673946,
      "balance_loss_mlp": 1.01675582,
      "epoch": 0.663500270546504,
      "flos": 23987753688960.0,
      "grad_norm": 3.3397039241567814,
      "language_loss": 0.74144554,
      "learning_rate": 1.0746415390775893e-06,
      "loss": 0.76340878,
      "num_input_tokens_seen": 118677025,
      "step": 5518,
      "time_per_iteration": 2.617201805114746
    },
    {
      "auxiliary_loss_clip": 0.01171345,
      "auxiliary_loss_mlp": 0.01023989,
      "balance_loss_clip": 1.04976737,
      "balance_loss_mlp": 1.01671147,
      "epoch": 0.663620513437143,
      "flos": 17932389050880.0,
      "grad_norm": 1.9381324307562664,
      "language_loss": 0.76520085,
      "learning_rate": 1.0739510308485939e-06,
      "loss": 0.7871542,
      "num_input_tokens_seen": 118694240,
      "step": 5519,
      "time_per_iteration": 4.396620512008667
    },
    {
      "auxiliary_loss_clip": 0.01045816,
      "auxiliary_loss_mlp": 0.01002047,
      "balance_loss_clip": 1.014691,
      "balance_loss_mlp": 1.00061631,
      "epoch": 0.6637407563277821,
      "flos": 57840241086720.0,
      "grad_norm": 0.8153966597613861,
      "language_loss": 0.62413418,
      "learning_rate": 1.07326066310049e-06,
      "loss": 0.64461279,
      "num_input_tokens_seen": 118758365,
      "step": 5520,
      "time_per_iteration": 3.387009859085083
    },
    {
      "auxiliary_loss_clip": 0.01125779,
      "auxiliary_loss_mlp": 0.0102966,
      "balance_loss_clip": 1.04103184,
      "balance_loss_mlp": 1.0216732,
      "epoch": 0.6638609992184212,
      "flos": 27306007079040.0,
      "grad_norm": 1.9136908741476681,
      "language_loss": 0.7922405,
      "learning_rate": 1.0725704359380059e-06,
      "loss": 0.81379485,
      "num_input_tokens_seen": 118778220,
      "step": 5521,
      "time_per_iteration": 2.766756296157837
    },
    {
      "auxiliary_loss_clip": 0.01170267,
      "auxiliary_loss_mlp": 0.01024776,
      "balance_loss_clip": 1.0468179,
      "balance_loss_mlp": 1.01737332,
      "epoch": 0.6639812421090603,
      "flos": 18624854419200.0,
      "grad_norm": 1.761748808697812,
      "language_loss": 0.7212283,
      "learning_rate": 1.0718803494658497e-06,
      "loss": 0.74317873,
      "num_input_tokens_seen": 118797110,
      "step": 5522,
      "time_per_iteration": 3.6498806476593018
    },
    {
      "auxiliary_loss_clip": 0.01069739,
      "auxiliary_loss_mlp": 0.01031858,
      "balance_loss_clip": 1.03992057,
      "balance_loss_mlp": 1.02424049,
      "epoch": 0.6641014849996993,
      "flos": 15924479806080.0,
      "grad_norm": 2.2119213158620648,
      "language_loss": 0.83493382,
      "learning_rate": 1.071190403788707e-06,
      "loss": 0.85594982,
      "num_input_tokens_seen": 118812415,
      "step": 5523,
      "time_per_iteration": 3.697870969772339
    },
    {
      "auxiliary_loss_clip": 0.01137092,
      "auxiliary_loss_mlp": 0.01030202,
      "balance_loss_clip": 1.04632068,
      "balance_loss_mlp": 1.02227414,
      "epoch": 0.6642217278903385,
      "flos": 26505486622080.0,
      "grad_norm": 2.9087334353983776,
      "language_loss": 0.75687087,
      "learning_rate": 1.0705005990112415e-06,
      "loss": 0.77854389,
      "num_input_tokens_seen": 118832195,
      "step": 5524,
      "time_per_iteration": 2.728175401687622
    },
    {
      "auxiliary_loss_clip": 0.01104168,
      "auxiliary_loss_mlp": 0.01022789,
      "balance_loss_clip": 1.04182625,
      "balance_loss_mlp": 1.01499271,
      "epoch": 0.6643419707809776,
      "flos": 15377308951680.0,
      "grad_norm": 2.3802315807045975,
      "language_loss": 0.74571395,
      "learning_rate": 1.0698109352380957e-06,
      "loss": 0.76698351,
      "num_input_tokens_seen": 118849795,
      "step": 5525,
      "time_per_iteration": 2.696197032928467
    },
    {
      "auxiliary_loss_clip": 0.01171618,
      "auxiliary_loss_mlp": 0.01025456,
      "balance_loss_clip": 1.0481056,
      "balance_loss_mlp": 1.01827109,
      "epoch": 0.6644622136716166,
      "flos": 25117610970240.0,
      "grad_norm": 2.7207388164748583,
      "language_loss": 0.78470778,
      "learning_rate": 1.0691214125738909e-06,
      "loss": 0.80667853,
      "num_input_tokens_seen": 118870000,
      "step": 5526,
      "time_per_iteration": 3.573052406311035
    },
    {
      "auxiliary_loss_clip": 0.0107353,
      "auxiliary_loss_mlp": 0.01001616,
      "balance_loss_clip": 1.01535451,
      "balance_loss_mlp": 1.00034082,
      "epoch": 0.6645824565622558,
      "flos": 66201717680640.0,
      "grad_norm": 0.7827417595321566,
      "language_loss": 0.57508975,
      "learning_rate": 1.0684320311232287e-06,
      "loss": 0.59584117,
      "num_input_tokens_seen": 118932905,
      "step": 5527,
      "time_per_iteration": 3.255934238433838
    },
    {
      "auxiliary_loss_clip": 0.01138365,
      "auxiliary_loss_mlp": 0.01022295,
      "balance_loss_clip": 1.04319286,
      "balance_loss_mlp": 1.01442766,
      "epoch": 0.6647026994528948,
      "flos": 25082131311360.0,
      "grad_norm": 1.864349214330365,
      "language_loss": 0.8138752,
      "learning_rate": 1.0677427909906865e-06,
      "loss": 0.83548182,
      "num_input_tokens_seen": 118953355,
      "step": 5528,
      "time_per_iteration": 2.7268123626708984
    },
    {
      "auxiliary_loss_clip": 0.01176423,
      "auxiliary_loss_mlp": 0.01024992,
      "balance_loss_clip": 1.04935682,
      "balance_loss_mlp": 1.01758873,
      "epoch": 0.6648229423435339,
      "flos": 18222187979520.0,
      "grad_norm": 1.896701223612167,
      "language_loss": 0.72520018,
      "learning_rate": 1.0670536922808216e-06,
      "loss": 0.74721432,
      "num_input_tokens_seen": 118973480,
      "step": 5529,
      "time_per_iteration": 2.5958104133605957
    },
    {
      "auxiliary_loss_clip": 0.01143697,
      "auxiliary_loss_mlp": 0.01027434,
      "balance_loss_clip": 1.04495621,
      "balance_loss_mlp": 1.01985252,
      "epoch": 0.6649431852341731,
      "flos": 18296882311680.0,
      "grad_norm": 2.4139559885122197,
      "language_loss": 0.72043157,
      "learning_rate": 1.06636473509817e-06,
      "loss": 0.7421428,
      "num_input_tokens_seen": 118989860,
      "step": 5530,
      "time_per_iteration": 2.6316230297088623
    },
    {
      "auxiliary_loss_clip": 0.01139103,
      "auxiliary_loss_mlp": 0.0076261,
      "balance_loss_clip": 1.0433414,
      "balance_loss_mlp": 1.00062394,
      "epoch": 0.6650634281248121,
      "flos": 17019575700480.0,
      "grad_norm": 1.8292826404799298,
      "language_loss": 0.80721831,
      "learning_rate": 1.0656759195472447e-06,
      "loss": 0.82623541,
      "num_input_tokens_seen": 119007150,
      "step": 5531,
      "time_per_iteration": 2.6730198860168457
    },
    {
      "auxiliary_loss_clip": 0.0104849,
      "auxiliary_loss_mlp": 0.01003192,
      "balance_loss_clip": 1.01320267,
      "balance_loss_mlp": 1.00189841,
      "epoch": 0.6651836710154512,
      "flos": 69294810666240.0,
      "grad_norm": 0.7991494358614435,
      "language_loss": 0.59760058,
      "learning_rate": 1.0649872457325414e-06,
      "loss": 0.61811739,
      "num_input_tokens_seen": 119068435,
      "step": 5532,
      "time_per_iteration": 3.1604621410369873
    },
    {
      "auxiliary_loss_clip": 0.01064672,
      "auxiliary_loss_mlp": 0.01001645,
      "balance_loss_clip": 1.01547956,
      "balance_loss_mlp": 1.00035751,
      "epoch": 0.6653039139060903,
      "flos": 66883444882560.0,
      "grad_norm": 0.8494168811139466,
      "language_loss": 0.55120599,
      "learning_rate": 1.0642987137585278e-06,
      "loss": 0.57186913,
      "num_input_tokens_seen": 119127960,
      "step": 5533,
      "time_per_iteration": 3.1835052967071533
    },
    {
      "auxiliary_loss_clip": 0.0114265,
      "auxiliary_loss_mlp": 0.01029678,
      "balance_loss_clip": 1.04473555,
      "balance_loss_mlp": 1.02182221,
      "epoch": 0.6654241567967294,
      "flos": 21470056669440.0,
      "grad_norm": 2.090779350075203,
      "language_loss": 0.82868552,
      "learning_rate": 1.0636103237296561e-06,
      "loss": 0.85040885,
      "num_input_tokens_seen": 119146885,
      "step": 5534,
      "time_per_iteration": 2.699167490005493
    },
    {
      "auxiliary_loss_clip": 0.0115853,
      "auxiliary_loss_mlp": 0.01032075,
      "balance_loss_clip": 1.04997134,
      "balance_loss_mlp": 1.02448773,
      "epoch": 0.6655443996873684,
      "flos": 25119514391040.0,
      "grad_norm": 1.9465740379906644,
      "language_loss": 0.84064072,
      "learning_rate": 1.062922075750353e-06,
      "loss": 0.86254674,
      "num_input_tokens_seen": 119166900,
      "step": 5535,
      "time_per_iteration": 2.6890530586242676
    },
    {
      "auxiliary_loss_clip": 0.01131999,
      "auxiliary_loss_mlp": 0.01025489,
      "balance_loss_clip": 1.04315042,
      "balance_loss_mlp": 1.01850605,
      "epoch": 0.6656646425780076,
      "flos": 17457326749440.0,
      "grad_norm": 2.5029201574239424,
      "language_loss": 0.7180481,
      "learning_rate": 1.0622339699250267e-06,
      "loss": 0.73962295,
      "num_input_tokens_seen": 119184820,
      "step": 5536,
      "time_per_iteration": 2.7330074310302734
    },
    {
      "auxiliary_loss_clip": 0.01128637,
      "auxiliary_loss_mlp": 0.01020572,
      "balance_loss_clip": 1.04257226,
      "balance_loss_mlp": 1.01330018,
      "epoch": 0.6657848854686467,
      "flos": 23434190213760.0,
      "grad_norm": 2.8347360900146574,
      "language_loss": 0.79378927,
      "learning_rate": 1.0615460063580624e-06,
      "loss": 0.81528133,
      "num_input_tokens_seen": 119203295,
      "step": 5537,
      "time_per_iteration": 2.7683849334716797
    },
    {
      "auxiliary_loss_clip": 0.01144919,
      "auxiliary_loss_mlp": 0.01025352,
      "balance_loss_clip": 1.04476249,
      "balance_loss_mlp": 1.01809549,
      "epoch": 0.6659051283592857,
      "flos": 11509909459200.0,
      "grad_norm": 1.9400198760588567,
      "language_loss": 0.73412549,
      "learning_rate": 1.060858185153821e-06,
      "loss": 0.75582814,
      "num_input_tokens_seen": 119221395,
      "step": 5538,
      "time_per_iteration": 2.6868796348571777
    },
    {
      "auxiliary_loss_clip": 0.01148858,
      "auxiliary_loss_mlp": 0.01028213,
      "balance_loss_clip": 1.04588974,
      "balance_loss_mlp": 1.01997614,
      "epoch": 0.6660253712499249,
      "flos": 20594554571520.0,
      "grad_norm": 2.1910162045180717,
      "language_loss": 0.76172626,
      "learning_rate": 1.0601705064166474e-06,
      "loss": 0.78349698,
      "num_input_tokens_seen": 119239790,
      "step": 5539,
      "time_per_iteration": 2.6174168586730957
    },
    {
      "auxiliary_loss_clip": 0.01139076,
      "auxiliary_loss_mlp": 0.01024619,
      "balance_loss_clip": 1.04600489,
      "balance_loss_mlp": 1.0171268,
      "epoch": 0.666145614140564,
      "flos": 21251504367360.0,
      "grad_norm": 2.008897351834559,
      "language_loss": 0.73215079,
      "learning_rate": 1.0594829702508596e-06,
      "loss": 0.75378776,
      "num_input_tokens_seen": 119257505,
      "step": 5540,
      "time_per_iteration": 2.739698886871338
    },
    {
      "auxiliary_loss_clip": 0.01132421,
      "auxiliary_loss_mlp": 0.01022136,
      "balance_loss_clip": 1.04368496,
      "balance_loss_mlp": 1.01498985,
      "epoch": 0.666265857031203,
      "flos": 33726188200320.0,
      "grad_norm": 1.7255117937733588,
      "language_loss": 0.5519594,
      "learning_rate": 1.0587955767607592e-06,
      "loss": 0.57350492,
      "num_input_tokens_seen": 119279365,
      "step": 5541,
      "time_per_iteration": 2.832576036453247
    },
    {
      "auxiliary_loss_clip": 0.01172462,
      "auxiliary_loss_mlp": 0.01030938,
      "balance_loss_clip": 1.04773057,
      "balance_loss_mlp": 1.02323711,
      "epoch": 0.6663860999218422,
      "flos": 17456644391040.0,
      "grad_norm": 2.1127281204264547,
      "language_loss": 0.77447498,
      "learning_rate": 1.0581083260506206e-06,
      "loss": 0.79650897,
      "num_input_tokens_seen": 119296150,
      "step": 5542,
      "time_per_iteration": 2.6830027103424072
    },
    {
      "auxiliary_loss_clip": 0.01142127,
      "auxiliary_loss_mlp": 0.01023609,
      "balance_loss_clip": 1.0442369,
      "balance_loss_mlp": 1.01627159,
      "epoch": 0.6665063428124812,
      "flos": 17676740977920.0,
      "grad_norm": 2.3251671552687005,
      "language_loss": 0.75991321,
      "learning_rate": 1.0574212182246993e-06,
      "loss": 0.78157055,
      "num_input_tokens_seen": 119314845,
      "step": 5543,
      "time_per_iteration": 2.8738317489624023
    },
    {
      "auxiliary_loss_clip": 0.01149262,
      "auxiliary_loss_mlp": 0.01027854,
      "balance_loss_clip": 1.04489005,
      "balance_loss_mlp": 1.01980126,
      "epoch": 0.6666265857031203,
      "flos": 27673265687040.0,
      "grad_norm": 2.399282588198003,
      "language_loss": 0.76197147,
      "learning_rate": 1.0567342533872303e-06,
      "loss": 0.78374267,
      "num_input_tokens_seen": 119334875,
      "step": 5544,
      "time_per_iteration": 3.6982035636901855
    },
    {
      "auxiliary_loss_clip": 0.01145428,
      "auxiliary_loss_mlp": 0.01028874,
      "balance_loss_clip": 1.04705906,
      "balance_loss_mlp": 1.02158165,
      "epoch": 0.6667468285937594,
      "flos": 25046831220480.0,
      "grad_norm": 1.998371669579996,
      "language_loss": 0.8102715,
      "learning_rate": 1.0560474316424255e-06,
      "loss": 0.83201456,
      "num_input_tokens_seen": 119354635,
      "step": 5545,
      "time_per_iteration": 2.778339385986328
    },
    {
      "auxiliary_loss_clip": 0.01144634,
      "auxiliary_loss_mlp": 0.01024495,
      "balance_loss_clip": 1.04348445,
      "balance_loss_mlp": 1.01648974,
      "epoch": 0.6668670714843985,
      "flos": 22780472641920.0,
      "grad_norm": 2.2428927124532914,
      "language_loss": 0.73888534,
      "learning_rate": 1.0553607530944746e-06,
      "loss": 0.76057661,
      "num_input_tokens_seen": 119372690,
      "step": 5546,
      "time_per_iteration": 2.662250280380249
    },
    {
      "auxiliary_loss_clip": 0.01132408,
      "auxiliary_loss_mlp": 0.01028113,
      "balance_loss_clip": 1.0431782,
      "balance_loss_mlp": 1.02036488,
      "epoch": 0.6669873143750376,
      "flos": 22163886754560.0,
      "grad_norm": 1.9205402199937027,
      "language_loss": 0.89528465,
      "learning_rate": 1.0546742178475463e-06,
      "loss": 0.91688991,
      "num_input_tokens_seen": 119391685,
      "step": 5547,
      "time_per_iteration": 3.6857218742370605
    },
    {
      "auxiliary_loss_clip": 0.01123155,
      "auxiliary_loss_mlp": 0.01028225,
      "balance_loss_clip": 1.04443324,
      "balance_loss_mlp": 1.02121532,
      "epoch": 0.6671075572656767,
      "flos": 20514832335360.0,
      "grad_norm": 2.06214991182939,
      "language_loss": 0.86415339,
      "learning_rate": 1.0539878260057868e-06,
      "loss": 0.8856672,
      "num_input_tokens_seen": 119410725,
      "step": 5548,
      "time_per_iteration": 3.725162982940674
    },
    {
      "auxiliary_loss_clip": 0.01164596,
      "auxiliary_loss_mlp": 0.01029181,
      "balance_loss_clip": 1.05139935,
      "balance_loss_mlp": 1.02051425,
      "epoch": 0.6672278001563158,
      "flos": 17931203902080.0,
      "grad_norm": 2.642008394412806,
      "language_loss": 0.68730009,
      "learning_rate": 1.0533015776733226e-06,
      "loss": 0.70923787,
      "num_input_tokens_seen": 119426875,
      "step": 5549,
      "time_per_iteration": 2.6075572967529297
    },
    {
      "auxiliary_loss_clip": 0.01143848,
      "auxiliary_loss_mlp": 0.01025145,
      "balance_loss_clip": 1.04873848,
      "balance_loss_mlp": 1.01732492,
      "epoch": 0.6673480430469548,
      "flos": 22342146975360.0,
      "grad_norm": 2.2557921307871323,
      "language_loss": 0.79123092,
      "learning_rate": 1.0526154729542566e-06,
      "loss": 0.81292081,
      "num_input_tokens_seen": 119446935,
      "step": 5550,
      "time_per_iteration": 2.848194122314453
    },
    {
      "auxiliary_loss_clip": 0.01132672,
      "auxiliary_loss_mlp": 0.01032247,
      "balance_loss_clip": 1.04733503,
      "balance_loss_mlp": 1.02440262,
      "epoch": 0.6674682859375939,
      "flos": 20703830722560.0,
      "grad_norm": 2.4625949993221026,
      "language_loss": 0.79789776,
      "learning_rate": 1.0519295119526699e-06,
      "loss": 0.81954694,
      "num_input_tokens_seen": 119463240,
      "step": 5551,
      "time_per_iteration": 2.684818744659424
    },
    {
      "auxiliary_loss_clip": 0.01148798,
      "auxiliary_loss_mlp": 0.01028073,
      "balance_loss_clip": 1.04641473,
      "balance_loss_mlp": 1.02062261,
      "epoch": 0.667588528828233,
      "flos": 26206673379840.0,
      "grad_norm": 1.6745406205892186,
      "language_loss": 0.8344866,
      "learning_rate": 1.0512436947726227e-06,
      "loss": 0.85625529,
      "num_input_tokens_seen": 119484655,
      "step": 5552,
      "time_per_iteration": 3.5992443561553955
    },
    {
      "auxiliary_loss_clip": 0.01131935,
      "auxiliary_loss_mlp": 0.01025893,
      "balance_loss_clip": 1.04267776,
      "balance_loss_mlp": 1.01820445,
      "epoch": 0.6677087717188721,
      "flos": 23071025756160.0,
      "grad_norm": 2.132148561135527,
      "language_loss": 0.65313941,
      "learning_rate": 1.0505580215181517e-06,
      "loss": 0.67471772,
      "num_input_tokens_seen": 119502895,
      "step": 5553,
      "time_per_iteration": 2.689302444458008
    },
    {
      "auxiliary_loss_clip": 0.01035998,
      "auxiliary_loss_mlp": 0.01003108,
      "balance_loss_clip": 1.01910472,
      "balance_loss_mlp": 1.00191009,
      "epoch": 0.6678290146095112,
      "flos": 70941315219840.0,
      "grad_norm": 0.7810518136881069,
      "language_loss": 0.56627226,
      "learning_rate": 1.0498724922932753e-06,
      "loss": 0.58666337,
      "num_input_tokens_seen": 119561010,
      "step": 5554,
      "time_per_iteration": 3.231783628463745
    },
    {
      "auxiliary_loss_clip": 0.01180063,
      "auxiliary_loss_mlp": 0.0103022,
      "balance_loss_clip": 1.05142224,
      "balance_loss_mlp": 1.02167869,
      "epoch": 0.6679492575001503,
      "flos": 18661088263680.0,
      "grad_norm": 2.4046113593420815,
      "language_loss": 0.86040312,
      "learning_rate": 1.0491871072019851e-06,
      "loss": 0.88250589,
      "num_input_tokens_seen": 119578900,
      "step": 5555,
      "time_per_iteration": 2.610520601272583
    },
    {
      "auxiliary_loss_clip": 0.01134965,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.04158652,
      "balance_loss_mlp": 1.01936173,
      "epoch": 0.6680695003907894,
      "flos": 29711985822720.0,
      "grad_norm": 1.6425556443736054,
      "language_loss": 0.6395846,
      "learning_rate": 1.0485018663482555e-06,
      "loss": 0.66120613,
      "num_input_tokens_seen": 119598920,
      "step": 5556,
      "time_per_iteration": 2.7486085891723633
    },
    {
      "auxiliary_loss_clip": 0.0115557,
      "auxiliary_loss_mlp": 0.01028927,
      "balance_loss_clip": 1.04736733,
      "balance_loss_mlp": 1.02050447,
      "epoch": 0.6681897432814284,
      "flos": 28218964083840.0,
      "grad_norm": 2.8387395700614073,
      "language_loss": 0.70111781,
      "learning_rate": 1.0478167698360354e-06,
      "loss": 0.7229628,
      "num_input_tokens_seen": 119618220,
      "step": 5557,
      "time_per_iteration": 2.7754080295562744
    },
    {
      "auxiliary_loss_clip": 0.01150803,
      "auxiliary_loss_mlp": 0.01029571,
      "balance_loss_clip": 1.04415524,
      "balance_loss_mlp": 1.02157211,
      "epoch": 0.6683099861720676,
      "flos": 25046543911680.0,
      "grad_norm": 2.006459595991925,
      "language_loss": 0.70432043,
      "learning_rate": 1.0471318177692556e-06,
      "loss": 0.72612417,
      "num_input_tokens_seen": 119638520,
      "step": 5558,
      "time_per_iteration": 2.7546021938323975
    },
    {
      "auxiliary_loss_clip": 0.01118769,
      "auxiliary_loss_mlp": 0.01028653,
      "balance_loss_clip": 1.04133642,
      "balance_loss_mlp": 1.02080655,
      "epoch": 0.6684302290627067,
      "flos": 22996977868800.0,
      "grad_norm": 2.3825292521234744,
      "language_loss": 0.75916195,
      "learning_rate": 1.046447010251821e-06,
      "loss": 0.78063619,
      "num_input_tokens_seen": 119655850,
      "step": 5559,
      "time_per_iteration": 2.6862878799438477
    },
    {
      "auxiliary_loss_clip": 0.01145705,
      "auxiliary_loss_mlp": 0.01025244,
      "balance_loss_clip": 1.04787004,
      "balance_loss_mlp": 1.01828527,
      "epoch": 0.6685504719533457,
      "flos": 26573824247040.0,
      "grad_norm": 1.6968736109727582,
      "language_loss": 0.75716984,
      "learning_rate": 1.0457623473876157e-06,
      "loss": 0.77887928,
      "num_input_tokens_seen": 119675355,
      "step": 5560,
      "time_per_iteration": 2.732908010482788
    },
    {
      "auxiliary_loss_clip": 0.01172653,
      "auxiliary_loss_mlp": 0.01029955,
      "balance_loss_clip": 1.04925632,
      "balance_loss_mlp": 1.0226655,
      "epoch": 0.6686707148439849,
      "flos": 28986087870720.0,
      "grad_norm": 2.4125437328347967,
      "language_loss": 0.70816654,
      "learning_rate": 1.0450778292805046e-06,
      "loss": 0.73019266,
      "num_input_tokens_seen": 119695340,
      "step": 5561,
      "time_per_iteration": 2.6331405639648438
    },
    {
      "auxiliary_loss_clip": 0.01159394,
      "auxiliary_loss_mlp": 0.01028929,
      "balance_loss_clip": 1.04498947,
      "balance_loss_mlp": 1.02133799,
      "epoch": 0.6687909577346239,
      "flos": 23623152687360.0,
      "grad_norm": 2.046894057621575,
      "language_loss": 0.78544241,
      "learning_rate": 1.0443934560343267e-06,
      "loss": 0.80732566,
      "num_input_tokens_seen": 119716750,
      "step": 5562,
      "time_per_iteration": 2.6977884769439697
    },
    {
      "auxiliary_loss_clip": 0.01118841,
      "auxiliary_loss_mlp": 0.01024806,
      "balance_loss_clip": 1.04221749,
      "balance_loss_mlp": 1.0173676,
      "epoch": 0.668911200625263,
      "flos": 23148593176320.0,
      "grad_norm": 1.9017394458728372,
      "language_loss": 0.77942491,
      "learning_rate": 1.0437092277529034e-06,
      "loss": 0.8008613,
      "num_input_tokens_seen": 119736005,
      "step": 5563,
      "time_per_iteration": 2.7570395469665527
    },
    {
      "auxiliary_loss_clip": 0.01137751,
      "auxiliary_loss_mlp": 0.01033949,
      "balance_loss_clip": 1.04308224,
      "balance_loss_mlp": 1.02600968,
      "epoch": 0.6690314435159022,
      "flos": 18551919853440.0,
      "grad_norm": 3.961189391444136,
      "language_loss": 0.7369653,
      "learning_rate": 1.0430251445400292e-06,
      "loss": 0.75868225,
      "num_input_tokens_seen": 119754050,
      "step": 5564,
      "time_per_iteration": 2.6391401290893555
    },
    {
      "auxiliary_loss_clip": 0.01077736,
      "auxiliary_loss_mlp": 0.01026292,
      "balance_loss_clip": 1.04191935,
      "balance_loss_mlp": 1.01837027,
      "epoch": 0.6691516864065412,
      "flos": 31759540704000.0,
      "grad_norm": 2.111815697224342,
      "language_loss": 0.62629253,
      "learning_rate": 1.0423412064994787e-06,
      "loss": 0.64733279,
      "num_input_tokens_seen": 119774820,
      "step": 5565,
      "time_per_iteration": 3.149540662765503
    },
    {
      "auxiliary_loss_clip": 0.01132235,
      "auxiliary_loss_mlp": 0.01026322,
      "balance_loss_clip": 1.04329014,
      "balance_loss_mlp": 1.01967049,
      "epoch": 0.6692719292971803,
      "flos": 34933864296960.0,
      "grad_norm": 1.9897151254567262,
      "language_loss": 0.73970068,
      "learning_rate": 1.0416574137350064e-06,
      "loss": 0.76128626,
      "num_input_tokens_seen": 119795525,
      "step": 5566,
      "time_per_iteration": 3.1727914810180664
    },
    {
      "auxiliary_loss_clip": 0.01151254,
      "auxiliary_loss_mlp": 0.0102348,
      "balance_loss_clip": 1.0451107,
      "balance_loss_mlp": 1.01611841,
      "epoch": 0.6693921721878194,
      "flos": 20449188230400.0,
      "grad_norm": 2.1480724538604474,
      "language_loss": 0.80817735,
      "learning_rate": 1.0409737663503428e-06,
      "loss": 0.8299247,
      "num_input_tokens_seen": 119813905,
      "step": 5567,
      "time_per_iteration": 2.671480178833008
    },
    {
      "auxiliary_loss_clip": 0.01156048,
      "auxiliary_loss_mlp": 0.01025116,
      "balance_loss_clip": 1.04544997,
      "balance_loss_mlp": 1.01783574,
      "epoch": 0.6695124150784585,
      "flos": 16614538963200.0,
      "grad_norm": 1.83301664102826,
      "language_loss": 0.82930779,
      "learning_rate": 1.040290264449196e-06,
      "loss": 0.85111946,
      "num_input_tokens_seen": 119832010,
      "step": 5568,
      "time_per_iteration": 2.7063639163970947
    },
    {
      "auxiliary_loss_clip": 0.01154908,
      "auxiliary_loss_mlp": 0.01026937,
      "balance_loss_clip": 1.04830527,
      "balance_loss_mlp": 1.01965022,
      "epoch": 0.6696326579690975,
      "flos": 26652145852800.0,
      "grad_norm": 2.049665930997182,
      "language_loss": 0.64095235,
      "learning_rate": 1.0396069081352532e-06,
      "loss": 0.66277075,
      "num_input_tokens_seen": 119851165,
      "step": 5569,
      "time_per_iteration": 2.6890246868133545
    },
    {
      "auxiliary_loss_clip": 0.01072215,
      "auxiliary_loss_mlp": 0.01001855,
      "balance_loss_clip": 1.01441264,
      "balance_loss_mlp": 1.00057948,
      "epoch": 0.6697529008597367,
      "flos": 66964603662720.0,
      "grad_norm": 0.77194488483529,
      "language_loss": 0.55996835,
      "learning_rate": 1.0389236975121782e-06,
      "loss": 0.5807091,
      "num_input_tokens_seen": 119906015,
      "step": 5570,
      "time_per_iteration": 4.203742027282715
    },
    {
      "auxiliary_loss_clip": 0.01174761,
      "auxiliary_loss_mlp": 0.01025409,
      "balance_loss_clip": 1.04887116,
      "balance_loss_mlp": 1.01751173,
      "epoch": 0.6698731437503758,
      "flos": 20886939279360.0,
      "grad_norm": 2.2111947541401884,
      "language_loss": 0.71847427,
      "learning_rate": 1.0382406326836147e-06,
      "loss": 0.74047595,
      "num_input_tokens_seen": 119925160,
      "step": 5571,
      "time_per_iteration": 2.9389092922210693
    },
    {
      "auxiliary_loss_clip": 0.01165885,
      "auxiliary_loss_mlp": 0.01025774,
      "balance_loss_clip": 1.04897213,
      "balance_loss_mlp": 1.01793575,
      "epoch": 0.6699933866410148,
      "flos": 20409470766720.0,
      "grad_norm": 1.974523727208466,
      "language_loss": 0.75804532,
      "learning_rate": 1.0375577137531828e-06,
      "loss": 0.77996194,
      "num_input_tokens_seen": 119943720,
      "step": 5572,
      "time_per_iteration": 2.732252597808838
    },
    {
      "auxiliary_loss_clip": 0.01144717,
      "auxiliary_loss_mlp": 0.01028952,
      "balance_loss_clip": 1.04474759,
      "balance_loss_mlp": 1.02117705,
      "epoch": 0.670113629531654,
      "flos": 29023075900800.0,
      "grad_norm": 1.5336609638190344,
      "language_loss": 0.72237623,
      "learning_rate": 1.0368749408244802e-06,
      "loss": 0.74411297,
      "num_input_tokens_seen": 119966640,
      "step": 5573,
      "time_per_iteration": 3.6708884239196777
    },
    {
      "auxiliary_loss_clip": 0.01152566,
      "auxiliary_loss_mlp": 0.01021279,
      "balance_loss_clip": 1.04675555,
      "balance_loss_mlp": 1.01441002,
      "epoch": 0.670233872422293,
      "flos": 19791699730560.0,
      "grad_norm": 2.6529943433730665,
      "language_loss": 0.78822947,
      "learning_rate": 1.0361923140010836e-06,
      "loss": 0.80996799,
      "num_input_tokens_seen": 119985125,
      "step": 5574,
      "time_per_iteration": 3.6088156700134277
    },
    {
      "auxiliary_loss_clip": 0.01165965,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.04881418,
      "balance_loss_mlp": 1.01885831,
      "epoch": 0.6703541153129321,
      "flos": 24243689070720.0,
      "grad_norm": 2.0973974730599747,
      "language_loss": 0.63831913,
      "learning_rate": 1.0355098333865455e-06,
      "loss": 0.66024256,
      "num_input_tokens_seen": 120004355,
      "step": 5575,
      "time_per_iteration": 2.633697509765625
    },
    {
      "auxiliary_loss_clip": 0.01157807,
      "auxiliary_loss_mlp": 0.01024079,
      "balance_loss_clip": 1.05144298,
      "balance_loss_mlp": 1.01684046,
      "epoch": 0.6704743582035713,
      "flos": 26688523351680.0,
      "grad_norm": 1.6683597898468823,
      "language_loss": 0.69136381,
      "learning_rate": 1.0348274990844006e-06,
      "loss": 0.71318263,
      "num_input_tokens_seen": 120027115,
      "step": 5576,
      "time_per_iteration": 2.6844990253448486
    },
    {
      "auxiliary_loss_clip": 0.01159068,
      "auxiliary_loss_mlp": 0.01024004,
      "balance_loss_clip": 1.04876018,
      "balance_loss_mlp": 1.01665473,
      "epoch": 0.6705946010942103,
      "flos": 23514379326720.0,
      "grad_norm": 1.701636998764028,
      "language_loss": 0.72359014,
      "learning_rate": 1.034145311198155e-06,
      "loss": 0.74542087,
      "num_input_tokens_seen": 120047130,
      "step": 5577,
      "time_per_iteration": 2.6996164321899414
    },
    {
      "auxiliary_loss_clip": 0.01171117,
      "auxiliary_loss_mlp": 0.010253,
      "balance_loss_clip": 1.04890621,
      "balance_loss_mlp": 1.01829624,
      "epoch": 0.6707148439848494,
      "flos": 24061011477120.0,
      "grad_norm": 1.8395568624541752,
      "language_loss": 0.639714,
      "learning_rate": 1.0334632698312989e-06,
      "loss": 0.66167814,
      "num_input_tokens_seen": 120067925,
      "step": 5578,
      "time_per_iteration": 3.5222008228302
    },
    {
      "auxiliary_loss_clip": 0.01138315,
      "auxiliary_loss_mlp": 0.0102515,
      "balance_loss_clip": 1.04449928,
      "balance_loss_mlp": 1.01754737,
      "epoch": 0.6708350868754885,
      "flos": 22528667324160.0,
      "grad_norm": 1.9985962775434556,
      "language_loss": 0.75043499,
      "learning_rate": 1.032781375087295e-06,
      "loss": 0.77206963,
      "num_input_tokens_seen": 120087825,
      "step": 5579,
      "time_per_iteration": 2.7018957138061523
    },
    {
      "auxiliary_loss_clip": 0.01146169,
      "auxiliary_loss_mlp": 0.01025597,
      "balance_loss_clip": 1.04732573,
      "balance_loss_mlp": 1.01862001,
      "epoch": 0.6709553297661276,
      "flos": 25227749047680.0,
      "grad_norm": 1.3840362967114743,
      "language_loss": 0.6737597,
      "learning_rate": 1.0320996270695891e-06,
      "loss": 0.69547737,
      "num_input_tokens_seen": 120108895,
      "step": 5580,
      "time_per_iteration": 2.7406699657440186
    },
    {
      "auxiliary_loss_clip": 0.01129256,
      "auxiliary_loss_mlp": 0.01029928,
      "balance_loss_clip": 1.04245448,
      "balance_loss_mlp": 1.02161884,
      "epoch": 0.6710755726567667,
      "flos": 20448757267200.0,
      "grad_norm": 1.7452227913957243,
      "language_loss": 0.73362339,
      "learning_rate": 1.0314180258815998e-06,
      "loss": 0.75521517,
      "num_input_tokens_seen": 120127535,
      "step": 5581,
      "time_per_iteration": 2.6912853717803955
    },
    {
      "auxiliary_loss_clip": 0.01123523,
      "auxiliary_loss_mlp": 0.01027911,
      "balance_loss_clip": 1.04335976,
      "balance_loss_mlp": 1.02085686,
      "epoch": 0.6711958155474057,
      "flos": 25995411538560.0,
      "grad_norm": 1.8354039945466019,
      "language_loss": 0.74463153,
      "learning_rate": 1.0307365716267247e-06,
      "loss": 0.76614583,
      "num_input_tokens_seen": 120147980,
      "step": 5582,
      "time_per_iteration": 2.79166579246521
    },
    {
      "auxiliary_loss_clip": 0.01158256,
      "auxiliary_loss_mlp": 0.01024412,
      "balance_loss_clip": 1.04879987,
      "balance_loss_mlp": 1.01626682,
      "epoch": 0.6713160584380449,
      "flos": 19937712516480.0,
      "grad_norm": 5.65532134602039,
      "language_loss": 0.78064108,
      "learning_rate": 1.0300552644083423e-06,
      "loss": 0.8024677,
      "num_input_tokens_seen": 120166905,
      "step": 5583,
      "time_per_iteration": 2.647261619567871
    },
    {
      "auxiliary_loss_clip": 0.01135029,
      "auxiliary_loss_mlp": 0.01027266,
      "balance_loss_clip": 1.04789531,
      "balance_loss_mlp": 1.0187304,
      "epoch": 0.6714363013286839,
      "flos": 18223373128320.0,
      "grad_norm": 2.5008472623433744,
      "language_loss": 0.72625077,
      "learning_rate": 1.0293741043298036e-06,
      "loss": 0.74787378,
      "num_input_tokens_seen": 120185255,
      "step": 5584,
      "time_per_iteration": 2.7223284244537354
    },
    {
      "auxiliary_loss_clip": 0.01135638,
      "auxiliary_loss_mlp": 0.01030941,
      "balance_loss_clip": 1.04866767,
      "balance_loss_mlp": 1.02281666,
      "epoch": 0.671556544219323,
      "flos": 25812374808960.0,
      "grad_norm": 4.638312229081315,
      "language_loss": 0.71384537,
      "learning_rate": 1.0286930914944436e-06,
      "loss": 0.73551118,
      "num_input_tokens_seen": 120205070,
      "step": 5585,
      "time_per_iteration": 2.7481961250305176
    },
    {
      "auxiliary_loss_clip": 0.01171863,
      "auxiliary_loss_mlp": 0.01022665,
      "balance_loss_clip": 1.04697692,
      "balance_loss_mlp": 1.01511884,
      "epoch": 0.6716767871099621,
      "flos": 15850431918720.0,
      "grad_norm": 2.6387821665956985,
      "language_loss": 0.77316111,
      "learning_rate": 1.0280122260055684e-06,
      "loss": 0.79510641,
      "num_input_tokens_seen": 120220780,
      "step": 5586,
      "time_per_iteration": 2.5618083477020264
    },
    {
      "auxiliary_loss_clip": 0.01174133,
      "auxiliary_loss_mlp": 0.0102469,
      "balance_loss_clip": 1.04990554,
      "balance_loss_mlp": 1.01681042,
      "epoch": 0.6717970300006012,
      "flos": 19756112330880.0,
      "grad_norm": 1.9781844718713213,
      "language_loss": 0.82349598,
      "learning_rate": 1.0273315079664652e-06,
      "loss": 0.84548414,
      "num_input_tokens_seen": 120238735,
      "step": 5587,
      "time_per_iteration": 2.561619520187378
    },
    {
      "auxiliary_loss_clip": 0.01160297,
      "auxiliary_loss_mlp": 0.01028199,
      "balance_loss_clip": 1.04704726,
      "balance_loss_mlp": 1.02090645,
      "epoch": 0.6719172728912403,
      "flos": 25485049146240.0,
      "grad_norm": 2.480801711658586,
      "language_loss": 0.74358392,
      "learning_rate": 1.0266509374803992e-06,
      "loss": 0.7654689,
      "num_input_tokens_seen": 120259895,
      "step": 5588,
      "time_per_iteration": 2.7013163566589355
    },
    {
      "auxiliary_loss_clip": 0.01173365,
      "auxiliary_loss_mlp": 0.00763063,
      "balance_loss_clip": 1.04824317,
      "balance_loss_mlp": 1.00061631,
      "epoch": 0.6720375157818794,
      "flos": 15880344969600.0,
      "grad_norm": 2.295028816792916,
      "language_loss": 0.84097105,
      "learning_rate": 1.0259705146506123e-06,
      "loss": 0.86033535,
      "num_input_tokens_seen": 120274790,
      "step": 5589,
      "time_per_iteration": 2.5371172428131104
    },
    {
      "auxiliary_loss_clip": 0.01159391,
      "auxiliary_loss_mlp": 0.01027818,
      "balance_loss_clip": 1.0460279,
      "balance_loss_mlp": 1.01998281,
      "epoch": 0.6721577586725185,
      "flos": 32010843231360.0,
      "grad_norm": 2.644940252836303,
      "language_loss": 0.77787066,
      "learning_rate": 1.025290239580324e-06,
      "loss": 0.79974282,
      "num_input_tokens_seen": 120295460,
      "step": 5590,
      "time_per_iteration": 2.732830286026001
    },
    {
      "auxiliary_loss_clip": 0.01115731,
      "auxiliary_loss_mlp": 0.01022062,
      "balance_loss_clip": 1.04243672,
      "balance_loss_mlp": 1.01473343,
      "epoch": 0.6722780015631575,
      "flos": 20737873837440.0,
      "grad_norm": 1.7007563597323496,
      "language_loss": 0.75632882,
      "learning_rate": 1.0246101123727313e-06,
      "loss": 0.77770668,
      "num_input_tokens_seen": 120314440,
      "step": 5591,
      "time_per_iteration": 2.8272366523742676
    },
    {
      "auxiliary_loss_clip": 0.01159759,
      "auxiliary_loss_mlp": 0.01024891,
      "balance_loss_clip": 1.04777563,
      "balance_loss_mlp": 1.01782763,
      "epoch": 0.6723982444537967,
      "flos": 16909617191040.0,
      "grad_norm": 1.9507592644277296,
      "language_loss": 0.78894389,
      "learning_rate": 1.0239301331310085e-06,
      "loss": 0.81079042,
      "num_input_tokens_seen": 120332060,
      "step": 5592,
      "time_per_iteration": 2.6467511653900146
    },
    {
      "auxiliary_loss_clip": 0.01156266,
      "auxiliary_loss_mlp": 0.01026384,
      "balance_loss_clip": 1.04741931,
      "balance_loss_mlp": 1.01888609,
      "epoch": 0.6725184873444358,
      "flos": 20667812359680.0,
      "grad_norm": 2.2218773924535764,
      "language_loss": 0.88432622,
      "learning_rate": 1.0232503019583088e-06,
      "loss": 0.90615273,
      "num_input_tokens_seen": 120351670,
      "step": 5593,
      "time_per_iteration": 2.653676748275757
    },
    {
      "auxiliary_loss_clip": 0.01155032,
      "auxiliary_loss_mlp": 0.01029669,
      "balance_loss_clip": 1.04781914,
      "balance_loss_mlp": 1.02135766,
      "epoch": 0.6726387302350748,
      "flos": 23727616416000.0,
      "grad_norm": 1.8048514000805427,
      "language_loss": 0.69855452,
      "learning_rate": 1.0225706189577619e-06,
      "loss": 0.72040153,
      "num_input_tokens_seen": 120370195,
      "step": 5594,
      "time_per_iteration": 2.6803197860717773
    },
    {
      "auxiliary_loss_clip": 0.0116055,
      "auxiliary_loss_mlp": 0.01024622,
      "balance_loss_clip": 1.04869354,
      "balance_loss_mlp": 1.01669478,
      "epoch": 0.672758973125714,
      "flos": 15188274650880.0,
      "grad_norm": 2.0774472263956296,
      "language_loss": 0.74735409,
      "learning_rate": 1.021891084232475e-06,
      "loss": 0.76920581,
      "num_input_tokens_seen": 120388130,
      "step": 5595,
      "time_per_iteration": 2.640676736831665
    },
    {
      "auxiliary_loss_clip": 0.01159559,
      "auxiliary_loss_mlp": 0.01025277,
      "balance_loss_clip": 1.04673696,
      "balance_loss_mlp": 1.01788592,
      "epoch": 0.672879216016353,
      "flos": 18077252601600.0,
      "grad_norm": 2.5382746610715645,
      "language_loss": 0.80094409,
      "learning_rate": 1.0212116978855325e-06,
      "loss": 0.82279247,
      "num_input_tokens_seen": 120406145,
      "step": 5596,
      "time_per_iteration": 3.6259071826934814
    },
    {
      "auxiliary_loss_clip": 0.01127309,
      "auxiliary_loss_mlp": 0.01027907,
      "balance_loss_clip": 1.04250932,
      "balance_loss_mlp": 1.02058721,
      "epoch": 0.6729994589069921,
      "flos": 23476349802240.0,
      "grad_norm": 1.6732178573438625,
      "language_loss": 0.78808749,
      "learning_rate": 1.020532460019997e-06,
      "loss": 0.80963969,
      "num_input_tokens_seen": 120425395,
      "step": 5597,
      "time_per_iteration": 2.7591142654418945
    },
    {
      "auxiliary_loss_clip": 0.01091124,
      "auxiliary_loss_mlp": 0.01030202,
      "balance_loss_clip": 1.04088092,
      "balance_loss_mlp": 1.02293622,
      "epoch": 0.6731197017976313,
      "flos": 26322018929280.0,
      "grad_norm": 1.9225533321404857,
      "language_loss": 0.70734155,
      "learning_rate": 1.0198533707389096e-06,
      "loss": 0.72855479,
      "num_input_tokens_seen": 120446270,
      "step": 5598,
      "time_per_iteration": 2.962085723876953
    },
    {
      "auxiliary_loss_clip": 0.01156736,
      "auxiliary_loss_mlp": 0.00762741,
      "balance_loss_clip": 1.04813623,
      "balance_loss_mlp": 1.00054765,
      "epoch": 0.6732399446882703,
      "flos": 21616428591360.0,
      "grad_norm": 1.979576538143776,
      "language_loss": 0.73137981,
      "learning_rate": 1.0191744301452853e-06,
      "loss": 0.75057459,
      "num_input_tokens_seen": 120465570,
      "step": 5599,
      "time_per_iteration": 3.135270118713379
    },
    {
      "auxiliary_loss_clip": 0.01171699,
      "auxiliary_loss_mlp": 0.01027092,
      "balance_loss_clip": 1.04660749,
      "balance_loss_mlp": 1.01965916,
      "epoch": 0.6733601875789094,
      "flos": 25880173729920.0,
      "grad_norm": 1.7644782552543798,
      "language_loss": 0.704988,
      "learning_rate": 1.0184956383421208e-06,
      "loss": 0.72697592,
      "num_input_tokens_seen": 120484220,
      "step": 5600,
      "time_per_iteration": 4.532362461090088
    },
    {
      "auxiliary_loss_clip": 0.01163875,
      "auxiliary_loss_mlp": 0.01030475,
      "balance_loss_clip": 1.05010533,
      "balance_loss_mlp": 1.0225774,
      "epoch": 0.6734804304695485,
      "flos": 22929573997440.0,
      "grad_norm": 2.980124969205153,
      "language_loss": 0.65551782,
      "learning_rate": 1.017816995432387e-06,
      "loss": 0.67746127,
      "num_input_tokens_seen": 120503320,
      "step": 5601,
      "time_per_iteration": 2.6792614459991455
    },
    {
      "auxiliary_loss_clip": 0.0114318,
      "auxiliary_loss_mlp": 0.01025237,
      "balance_loss_clip": 1.04496193,
      "balance_loss_mlp": 1.0177747,
      "epoch": 0.6736006733601876,
      "flos": 18697968552960.0,
      "grad_norm": 2.3164402357509775,
      "language_loss": 0.74617624,
      "learning_rate": 1.0171385015190353e-06,
      "loss": 0.76786035,
      "num_input_tokens_seen": 120523180,
      "step": 5602,
      "time_per_iteration": 2.7164807319641113
    },
    {
      "auxiliary_loss_clip": 0.01140382,
      "auxiliary_loss_mlp": 0.00761838,
      "balance_loss_clip": 1.04821968,
      "balance_loss_mlp": 1.00047779,
      "epoch": 0.6737209162508266,
      "flos": 19427745173760.0,
      "grad_norm": 2.0824687294503494,
      "language_loss": 0.73097634,
      "learning_rate": 1.0164601567049908e-06,
      "loss": 0.74999857,
      "num_input_tokens_seen": 120541710,
      "step": 5603,
      "time_per_iteration": 2.657412052154541
    },
    {
      "auxiliary_loss_clip": 0.01143356,
      "auxiliary_loss_mlp": 0.01035625,
      "balance_loss_clip": 1.04636931,
      "balance_loss_mlp": 1.02740562,
      "epoch": 0.6738411591414658,
      "flos": 20158060498560.0,
      "grad_norm": 1.7380537021400595,
      "language_loss": 0.80120832,
      "learning_rate": 1.015781961093158e-06,
      "loss": 0.82299817,
      "num_input_tokens_seen": 120561030,
      "step": 5604,
      "time_per_iteration": 3.6905157566070557
    },
    {
      "auxiliary_loss_clip": 0.01146615,
      "auxiliary_loss_mlp": 0.01026144,
      "balance_loss_clip": 1.04210949,
      "balance_loss_mlp": 1.0189141,
      "epoch": 0.6739614020321049,
      "flos": 21653847584640.0,
      "grad_norm": 1.6547121668972335,
      "language_loss": 0.77087247,
      "learning_rate": 1.0151039147864197e-06,
      "loss": 0.7926001,
      "num_input_tokens_seen": 120581005,
      "step": 5605,
      "time_per_iteration": 2.735487222671509
    },
    {
      "auxiliary_loss_clip": 0.01086711,
      "auxiliary_loss_mlp": 0.01021433,
      "balance_loss_clip": 1.04620957,
      "balance_loss_mlp": 1.0133028,
      "epoch": 0.6740816449227439,
      "flos": 19171702051200.0,
      "grad_norm": 1.9949174297970291,
      "language_loss": 0.66060144,
      "learning_rate": 1.0144260178876336e-06,
      "loss": 0.68168288,
      "num_input_tokens_seen": 120600350,
      "step": 5606,
      "time_per_iteration": 2.8177146911621094
    },
    {
      "auxiliary_loss_clip": 0.01151305,
      "auxiliary_loss_mlp": 0.01028392,
      "balance_loss_clip": 1.04727447,
      "balance_loss_mlp": 1.02103949,
      "epoch": 0.6742018878133831,
      "flos": 21097015971840.0,
      "grad_norm": 2.4496737697392774,
      "language_loss": 0.66957605,
      "learning_rate": 1.0137482704996388e-06,
      "loss": 0.69137299,
      "num_input_tokens_seen": 120614700,
      "step": 5607,
      "time_per_iteration": 2.681610345840454
    },
    {
      "auxiliary_loss_clip": 0.01130912,
      "auxiliary_loss_mlp": 0.01025113,
      "balance_loss_clip": 1.04421699,
      "balance_loss_mlp": 1.01700675,
      "epoch": 0.6743221307040221,
      "flos": 23549966726400.0,
      "grad_norm": 2.164569296270932,
      "language_loss": 0.78940463,
      "learning_rate": 1.0130706727252461e-06,
      "loss": 0.81096482,
      "num_input_tokens_seen": 120631755,
      "step": 5608,
      "time_per_iteration": 2.689283847808838
    },
    {
      "auxiliary_loss_clip": 0.01135769,
      "auxiliary_loss_mlp": 0.01030542,
      "balance_loss_clip": 1.04579091,
      "balance_loss_mlp": 1.02293038,
      "epoch": 0.6744423735946612,
      "flos": 16249542912000.0,
      "grad_norm": 2.5762382794721397,
      "language_loss": 0.68218058,
      "learning_rate": 1.0123932246672468e-06,
      "loss": 0.70384371,
      "num_input_tokens_seen": 120645900,
      "step": 5609,
      "time_per_iteration": 2.722545623779297
    },
    {
      "auxiliary_loss_clip": 0.01026009,
      "auxiliary_loss_mlp": 0.00754235,
      "balance_loss_clip": 1.01242709,
      "balance_loss_mlp": 1.00050151,
      "epoch": 0.6745626164853004,
      "flos": 57843257829120.0,
      "grad_norm": 0.8028472716722818,
      "language_loss": 0.55793214,
      "learning_rate": 1.0117159264284114e-06,
      "loss": 0.57573462,
      "num_input_tokens_seen": 120709070,
      "step": 5610,
      "time_per_iteration": 3.229762554168701
    },
    {
      "auxiliary_loss_clip": 0.01146399,
      "auxiliary_loss_mlp": 0.01027652,
      "balance_loss_clip": 1.04584312,
      "balance_loss_mlp": 1.01959038,
      "epoch": 0.6746828593759394,
      "flos": 20485027025280.0,
      "grad_norm": 1.6199596471313982,
      "language_loss": 0.77036834,
      "learning_rate": 1.0110387781114837e-06,
      "loss": 0.79210889,
      "num_input_tokens_seen": 120727685,
      "step": 5611,
      "time_per_iteration": 2.740227460861206
    },
    {
      "auxiliary_loss_clip": 0.01171686,
      "auxiliary_loss_mlp": 0.01032665,
      "balance_loss_clip": 1.04830587,
      "balance_loss_mlp": 1.02478504,
      "epoch": 0.6748031022665785,
      "flos": 19208223204480.0,
      "grad_norm": 1.9751819457893176,
      "language_loss": 0.77098811,
      "learning_rate": 1.0103617798191872e-06,
      "loss": 0.79303169,
      "num_input_tokens_seen": 120747160,
      "step": 5612,
      "time_per_iteration": 2.667494058609009
    },
    {
      "auxiliary_loss_clip": 0.01141049,
      "auxiliary_loss_mlp": 0.01027562,
      "balance_loss_clip": 1.04592025,
      "balance_loss_mlp": 1.01973629,
      "epoch": 0.6749233451572175,
      "flos": 15195026407680.0,
      "grad_norm": 2.4898386321441786,
      "language_loss": 0.83021438,
      "learning_rate": 1.0096849316542217e-06,
      "loss": 0.85190046,
      "num_input_tokens_seen": 120763710,
      "step": 5613,
      "time_per_iteration": 2.6229987144470215
    },
    {
      "auxiliary_loss_clip": 0.01073739,
      "auxiliary_loss_mlp": 0.01025765,
      "balance_loss_clip": 1.03619325,
      "balance_loss_mlp": 1.0179987,
      "epoch": 0.6750435880478567,
      "flos": 26499489050880.0,
      "grad_norm": 5.548352424766525,
      "language_loss": 0.74533129,
      "learning_rate": 1.0090082337192643e-06,
      "loss": 0.76632631,
      "num_input_tokens_seen": 120783355,
      "step": 5614,
      "time_per_iteration": 2.8566932678222656
    },
    {
      "auxiliary_loss_clip": 0.01094433,
      "auxiliary_loss_mlp": 0.01020686,
      "balance_loss_clip": 1.03553474,
      "balance_loss_mlp": 1.01347947,
      "epoch": 0.6751638309384957,
      "flos": 23404313076480.0,
      "grad_norm": 1.9250115902088827,
      "language_loss": 0.78154719,
      "learning_rate": 1.0083316861169705e-06,
      "loss": 0.80269837,
      "num_input_tokens_seen": 120802090,
      "step": 5615,
      "time_per_iteration": 2.7658333778381348
    },
    {
      "auxiliary_loss_clip": 0.01135365,
      "auxiliary_loss_mlp": 0.01028758,
      "balance_loss_clip": 1.04158676,
      "balance_loss_mlp": 1.02096164,
      "epoch": 0.6752840738291348,
      "flos": 23441408847360.0,
      "grad_norm": 2.055620239732859,
      "language_loss": 0.71661985,
      "learning_rate": 1.0076552889499713e-06,
      "loss": 0.7382611,
      "num_input_tokens_seen": 120822855,
      "step": 5616,
      "time_per_iteration": 2.7836246490478516
    },
    {
      "auxiliary_loss_clip": 0.01158776,
      "auxiliary_loss_mlp": 0.01029412,
      "balance_loss_clip": 1.05043197,
      "balance_loss_mlp": 1.02183604,
      "epoch": 0.675404316719774,
      "flos": 30335826257280.0,
      "grad_norm": 2.0190496489226657,
      "language_loss": 0.73920786,
      "learning_rate": 1.006979042320876e-06,
      "loss": 0.7610898,
      "num_input_tokens_seen": 120843070,
      "step": 5617,
      "time_per_iteration": 2.713481903076172
    },
    {
      "auxiliary_loss_clip": 0.01136754,
      "auxiliary_loss_mlp": 0.0102295,
      "balance_loss_clip": 1.04049695,
      "balance_loss_mlp": 1.01558542,
      "epoch": 0.675524559610413,
      "flos": 23622613983360.0,
      "grad_norm": 2.328651265977191,
      "language_loss": 0.63237411,
      "learning_rate": 1.0063029463322702e-06,
      "loss": 0.6539712,
      "num_input_tokens_seen": 120863345,
      "step": 5618,
      "time_per_iteration": 2.6995837688446045
    },
    {
      "auxiliary_loss_clip": 0.01110937,
      "auxiliary_loss_mlp": 0.00762991,
      "balance_loss_clip": 1.0410403,
      "balance_loss_mlp": 1.00058484,
      "epoch": 0.6756448025010521,
      "flos": 21248631279360.0,
      "grad_norm": 2.3740472012629104,
      "language_loss": 0.75816512,
      "learning_rate": 1.0056270010867164e-06,
      "loss": 0.77690446,
      "num_input_tokens_seen": 120880915,
      "step": 5619,
      "time_per_iteration": 2.740149736404419
    },
    {
      "auxiliary_loss_clip": 0.0114697,
      "auxiliary_loss_mlp": 0.01024427,
      "balance_loss_clip": 1.04335666,
      "balance_loss_mlp": 1.01630926,
      "epoch": 0.6757650453916912,
      "flos": 21646521210240.0,
      "grad_norm": 2.6011823669195886,
      "language_loss": 0.78155828,
      "learning_rate": 1.004951206686758e-06,
      "loss": 0.80327225,
      "num_input_tokens_seen": 120899190,
      "step": 5620,
      "time_per_iteration": 2.673902750015259
    },
    {
      "auxiliary_loss_clip": 0.01155951,
      "auxiliary_loss_mlp": 0.01032167,
      "balance_loss_clip": 1.04764855,
      "balance_loss_mlp": 1.0249306,
      "epoch": 0.6758852882823303,
      "flos": 21795658479360.0,
      "grad_norm": 2.49675271649647,
      "language_loss": 0.71811426,
      "learning_rate": 1.0042755632349087e-06,
      "loss": 0.73999548,
      "num_input_tokens_seen": 120916080,
      "step": 5621,
      "time_per_iteration": 2.653812885284424
    },
    {
      "auxiliary_loss_clip": 0.01131986,
      "auxiliary_loss_mlp": 0.01030446,
      "balance_loss_clip": 1.04598165,
      "balance_loss_mlp": 1.02251828,
      "epoch": 0.6760055311729694,
      "flos": 27088783580160.0,
      "grad_norm": 2.2330127973169156,
      "language_loss": 0.62687755,
      "learning_rate": 1.0036000708336653e-06,
      "loss": 0.64850187,
      "num_input_tokens_seen": 120935210,
      "step": 5622,
      "time_per_iteration": 2.710482597351074
    },
    {
      "auxiliary_loss_clip": 0.01148303,
      "auxiliary_loss_mlp": 0.01025053,
      "balance_loss_clip": 1.04722953,
      "balance_loss_mlp": 1.01777864,
      "epoch": 0.6761257740636085,
      "flos": 17999792922240.0,
      "grad_norm": 3.943259077005317,
      "language_loss": 0.79268008,
      "learning_rate": 1.0029247295854984e-06,
      "loss": 0.81441367,
      "num_input_tokens_seen": 120951830,
      "step": 5623,
      "time_per_iteration": 3.613487482070923
    },
    {
      "auxiliary_loss_clip": 0.0113431,
      "auxiliary_loss_mlp": 0.01027593,
      "balance_loss_clip": 1.04503393,
      "balance_loss_mlp": 1.02085233,
      "epoch": 0.6762460169542476,
      "flos": 15121912273920.0,
      "grad_norm": 1.830363545364406,
      "language_loss": 0.71552223,
      "learning_rate": 1.0022495395928588e-06,
      "loss": 0.73714125,
      "num_input_tokens_seen": 120970310,
      "step": 5624,
      "time_per_iteration": 2.7659568786621094
    },
    {
      "auxiliary_loss_clip": 0.01072208,
      "auxiliary_loss_mlp": 0.01001471,
      "balance_loss_clip": 1.01518321,
      "balance_loss_mlp": 1.00026691,
      "epoch": 0.6763662598448866,
      "flos": 67886970030720.0,
      "grad_norm": 0.836156658565272,
      "language_loss": 0.62328517,
      "learning_rate": 1.0015745009581697e-06,
      "loss": 0.64402199,
      "num_input_tokens_seen": 121031915,
      "step": 5625,
      "time_per_iteration": 4.278688669204712
    },
    {
      "auxiliary_loss_clip": 0.01158074,
      "auxiliary_loss_mlp": 0.01025334,
      "balance_loss_clip": 1.04876542,
      "balance_loss_mlp": 1.01793969,
      "epoch": 0.6764865027355258,
      "flos": 20631829910400.0,
      "grad_norm": 2.802326638138946,
      "language_loss": 0.66845995,
      "learning_rate": 1.0008996137838343e-06,
      "loss": 0.69029403,
      "num_input_tokens_seen": 121050890,
      "step": 5626,
      "time_per_iteration": 3.5700061321258545
    },
    {
      "auxiliary_loss_clip": 0.01176387,
      "auxiliary_loss_mlp": 0.01025554,
      "balance_loss_clip": 1.04930329,
      "balance_loss_mlp": 1.01737905,
      "epoch": 0.6766067456261649,
      "flos": 21215809226880.0,
      "grad_norm": 1.9251392710950224,
      "language_loss": 0.79554623,
      "learning_rate": 1.000224878172234e-06,
      "loss": 0.81756568,
      "num_input_tokens_seen": 121070015,
      "step": 5627,
      "time_per_iteration": 2.637594699859619
    },
    {
      "auxiliary_loss_clip": 0.01160703,
      "auxiliary_loss_mlp": 0.0102381,
      "balance_loss_clip": 1.04735982,
      "balance_loss_mlp": 1.01661563,
      "epoch": 0.6767269885168039,
      "flos": 19938251220480.0,
      "grad_norm": 1.9654144533019624,
      "language_loss": 0.72909075,
      "learning_rate": 9.99550294225724e-07,
      "loss": 0.75093585,
      "num_input_tokens_seen": 121089170,
      "step": 5628,
      "time_per_iteration": 2.62931227684021
    },
    {
      "auxiliary_loss_clip": 0.01117666,
      "auxiliary_loss_mlp": 0.01025496,
      "balance_loss_clip": 1.04049051,
      "balance_loss_mlp": 1.01773608,
      "epoch": 0.6768472314074431,
      "flos": 20814076540800.0,
      "grad_norm": 2.137186621327736,
      "language_loss": 0.7285136,
      "learning_rate": 9.988758620466402e-07,
      "loss": 0.74994516,
      "num_input_tokens_seen": 121108040,
      "step": 5629,
      "time_per_iteration": 2.7638378143310547
    },
    {
      "auxiliary_loss_clip": 0.01110145,
      "auxiliary_loss_mlp": 0.01026942,
      "balance_loss_clip": 1.04242206,
      "balance_loss_mlp": 1.02050757,
      "epoch": 0.6769674742980821,
      "flos": 23186012169600.0,
      "grad_norm": 1.5324629585452674,
      "language_loss": 0.76249516,
      "learning_rate": 9.982015817372917e-07,
      "loss": 0.78386599,
      "num_input_tokens_seen": 121128480,
      "step": 5630,
      "time_per_iteration": 3.647925615310669
    },
    {
      "auxiliary_loss_clip": 0.01115857,
      "auxiliary_loss_mlp": 0.01026612,
      "balance_loss_clip": 1.04196262,
      "balance_loss_mlp": 1.01862526,
      "epoch": 0.6770877171887212,
      "flos": 24242934885120.0,
      "grad_norm": 2.25630976152184,
      "language_loss": 0.82313728,
      "learning_rate": 9.975274533999657e-07,
      "loss": 0.84456193,
      "num_input_tokens_seen": 121148010,
      "step": 5631,
      "time_per_iteration": 2.7790493965148926
    },
    {
      "auxiliary_loss_clip": 0.01175293,
      "auxiliary_loss_mlp": 0.01021043,
      "balance_loss_clip": 1.04929769,
      "balance_loss_mlp": 1.01350284,
      "epoch": 0.6772079600793603,
      "flos": 18141567903360.0,
      "grad_norm": 2.5255062704283384,
      "language_loss": 0.83988953,
      "learning_rate": 9.96853477136929e-07,
      "loss": 0.86185288,
      "num_input_tokens_seen": 121162755,
      "step": 5632,
      "time_per_iteration": 2.5044758319854736
    },
    {
      "auxiliary_loss_clip": 0.01121418,
      "auxiliary_loss_mlp": 0.01027861,
      "balance_loss_clip": 1.04045367,
      "balance_loss_mlp": 1.02011228,
      "epoch": 0.6773282029699994,
      "flos": 22452069571200.0,
      "grad_norm": 2.597752941017765,
      "language_loss": 0.75254214,
      "learning_rate": 9.96179653050422e-07,
      "loss": 0.77403498,
      "num_input_tokens_seen": 121182915,
      "step": 5633,
      "time_per_iteration": 2.7011470794677734
    },
    {
      "auxiliary_loss_clip": 0.01122076,
      "auxiliary_loss_mlp": 0.01023913,
      "balance_loss_clip": 1.04330277,
      "balance_loss_mlp": 1.01656973,
      "epoch": 0.6774484458606385,
      "flos": 18693730748160.0,
      "grad_norm": 2.069292776332559,
      "language_loss": 0.74108005,
      "learning_rate": 9.955059812426635e-07,
      "loss": 0.76253998,
      "num_input_tokens_seen": 121200445,
      "step": 5634,
      "time_per_iteration": 2.7119178771972656
    },
    {
      "auxiliary_loss_clip": 0.01175202,
      "auxiliary_loss_mlp": 0.01021383,
      "balance_loss_clip": 1.05220079,
      "balance_loss_mlp": 1.01373613,
      "epoch": 0.6775686887512776,
      "flos": 25994046821760.0,
      "grad_norm": 2.1806452243813754,
      "language_loss": 0.83088183,
      "learning_rate": 9.948324618158493e-07,
      "loss": 0.8528477,
      "num_input_tokens_seen": 121220785,
      "step": 5635,
      "time_per_iteration": 2.6580445766448975
    },
    {
      "auxiliary_loss_clip": 0.01159701,
      "auxiliary_loss_mlp": 0.01021827,
      "balance_loss_clip": 1.04570699,
      "balance_loss_mlp": 1.01421213,
      "epoch": 0.6776889316419167,
      "flos": 13587987922560.0,
      "grad_norm": 2.5433321248273475,
      "language_loss": 0.77866387,
      "learning_rate": 9.941590948721502e-07,
      "loss": 0.80047917,
      "num_input_tokens_seen": 121237985,
      "step": 5636,
      "time_per_iteration": 2.7068536281585693
    },
    {
      "auxiliary_loss_clip": 0.01139215,
      "auxiliary_loss_mlp": 0.01024649,
      "balance_loss_clip": 1.04508567,
      "balance_loss_mlp": 1.01782393,
      "epoch": 0.6778091745325557,
      "flos": 27601121220480.0,
      "grad_norm": 3.1535602705567243,
      "language_loss": 0.76572204,
      "learning_rate": 9.934858805137188e-07,
      "loss": 0.78736067,
      "num_input_tokens_seen": 121258635,
      "step": 5637,
      "time_per_iteration": 2.701462507247925
    },
    {
      "auxiliary_loss_clip": 0.01155338,
      "auxiliary_loss_mlp": 0.01024991,
      "balance_loss_clip": 1.04881716,
      "balance_loss_mlp": 1.01788664,
      "epoch": 0.6779294174231949,
      "flos": 18734058743040.0,
      "grad_norm": 1.6146172095275984,
      "language_loss": 0.80485415,
      "learning_rate": 9.92812818842677e-07,
      "loss": 0.82665741,
      "num_input_tokens_seen": 121277810,
      "step": 5638,
      "time_per_iteration": 2.6393420696258545
    },
    {
      "auxiliary_loss_clip": 0.01154541,
      "auxiliary_loss_mlp": 0.01023544,
      "balance_loss_clip": 1.04615104,
      "balance_loss_mlp": 1.01603127,
      "epoch": 0.678049660313834,
      "flos": 45873797765760.0,
      "grad_norm": 1.6808301606635099,
      "language_loss": 0.64113212,
      "learning_rate": 9.921399099611306e-07,
      "loss": 0.66291308,
      "num_input_tokens_seen": 121298975,
      "step": 5639,
      "time_per_iteration": 2.8180606365203857
    },
    {
      "auxiliary_loss_clip": 0.01146285,
      "auxiliary_loss_mlp": 0.0102666,
      "balance_loss_clip": 1.04446542,
      "balance_loss_mlp": 1.01953411,
      "epoch": 0.678169903204473,
      "flos": 19974556892160.0,
      "grad_norm": 1.5023220297590467,
      "language_loss": 0.6898362,
      "learning_rate": 9.914671539711588e-07,
      "loss": 0.71156561,
      "num_input_tokens_seen": 121318495,
      "step": 5640,
      "time_per_iteration": 2.6690948009490967
    },
    {
      "auxiliary_loss_clip": 0.01076271,
      "auxiliary_loss_mlp": 0.00763117,
      "balance_loss_clip": 1.04072905,
      "balance_loss_mlp": 1.0005827,
      "epoch": 0.6782901460951122,
      "flos": 21395613732480.0,
      "grad_norm": 2.516338416050967,
      "language_loss": 0.78301537,
      "learning_rate": 9.90794550974817e-07,
      "loss": 0.80140924,
      "num_input_tokens_seen": 121338890,
      "step": 5641,
      "time_per_iteration": 2.991572618484497
    },
    {
      "auxiliary_loss_clip": 0.01125203,
      "auxiliary_loss_mlp": 0.01028683,
      "balance_loss_clip": 1.04332697,
      "balance_loss_mlp": 1.02106583,
      "epoch": 0.6784103889857512,
      "flos": 21434002392960.0,
      "grad_norm": 2.1587111522706843,
      "language_loss": 0.81142557,
      "learning_rate": 9.901221010741407e-07,
      "loss": 0.83296442,
      "num_input_tokens_seen": 121358210,
      "step": 5642,
      "time_per_iteration": 3.0120208263397217
    },
    {
      "auxiliary_loss_clip": 0.01159589,
      "auxiliary_loss_mlp": 0.01026701,
      "balance_loss_clip": 1.04600012,
      "balance_loss_mlp": 1.01890516,
      "epoch": 0.6785306318763903,
      "flos": 32671923091200.0,
      "grad_norm": 1.8589368385098235,
      "language_loss": 0.74617702,
      "learning_rate": 9.894498043711375e-07,
      "loss": 0.76803994,
      "num_input_tokens_seen": 121379955,
      "step": 5643,
      "time_per_iteration": 2.710144281387329
    },
    {
      "auxiliary_loss_clip": 0.01139678,
      "auxiliary_loss_mlp": 0.01024011,
      "balance_loss_clip": 1.04272366,
      "balance_loss_mlp": 1.01705503,
      "epoch": 0.6786508747670293,
      "flos": 25632139340160.0,
      "grad_norm": 2.181876029897192,
      "language_loss": 0.69281673,
      "learning_rate": 9.887776609677962e-07,
      "loss": 0.71445364,
      "num_input_tokens_seen": 121401325,
      "step": 5644,
      "time_per_iteration": 2.70442271232605
    },
    {
      "auxiliary_loss_clip": 0.01117848,
      "auxiliary_loss_mlp": 0.01024846,
      "balance_loss_clip": 1.03898442,
      "balance_loss_mlp": 1.01788163,
      "epoch": 0.6787711176576685,
      "flos": 19171881619200.0,
      "grad_norm": 1.8204059257647656,
      "language_loss": 0.72461414,
      "learning_rate": 9.88105670966079e-07,
      "loss": 0.74604106,
      "num_input_tokens_seen": 121419785,
      "step": 5645,
      "time_per_iteration": 2.6737377643585205
    },
    {
      "auxiliary_loss_clip": 0.01103784,
      "auxiliary_loss_mlp": 0.01029451,
      "balance_loss_clip": 1.04238033,
      "balance_loss_mlp": 1.02170277,
      "epoch": 0.6788913605483076,
      "flos": 13985159581440.0,
      "grad_norm": 1.865303635923441,
      "language_loss": 0.78592896,
      "learning_rate": 9.874338344679283e-07,
      "loss": 0.80726129,
      "num_input_tokens_seen": 121435630,
      "step": 5646,
      "time_per_iteration": 2.763004779815674
    },
    {
      "auxiliary_loss_clip": 0.01168932,
      "auxiliary_loss_mlp": 0.01021636,
      "balance_loss_clip": 1.04797268,
      "balance_loss_mlp": 1.01463842,
      "epoch": 0.6790116034389466,
      "flos": 22017586659840.0,
      "grad_norm": 1.9825905582205374,
      "language_loss": 0.73949498,
      "learning_rate": 9.86762151575259e-07,
      "loss": 0.7614007,
      "num_input_tokens_seen": 121455625,
      "step": 5647,
      "time_per_iteration": 2.5971169471740723
    },
    {
      "auxiliary_loss_clip": 0.01117395,
      "auxiliary_loss_mlp": 0.00761542,
      "balance_loss_clip": 1.04415154,
      "balance_loss_mlp": 1.00056803,
      "epoch": 0.6791318463295858,
      "flos": 20922454851840.0,
      "grad_norm": 1.5489370430254157,
      "language_loss": 0.80035377,
      "learning_rate": 9.860906223899651e-07,
      "loss": 0.81914318,
      "num_input_tokens_seen": 121475020,
      "step": 5648,
      "time_per_iteration": 4.093108415603638
    },
    {
      "auxiliary_loss_clip": 0.01149502,
      "auxiliary_loss_mlp": 0.01023873,
      "balance_loss_clip": 1.04635417,
      "balance_loss_mlp": 1.01621127,
      "epoch": 0.6792520892202248,
      "flos": 28512749422080.0,
      "grad_norm": 1.7122207010124544,
      "language_loss": 0.75649381,
      "learning_rate": 9.854192470139184e-07,
      "loss": 0.77822757,
      "num_input_tokens_seen": 121496500,
      "step": 5649,
      "time_per_iteration": 2.7800652980804443
    },
    {
      "auxiliary_loss_clip": 0.01142862,
      "auxiliary_loss_mlp": 0.01029395,
      "balance_loss_clip": 1.04773569,
      "balance_loss_mlp": 1.02199864,
      "epoch": 0.6793723321108639,
      "flos": 20011904058240.0,
      "grad_norm": 2.274358264515876,
      "language_loss": 0.71743375,
      "learning_rate": 9.847480255489645e-07,
      "loss": 0.73915637,
      "num_input_tokens_seen": 121515525,
      "step": 5650,
      "time_per_iteration": 2.6828715801239014
    },
    {
      "auxiliary_loss_clip": 0.01149634,
      "auxiliary_loss_mlp": 0.0102404,
      "balance_loss_clip": 1.04641342,
      "balance_loss_mlp": 1.01610708,
      "epoch": 0.6794925750015031,
      "flos": 26649488246400.0,
      "grad_norm": 2.001881194683384,
      "language_loss": 0.69215226,
      "learning_rate": 9.840769580969295e-07,
      "loss": 0.713889,
      "num_input_tokens_seen": 121535965,
      "step": 5651,
      "time_per_iteration": 3.757110357284546
    },
    {
      "auxiliary_loss_clip": 0.01150815,
      "auxiliary_loss_mlp": 0.01026317,
      "balance_loss_clip": 1.04547119,
      "balance_loss_mlp": 1.01937556,
      "epoch": 0.6796128178921421,
      "flos": 21580374314880.0,
      "grad_norm": 1.7948272544713886,
      "language_loss": 0.79974496,
      "learning_rate": 9.834060447596114e-07,
      "loss": 0.82151628,
      "num_input_tokens_seen": 121555235,
      "step": 5652,
      "time_per_iteration": 3.6410605907440186
    },
    {
      "auxiliary_loss_clip": 0.01158784,
      "auxiliary_loss_mlp": 0.01026511,
      "balance_loss_clip": 1.04612184,
      "balance_loss_mlp": 1.01904249,
      "epoch": 0.6797330607827812,
      "flos": 22492002516480.0,
      "grad_norm": 1.936913262209491,
      "language_loss": 0.78392363,
      "learning_rate": 9.827352856387868e-07,
      "loss": 0.8057766,
      "num_input_tokens_seen": 121574945,
      "step": 5653,
      "time_per_iteration": 2.703279972076416
    },
    {
      "auxiliary_loss_clip": 0.0102152,
      "auxiliary_loss_mlp": 0.01001352,
      "balance_loss_clip": 1.01080024,
      "balance_loss_mlp": 1.00034463,
      "epoch": 0.6798533036734203,
      "flos": 66306648286080.0,
      "grad_norm": 0.776518237138747,
      "language_loss": 0.64263749,
      "learning_rate": 9.820646808362118e-07,
      "loss": 0.66286623,
      "num_input_tokens_seen": 121641200,
      "step": 5654,
      "time_per_iteration": 3.5675435066223145
    },
    {
      "auxiliary_loss_clip": 0.01142564,
      "auxiliary_loss_mlp": 0.01025125,
      "balance_loss_clip": 1.04781771,
      "balance_loss_mlp": 1.01753092,
      "epoch": 0.6799735465640594,
      "flos": 16180163792640.0,
      "grad_norm": 3.0512343888240934,
      "language_loss": 0.72713155,
      "learning_rate": 9.813942304536154e-07,
      "loss": 0.74880838,
      "num_input_tokens_seen": 121659170,
      "step": 5655,
      "time_per_iteration": 2.9382100105285645
    },
    {
      "auxiliary_loss_clip": 0.01145245,
      "auxiliary_loss_mlp": 0.01025823,
      "balance_loss_clip": 1.04557276,
      "balance_loss_mlp": 1.01865304,
      "epoch": 0.6800937894546984,
      "flos": 22125749489280.0,
      "grad_norm": 1.9382594220587697,
      "language_loss": 0.63420528,
      "learning_rate": 9.807239345927043e-07,
      "loss": 0.65591598,
      "num_input_tokens_seen": 121679180,
      "step": 5656,
      "time_per_iteration": 3.642383337020874
    },
    {
      "auxiliary_loss_clip": 0.01143145,
      "auxiliary_loss_mlp": 0.01023079,
      "balance_loss_clip": 1.04080808,
      "balance_loss_mlp": 1.0159359,
      "epoch": 0.6802140323453376,
      "flos": 31612953300480.0,
      "grad_norm": 2.3544822754541346,
      "language_loss": 0.71936029,
      "learning_rate": 9.80053793355162e-07,
      "loss": 0.74102253,
      "num_input_tokens_seen": 121697875,
      "step": 5657,
      "time_per_iteration": 2.70095157623291
    },
    {
      "auxiliary_loss_clip": 0.01113406,
      "auxiliary_loss_mlp": 0.01028638,
      "balance_loss_clip": 1.04283965,
      "balance_loss_mlp": 1.02161658,
      "epoch": 0.6803342752359767,
      "flos": 17712938908800.0,
      "grad_norm": 1.9147920958907063,
      "language_loss": 0.75011462,
      "learning_rate": 9.793838068426472e-07,
      "loss": 0.77153504,
      "num_input_tokens_seen": 121715570,
      "step": 5658,
      "time_per_iteration": 2.699640989303589
    },
    {
      "auxiliary_loss_clip": 0.01172915,
      "auxiliary_loss_mlp": 0.0102626,
      "balance_loss_clip": 1.05005503,
      "balance_loss_mlp": 1.01858628,
      "epoch": 0.6804545181266157,
      "flos": 11326800902400.0,
      "grad_norm": 2.6163517279601365,
      "language_loss": 0.61370015,
      "learning_rate": 9.78713975156799e-07,
      "loss": 0.63569188,
      "num_input_tokens_seen": 121731435,
      "step": 5659,
      "time_per_iteration": 2.5309886932373047
    },
    {
      "auxiliary_loss_clip": 0.01129764,
      "auxiliary_loss_mlp": 0.01035833,
      "balance_loss_clip": 1.04552627,
      "balance_loss_mlp": 1.02764344,
      "epoch": 0.6805747610172549,
      "flos": 29350976181120.0,
      "grad_norm": 2.1139411281544898,
      "language_loss": 0.71942472,
      "learning_rate": 9.780442983992273e-07,
      "loss": 0.74108064,
      "num_input_tokens_seen": 121749950,
      "step": 5660,
      "time_per_iteration": 2.7488083839416504
    },
    {
      "auxiliary_loss_clip": 0.01136992,
      "auxiliary_loss_mlp": 0.01025463,
      "balance_loss_clip": 1.04471433,
      "balance_loss_mlp": 1.01806593,
      "epoch": 0.680695003907894,
      "flos": 37631868612480.0,
      "grad_norm": 1.6756182778319337,
      "language_loss": 0.71810937,
      "learning_rate": 9.773747766715238e-07,
      "loss": 0.73973393,
      "num_input_tokens_seen": 121770770,
      "step": 5661,
      "time_per_iteration": 2.76985239982605
    },
    {
      "auxiliary_loss_clip": 0.01145286,
      "auxiliary_loss_mlp": 0.0102618,
      "balance_loss_clip": 1.04491341,
      "balance_loss_mlp": 1.01906002,
      "epoch": 0.680815246798533,
      "flos": 22127365601280.0,
      "grad_norm": 1.8661195166632119,
      "language_loss": 0.80098188,
      "learning_rate": 9.767054100752536e-07,
      "loss": 0.82269651,
      "num_input_tokens_seen": 121790720,
      "step": 5662,
      "time_per_iteration": 2.683082103729248
    },
    {
      "auxiliary_loss_clip": 0.01135598,
      "auxiliary_loss_mlp": 0.0102786,
      "balance_loss_clip": 1.04683471,
      "balance_loss_mlp": 1.01979518,
      "epoch": 0.6809354896891722,
      "flos": 17201822330880.0,
      "grad_norm": 2.6649592074581823,
      "language_loss": 0.82033217,
      "learning_rate": 9.760361987119584e-07,
      "loss": 0.84196675,
      "num_input_tokens_seen": 121808455,
      "step": 5663,
      "time_per_iteration": 2.6260786056518555
    },
    {
      "auxiliary_loss_clip": 0.01141779,
      "auxiliary_loss_mlp": 0.01028435,
      "balance_loss_clip": 1.04434037,
      "balance_loss_mlp": 1.02111018,
      "epoch": 0.6810557325798112,
      "flos": 12458166554880.0,
      "grad_norm": 1.9163305750440445,
      "language_loss": 0.67881477,
      "learning_rate": 9.753671426831592e-07,
      "loss": 0.70051694,
      "num_input_tokens_seen": 121824470,
      "step": 5664,
      "time_per_iteration": 2.6702826023101807
    },
    {
      "auxiliary_loss_clip": 0.01150164,
      "auxiliary_loss_mlp": 0.01022969,
      "balance_loss_clip": 1.04367602,
      "balance_loss_mlp": 1.01538467,
      "epoch": 0.6811759754704503,
      "flos": 22156165330560.0,
      "grad_norm": 1.7746016606696051,
      "language_loss": 0.79496086,
      "learning_rate": 9.746982420903483e-07,
      "loss": 0.81669223,
      "num_input_tokens_seen": 121842665,
      "step": 5665,
      "time_per_iteration": 2.6420562267303467
    },
    {
      "auxiliary_loss_clip": 0.01153332,
      "auxiliary_loss_mlp": 0.01022843,
      "balance_loss_clip": 1.04755592,
      "balance_loss_mlp": 1.01566625,
      "epoch": 0.6812962183610894,
      "flos": 17525377065600.0,
      "grad_norm": 1.5641216031806788,
      "language_loss": 0.74531984,
      "learning_rate": 9.740294970349993e-07,
      "loss": 0.76708156,
      "num_input_tokens_seen": 121859080,
      "step": 5666,
      "time_per_iteration": 2.596804141998291
    },
    {
      "auxiliary_loss_clip": 0.01051439,
      "auxiliary_loss_mlp": 0.01001088,
      "balance_loss_clip": 1.01521552,
      "balance_loss_mlp": 0.99990219,
      "epoch": 0.6814164612517285,
      "flos": 60274480855680.0,
      "grad_norm": 0.8791396548599791,
      "language_loss": 0.609375,
      "learning_rate": 9.733609076185594e-07,
      "loss": 0.62990028,
      "num_input_tokens_seen": 121915485,
      "step": 5667,
      "time_per_iteration": 3.191171407699585
    },
    {
      "auxiliary_loss_clip": 0.01158668,
      "auxiliary_loss_mlp": 0.01029567,
      "balance_loss_clip": 1.04920232,
      "balance_loss_mlp": 1.02225983,
      "epoch": 0.6815367041423676,
      "flos": 19317750750720.0,
      "grad_norm": 2.3178167145229156,
      "language_loss": 0.83698928,
      "learning_rate": 9.72692473942455e-07,
      "loss": 0.85887158,
      "num_input_tokens_seen": 121932710,
      "step": 5668,
      "time_per_iteration": 2.5936012268066406
    },
    {
      "auxiliary_loss_clip": 0.01118334,
      "auxiliary_loss_mlp": 0.01031653,
      "balance_loss_clip": 1.0444665,
      "balance_loss_mlp": 1.02386284,
      "epoch": 0.6816569470330067,
      "flos": 22161696024960.0,
      "grad_norm": 1.8809093624210158,
      "language_loss": 0.77643377,
      "learning_rate": 9.720241961080849e-07,
      "loss": 0.79793364,
      "num_input_tokens_seen": 121952025,
      "step": 5669,
      "time_per_iteration": 2.7268805503845215
    },
    {
      "auxiliary_loss_clip": 0.01170842,
      "auxiliary_loss_mlp": 0.01023632,
      "balance_loss_clip": 1.04630017,
      "balance_loss_mlp": 1.01635385,
      "epoch": 0.6817771899236458,
      "flos": 41463501137280.0,
      "grad_norm": 2.3465616045051694,
      "language_loss": 0.73154581,
      "learning_rate": 9.713560742168259e-07,
      "loss": 0.75349057,
      "num_input_tokens_seen": 121974650,
      "step": 5670,
      "time_per_iteration": 2.7638742923736572
    },
    {
      "auxiliary_loss_clip": 0.01127728,
      "auxiliary_loss_mlp": 0.01028179,
      "balance_loss_clip": 1.04249024,
      "balance_loss_mlp": 1.02112162,
      "epoch": 0.6818974328142848,
      "flos": 21106138026240.0,
      "grad_norm": 2.1085115005182966,
      "language_loss": 0.71552289,
      "learning_rate": 9.706881083700333e-07,
      "loss": 0.73708194,
      "num_input_tokens_seen": 121994335,
      "step": 5671,
      "time_per_iteration": 2.7430710792541504
    },
    {
      "auxiliary_loss_clip": 0.01105009,
      "auxiliary_loss_mlp": 0.01029421,
      "balance_loss_clip": 1.04692984,
      "balance_loss_mlp": 1.02201843,
      "epoch": 0.682017675704924,
      "flos": 20441897769600.0,
      "grad_norm": 2.0297460941932193,
      "language_loss": 0.82766551,
      "learning_rate": 9.700202986690357e-07,
      "loss": 0.84900981,
      "num_input_tokens_seen": 122012635,
      "step": 5672,
      "time_per_iteration": 2.736157178878784
    },
    {
      "auxiliary_loss_clip": 0.0115523,
      "auxiliary_loss_mlp": 0.00763187,
      "balance_loss_clip": 1.04613256,
      "balance_loss_mlp": 1.00050354,
      "epoch": 0.682137918595563,
      "flos": 20044438801920.0,
      "grad_norm": 2.012110426060409,
      "language_loss": 0.66537356,
      "learning_rate": 9.693526452151413e-07,
      "loss": 0.68455768,
      "num_input_tokens_seen": 122031685,
      "step": 5673,
      "time_per_iteration": 2.618607759475708
    },
    {
      "auxiliary_loss_clip": 0.01134199,
      "auxiliary_loss_mlp": 0.01029013,
      "balance_loss_clip": 1.04316175,
      "balance_loss_mlp": 1.02069521,
      "epoch": 0.6822581614862021,
      "flos": 31684559063040.0,
      "grad_norm": 1.72202880667935,
      "language_loss": 0.75748175,
      "learning_rate": 9.686851481096305e-07,
      "loss": 0.77911389,
      "num_input_tokens_seen": 122052995,
      "step": 5674,
      "time_per_iteration": 2.7543230056762695
    },
    {
      "auxiliary_loss_clip": 0.01097837,
      "auxiliary_loss_mlp": 0.01025086,
      "balance_loss_clip": 1.04049587,
      "balance_loss_mlp": 1.01799941,
      "epoch": 0.6823784043768413,
      "flos": 23477570864640.0,
      "grad_norm": 3.241500907025541,
      "language_loss": 0.71582299,
      "learning_rate": 9.68017807453762e-07,
      "loss": 0.7370522,
      "num_input_tokens_seen": 122071740,
      "step": 5675,
      "time_per_iteration": 3.686222791671753
    },
    {
      "auxiliary_loss_clip": 0.01145806,
      "auxiliary_loss_mlp": 0.00761925,
      "balance_loss_clip": 1.04733658,
      "balance_loss_mlp": 1.00053525,
      "epoch": 0.6824986472674803,
      "flos": 14137134024960.0,
      "grad_norm": 1.7721310378460424,
      "language_loss": 0.73100466,
      "learning_rate": 9.673506233487721e-07,
      "loss": 0.7500819,
      "num_input_tokens_seen": 122089705,
      "step": 5676,
      "time_per_iteration": 2.654435396194458
    },
    {
      "auxiliary_loss_clip": 0.01139828,
      "auxiliary_loss_mlp": 0.00761364,
      "balance_loss_clip": 1.04323721,
      "balance_loss_mlp": 1.00043535,
      "epoch": 0.6826188901581194,
      "flos": 21504997624320.0,
      "grad_norm": 1.7545153933056559,
      "language_loss": 0.86041021,
      "learning_rate": 9.666835958958717e-07,
      "loss": 0.87942207,
      "num_input_tokens_seen": 122109025,
      "step": 5677,
      "time_per_iteration": 3.641615867614746
    },
    {
      "auxiliary_loss_clip": 0.01172598,
      "auxiliary_loss_mlp": 0.01028933,
      "balance_loss_clip": 1.04999781,
      "balance_loss_mlp": 1.02171826,
      "epoch": 0.6827391330487584,
      "flos": 20810126044800.0,
      "grad_norm": 2.0538615358795123,
      "language_loss": 0.80001456,
      "learning_rate": 9.660167251962484e-07,
      "loss": 0.82202989,
      "num_input_tokens_seen": 122127385,
      "step": 5678,
      "time_per_iteration": 3.876284122467041
    },
    {
      "auxiliary_loss_clip": 0.0113001,
      "auxiliary_loss_mlp": 0.01026598,
      "balance_loss_clip": 1.04181337,
      "balance_loss_mlp": 1.01905179,
      "epoch": 0.6828593759393976,
      "flos": 21688788539520.0,
      "grad_norm": 3.4666103398676515,
      "language_loss": 0.77829558,
      "learning_rate": 9.653500113510654e-07,
      "loss": 0.79986167,
      "num_input_tokens_seen": 122146500,
      "step": 5679,
      "time_per_iteration": 2.8046910762786865
    },
    {
      "auxiliary_loss_clip": 0.01138774,
      "auxiliary_loss_mlp": 0.01026323,
      "balance_loss_clip": 1.04438424,
      "balance_loss_mlp": 1.01851463,
      "epoch": 0.6829796188300367,
      "flos": 25337707557120.0,
      "grad_norm": 2.4728975050452493,
      "language_loss": 0.66926688,
      "learning_rate": 9.646834544614627e-07,
      "loss": 0.69091785,
      "num_input_tokens_seen": 122167000,
      "step": 5680,
      "time_per_iteration": 2.743065118789673
    },
    {
      "auxiliary_loss_clip": 0.01137556,
      "auxiliary_loss_mlp": 0.01026141,
      "balance_loss_clip": 1.04600501,
      "balance_loss_mlp": 1.01911712,
      "epoch": 0.6830998617206757,
      "flos": 20704800389760.0,
      "grad_norm": 1.8890178556877153,
      "language_loss": 0.7643345,
      "learning_rate": 9.64017054628558e-07,
      "loss": 0.78597152,
      "num_input_tokens_seen": 122185825,
      "step": 5681,
      "time_per_iteration": 2.7170660495758057
    },
    {
      "auxiliary_loss_clip": 0.01116844,
      "auxiliary_loss_mlp": 0.01022845,
      "balance_loss_clip": 1.0408299,
      "balance_loss_mlp": 1.01560903,
      "epoch": 0.6832201046113149,
      "flos": 21726638496000.0,
      "grad_norm": 1.6219080112646582,
      "language_loss": 0.79226035,
      "learning_rate": 9.63350811953441e-07,
      "loss": 0.81365728,
      "num_input_tokens_seen": 122206200,
      "step": 5682,
      "time_per_iteration": 3.703184127807617
    },
    {
      "auxiliary_loss_clip": 0.01128588,
      "auxiliary_loss_mlp": 0.01022768,
      "balance_loss_clip": 1.04280543,
      "balance_loss_mlp": 1.01560998,
      "epoch": 0.6833403475019539,
      "flos": 19536554448000.0,
      "grad_norm": 2.0095693061307154,
      "language_loss": 0.7041325,
      "learning_rate": 9.626847265371826e-07,
      "loss": 0.72564602,
      "num_input_tokens_seen": 122225520,
      "step": 5683,
      "time_per_iteration": 2.7108614444732666
    },
    {
      "auxiliary_loss_clip": 0.01135439,
      "auxiliary_loss_mlp": 0.01029047,
      "balance_loss_clip": 1.04371369,
      "balance_loss_mlp": 1.0219841,
      "epoch": 0.683460590392593,
      "flos": 19352153001600.0,
      "grad_norm": 2.22697690436209,
      "language_loss": 0.78513336,
      "learning_rate": 9.620187984808262e-07,
      "loss": 0.80677819,
      "num_input_tokens_seen": 122244320,
      "step": 5684,
      "time_per_iteration": 2.6867051124572754
    },
    {
      "auxiliary_loss_clip": 0.01141947,
      "auxiliary_loss_mlp": 0.00762254,
      "balance_loss_clip": 1.04582334,
      "balance_loss_mlp": 1.00049114,
      "epoch": 0.6835808332832322,
      "flos": 23288500650240.0,
      "grad_norm": 1.5923121859707137,
      "language_loss": 0.85760677,
      "learning_rate": 9.613530278853919e-07,
      "loss": 0.87664872,
      "num_input_tokens_seen": 122264295,
      "step": 5685,
      "time_per_iteration": 2.6723363399505615
    },
    {
      "auxiliary_loss_clip": 0.01155158,
      "auxiliary_loss_mlp": 0.01025441,
      "balance_loss_clip": 1.04608166,
      "balance_loss_mlp": 1.01794815,
      "epoch": 0.6837010761738712,
      "flos": 21653416621440.0,
      "grad_norm": 2.128330854449357,
      "language_loss": 0.74369961,
      "learning_rate": 9.60687414851879e-07,
      "loss": 0.76550555,
      "num_input_tokens_seen": 122285300,
      "step": 5686,
      "time_per_iteration": 2.721127510070801
    },
    {
      "auxiliary_loss_clip": 0.01146618,
      "auxiliary_loss_mlp": 0.01020524,
      "balance_loss_clip": 1.04677749,
      "balance_loss_mlp": 1.01340723,
      "epoch": 0.6838213190645103,
      "flos": 17566387418880.0,
      "grad_norm": 2.4030622748177386,
      "language_loss": 0.77631301,
      "learning_rate": 9.600219594812575e-07,
      "loss": 0.79798442,
      "num_input_tokens_seen": 122303240,
      "step": 5687,
      "time_per_iteration": 2.612215757369995
    },
    {
      "auxiliary_loss_clip": 0.01167789,
      "auxiliary_loss_mlp": 0.01021945,
      "balance_loss_clip": 1.04639864,
      "balance_loss_mlp": 1.01520669,
      "epoch": 0.6839415619551494,
      "flos": 23112538899840.0,
      "grad_norm": 1.6659532567714697,
      "language_loss": 0.73219025,
      "learning_rate": 9.593566618744786e-07,
      "loss": 0.75408763,
      "num_input_tokens_seen": 122323390,
      "step": 5688,
      "time_per_iteration": 2.697328567504883
    },
    {
      "auxiliary_loss_clip": 0.0117033,
      "auxiliary_loss_mlp": 0.01026277,
      "balance_loss_clip": 1.04767549,
      "balance_loss_mlp": 1.0192827,
      "epoch": 0.6840618048457885,
      "flos": 22127868391680.0,
      "grad_norm": 1.8070888608928692,
      "language_loss": 0.74025744,
      "learning_rate": 9.58691522132466e-07,
      "loss": 0.76222354,
      "num_input_tokens_seen": 122342200,
      "step": 5689,
      "time_per_iteration": 2.731128692626953
    },
    {
      "auxiliary_loss_clip": 0.01147689,
      "auxiliary_loss_mlp": 0.01028782,
      "balance_loss_clip": 1.04698014,
      "balance_loss_mlp": 1.02074695,
      "epoch": 0.6841820477364275,
      "flos": 22015898720640.0,
      "grad_norm": 1.8983394835647847,
      "language_loss": 0.84542161,
      "learning_rate": 9.58026540356123e-07,
      "loss": 0.86718631,
      "num_input_tokens_seen": 122360465,
      "step": 5690,
      "time_per_iteration": 2.6994621753692627
    },
    {
      "auxiliary_loss_clip": 0.01159485,
      "auxiliary_loss_mlp": 0.01030826,
      "balance_loss_clip": 1.0470922,
      "balance_loss_mlp": 1.0232501,
      "epoch": 0.6843022906270667,
      "flos": 24900531125760.0,
      "grad_norm": 1.7522910108898768,
      "language_loss": 0.86624551,
      "learning_rate": 9.573617166463246e-07,
      "loss": 0.88814867,
      "num_input_tokens_seen": 122381680,
      "step": 5691,
      "time_per_iteration": 2.6156387329101562
    },
    {
      "auxiliary_loss_clip": 0.01143867,
      "auxiliary_loss_mlp": 0.01022727,
      "balance_loss_clip": 1.04390967,
      "balance_loss_mlp": 1.015589,
      "epoch": 0.6844225335177058,
      "flos": 19969924037760.0,
      "grad_norm": 1.9802643085676674,
      "language_loss": 0.60204363,
      "learning_rate": 9.56697051103924e-07,
      "loss": 0.62370956,
      "num_input_tokens_seen": 122399120,
      "step": 5692,
      "time_per_iteration": 2.709254741668701
    },
    {
      "auxiliary_loss_clip": 0.01140398,
      "auxiliary_loss_mlp": 0.01023931,
      "balance_loss_clip": 1.04520822,
      "balance_loss_mlp": 1.0170083,
      "epoch": 0.6845427764083448,
      "flos": 25883334126720.0,
      "grad_norm": 2.6402464139957273,
      "language_loss": 0.81170344,
      "learning_rate": 9.560325438297522e-07,
      "loss": 0.83334672,
      "num_input_tokens_seen": 122417430,
      "step": 5693,
      "time_per_iteration": 2.727132797241211
    },
    {
      "auxiliary_loss_clip": 0.01145328,
      "auxiliary_loss_mlp": 0.01029598,
      "balance_loss_clip": 1.04922044,
      "balance_loss_mlp": 1.02237093,
      "epoch": 0.684663019298984,
      "flos": 18880143356160.0,
      "grad_norm": 1.793128764344129,
      "language_loss": 0.8682822,
      "learning_rate": 9.553681949246127e-07,
      "loss": 0.89003146,
      "num_input_tokens_seen": 122435055,
      "step": 5694,
      "time_per_iteration": 2.6896986961364746
    },
    {
      "auxiliary_loss_clip": 0.01133121,
      "auxiliary_loss_mlp": 0.01032732,
      "balance_loss_clip": 1.04412532,
      "balance_loss_mlp": 1.02498317,
      "epoch": 0.684783262189623,
      "flos": 54193725302400.0,
      "grad_norm": 3.1243937621601368,
      "language_loss": 0.7535789,
      "learning_rate": 9.547040044892886e-07,
      "loss": 0.77523744,
      "num_input_tokens_seen": 122462570,
      "step": 5695,
      "time_per_iteration": 2.9682672023773193
    },
    {
      "auxiliary_loss_clip": 0.01064069,
      "auxiliary_loss_mlp": 0.01001487,
      "balance_loss_clip": 1.01525855,
      "balance_loss_mlp": 1.00038993,
      "epoch": 0.6849035050802621,
      "flos": 63970264143360.0,
      "grad_norm": 0.889399092057962,
      "language_loss": 0.60026091,
      "learning_rate": 9.540399726245354e-07,
      "loss": 0.62091649,
      "num_input_tokens_seen": 122519275,
      "step": 5696,
      "time_per_iteration": 3.134632110595703
    },
    {
      "auxiliary_loss_clip": 0.01138132,
      "auxiliary_loss_mlp": 0.01033022,
      "balance_loss_clip": 1.04255843,
      "balance_loss_mlp": 1.02486205,
      "epoch": 0.6850237479709013,
      "flos": 25224121774080.0,
      "grad_norm": 1.715351670355798,
      "language_loss": 0.69027495,
      "learning_rate": 9.533760994310859e-07,
      "loss": 0.71198654,
      "num_input_tokens_seen": 122539675,
      "step": 5697,
      "time_per_iteration": 2.7083899974823
    },
    {
      "auxiliary_loss_clip": 0.01173102,
      "auxiliary_loss_mlp": 0.01027902,
      "balance_loss_clip": 1.0499891,
      "balance_loss_mlp": 1.02046001,
      "epoch": 0.6851439908615403,
      "flos": 19354128249600.0,
      "grad_norm": 2.0593683469824144,
      "language_loss": 0.75367975,
      "learning_rate": 9.527123850096508e-07,
      "loss": 0.77568984,
      "num_input_tokens_seen": 122558035,
      "step": 5698,
      "time_per_iteration": 2.596379280090332
    },
    {
      "auxiliary_loss_clip": 0.01160713,
      "auxiliary_loss_mlp": 0.01025849,
      "balance_loss_clip": 1.04713178,
      "balance_loss_mlp": 1.01818347,
      "epoch": 0.6852642337521794,
      "flos": 23182133500800.0,
      "grad_norm": 1.9454362585602496,
      "language_loss": 0.72072613,
      "learning_rate": 9.520488294609142e-07,
      "loss": 0.74259174,
      "num_input_tokens_seen": 122576815,
      "step": 5699,
      "time_per_iteration": 2.6695423126220703
    },
    {
      "auxiliary_loss_clip": 0.01028831,
      "auxiliary_loss_mlp": 0.01000353,
      "balance_loss_clip": 1.01564789,
      "balance_loss_mlp": 0.99925613,
      "epoch": 0.6853844766428185,
      "flos": 62647206583680.0,
      "grad_norm": 0.7460187283212975,
      "language_loss": 0.53850186,
      "learning_rate": 9.513854328855368e-07,
      "loss": 0.55879366,
      "num_input_tokens_seen": 122634690,
      "step": 5700,
      "time_per_iteration": 4.187113285064697
    },
    {
      "auxiliary_loss_clip": 0.01166535,
      "auxiliary_loss_mlp": 0.01024922,
      "balance_loss_clip": 1.04606843,
      "balance_loss_mlp": 1.01811206,
      "epoch": 0.6855047195334576,
      "flos": 23437242869760.0,
      "grad_norm": 2.240416034551167,
      "language_loss": 0.81059515,
      "learning_rate": 9.507221953841558e-07,
      "loss": 0.8325097,
      "num_input_tokens_seen": 122652320,
      "step": 5701,
      "time_per_iteration": 2.6337695121765137
    },
    {
      "auxiliary_loss_clip": 0.01158838,
      "auxiliary_loss_mlp": 0.0102879,
      "balance_loss_clip": 1.04937375,
      "balance_loss_mlp": 1.02112198,
      "epoch": 0.6856249624240967,
      "flos": 20664831530880.0,
      "grad_norm": 1.5458304651233254,
      "language_loss": 0.77718824,
      "learning_rate": 9.500591170573824e-07,
      "loss": 0.79906452,
      "num_input_tokens_seen": 122672340,
      "step": 5702,
      "time_per_iteration": 2.6468875408172607
    },
    {
      "auxiliary_loss_clip": 0.01110074,
      "auxiliary_loss_mlp": 0.01031455,
      "balance_loss_clip": 1.04255259,
      "balance_loss_mlp": 1.02405798,
      "epoch": 0.6857452053147358,
      "flos": 17087302794240.0,
      "grad_norm": 1.7882005014539861,
      "language_loss": 0.74121797,
      "learning_rate": 9.493961980058078e-07,
      "loss": 0.76263326,
      "num_input_tokens_seen": 122689935,
      "step": 5703,
      "time_per_iteration": 3.720088481903076
    },
    {
      "auxiliary_loss_clip": 0.01084907,
      "auxiliary_loss_mlp": 0.01029389,
      "balance_loss_clip": 1.03781486,
      "balance_loss_mlp": 1.02223933,
      "epoch": 0.6858654482053749,
      "flos": 30847266057600.0,
      "grad_norm": 2.006910432939039,
      "language_loss": 0.67408401,
      "learning_rate": 9.48733438329993e-07,
      "loss": 0.69522697,
      "num_input_tokens_seen": 122710200,
      "step": 5704,
      "time_per_iteration": 3.7168586254119873
    },
    {
      "auxiliary_loss_clip": 0.01165883,
      "auxiliary_loss_mlp": 0.00761941,
      "balance_loss_clip": 1.04607677,
      "balance_loss_mlp": 1.00047445,
      "epoch": 0.6859856910960139,
      "flos": 28877314510080.0,
      "grad_norm": 1.8309177183552732,
      "language_loss": 0.74290049,
      "learning_rate": 9.480708381304807e-07,
      "loss": 0.76217866,
      "num_input_tokens_seen": 122731495,
      "step": 5705,
      "time_per_iteration": 2.716273546218872
    },
    {
      "auxiliary_loss_clip": 0.01111941,
      "auxiliary_loss_mlp": 0.01028979,
      "balance_loss_clip": 1.04432106,
      "balance_loss_mlp": 1.02158821,
      "epoch": 0.6861059339866531,
      "flos": 19354523299200.0,
      "grad_norm": 2.080948890723887,
      "language_loss": 0.83523762,
      "learning_rate": 9.474083975077858e-07,
      "loss": 0.85664678,
      "num_input_tokens_seen": 122748620,
      "step": 5706,
      "time_per_iteration": 2.758047103881836
    },
    {
      "auxiliary_loss_clip": 0.01148495,
      "auxiliary_loss_mlp": 0.01027597,
      "balance_loss_clip": 1.04325771,
      "balance_loss_mlp": 1.01974118,
      "epoch": 0.6862261768772921,
      "flos": 22199976944640.0,
      "grad_norm": 2.4798279225019964,
      "language_loss": 0.79818344,
      "learning_rate": 9.467461165623994e-07,
      "loss": 0.81994438,
      "num_input_tokens_seen": 122767670,
      "step": 5707,
      "time_per_iteration": 2.684525966644287
    },
    {
      "auxiliary_loss_clip": 0.01157608,
      "auxiliary_loss_mlp": 0.01022942,
      "balance_loss_clip": 1.04471564,
      "balance_loss_mlp": 1.01603365,
      "epoch": 0.6863464197679312,
      "flos": 26285677344000.0,
      "grad_norm": 1.9734357954421704,
      "language_loss": 0.79620779,
      "learning_rate": 9.46083995394791e-07,
      "loss": 0.81801331,
      "num_input_tokens_seen": 122785480,
      "step": 5708,
      "time_per_iteration": 3.5906710624694824
    },
    {
      "auxiliary_loss_clip": 0.01155023,
      "auxiliary_loss_mlp": 0.00762429,
      "balance_loss_clip": 1.04584002,
      "balance_loss_mlp": 1.00050402,
      "epoch": 0.6864666626585703,
      "flos": 37815228564480.0,
      "grad_norm": 1.861802094353502,
      "language_loss": 0.63544214,
      "learning_rate": 9.454220341054012e-07,
      "loss": 0.65461659,
      "num_input_tokens_seen": 122810265,
      "step": 5709,
      "time_per_iteration": 2.801804304122925
    },
    {
      "auxiliary_loss_clip": 0.0112688,
      "auxiliary_loss_mlp": 0.01020225,
      "balance_loss_clip": 1.04417765,
      "balance_loss_mlp": 1.01297688,
      "epoch": 0.6865869055492094,
      "flos": 19391152193280.0,
      "grad_norm": 2.1575769234007667,
      "language_loss": 0.808079,
      "learning_rate": 9.447602327946512e-07,
      "loss": 0.82955009,
      "num_input_tokens_seen": 122828905,
      "step": 5710,
      "time_per_iteration": 2.7218356132507324
    },
    {
      "auxiliary_loss_clip": 0.01139465,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.04249001,
      "balance_loss_mlp": 1.02236855,
      "epoch": 0.6867071484398485,
      "flos": 20375966355840.0,
      "grad_norm": 2.0236449765862985,
      "language_loss": 0.76173794,
      "learning_rate": 9.440985915629338e-07,
      "loss": 0.78342831,
      "num_input_tokens_seen": 122846235,
      "step": 5711,
      "time_per_iteration": 2.75539231300354
    },
    {
      "auxiliary_loss_clip": 0.01170466,
      "auxiliary_loss_mlp": 0.01029888,
      "balance_loss_clip": 1.05113208,
      "balance_loss_mlp": 1.02255082,
      "epoch": 0.6868273913304875,
      "flos": 15889143801600.0,
      "grad_norm": 1.984058464985516,
      "language_loss": 0.72873223,
      "learning_rate": 9.434371105106223e-07,
      "loss": 0.75073576,
      "num_input_tokens_seen": 122863835,
      "step": 5712,
      "time_per_iteration": 2.582150936126709
    },
    {
      "auxiliary_loss_clip": 0.01125513,
      "auxiliary_loss_mlp": 0.0102871,
      "balance_loss_clip": 1.04236531,
      "balance_loss_mlp": 1.02134025,
      "epoch": 0.6869476342211267,
      "flos": 24462492768000.0,
      "grad_norm": 1.8414592482800163,
      "language_loss": 0.70862424,
      "learning_rate": 9.427757897380602e-07,
      "loss": 0.73016655,
      "num_input_tokens_seen": 122883235,
      "step": 5713,
      "time_per_iteration": 2.703075647354126
    },
    {
      "auxiliary_loss_clip": 0.01124297,
      "auxiliary_loss_mlp": 0.01028468,
      "balance_loss_clip": 1.04399896,
      "balance_loss_mlp": 1.02109444,
      "epoch": 0.6870678771117658,
      "flos": 18442571875200.0,
      "grad_norm": 2.324107640267756,
      "language_loss": 0.85204375,
      "learning_rate": 9.421146293455695e-07,
      "loss": 0.8735714,
      "num_input_tokens_seen": 122898975,
      "step": 5714,
      "time_per_iteration": 2.6754708290100098
    },
    {
      "auxiliary_loss_clip": 0.0113927,
      "auxiliary_loss_mlp": 0.01026785,
      "balance_loss_clip": 1.04270577,
      "balance_loss_mlp": 1.01930773,
      "epoch": 0.6871881200024048,
      "flos": 22200371994240.0,
      "grad_norm": 2.2832674638598793,
      "language_loss": 0.68466806,
      "learning_rate": 9.414536294334489e-07,
      "loss": 0.70632863,
      "num_input_tokens_seen": 122918995,
      "step": 5715,
      "time_per_iteration": 2.6638376712799072
    },
    {
      "auxiliary_loss_clip": 0.01142655,
      "auxiliary_loss_mlp": 0.01021432,
      "balance_loss_clip": 1.04214966,
      "balance_loss_mlp": 1.0142529,
      "epoch": 0.687308362893044,
      "flos": 22127724737280.0,
      "grad_norm": 3.056115292540584,
      "language_loss": 0.6950435,
      "learning_rate": 9.407927901019708e-07,
      "loss": 0.7166844,
      "num_input_tokens_seen": 122938125,
      "step": 5716,
      "time_per_iteration": 2.6696979999542236
    },
    {
      "auxiliary_loss_clip": 0.01156576,
      "auxiliary_loss_mlp": 0.0101916,
      "balance_loss_clip": 1.04554141,
      "balance_loss_mlp": 1.01202512,
      "epoch": 0.687428605783683,
      "flos": 25040546340480.0,
      "grad_norm": 1.9491692070173496,
      "language_loss": 0.7668159,
      "learning_rate": 9.401321114513854e-07,
      "loss": 0.78857321,
      "num_input_tokens_seen": 122957020,
      "step": 5717,
      "time_per_iteration": 2.7206945419311523
    },
    {
      "auxiliary_loss_clip": 0.01170404,
      "auxiliary_loss_mlp": 0.01022601,
      "balance_loss_clip": 1.04830825,
      "balance_loss_mlp": 1.01526093,
      "epoch": 0.6875488486743221,
      "flos": 23770063313280.0,
      "grad_norm": 1.5735856314233618,
      "language_loss": 0.75703877,
      "learning_rate": 9.394715935819155e-07,
      "loss": 0.77896881,
      "num_input_tokens_seen": 122977410,
      "step": 5718,
      "time_per_iteration": 2.694150924682617
    },
    {
      "auxiliary_loss_clip": 0.01158226,
      "auxiliary_loss_mlp": 0.01029792,
      "balance_loss_clip": 1.04577506,
      "balance_loss_mlp": 1.02249658,
      "epoch": 0.6876690915649613,
      "flos": 25516937445120.0,
      "grad_norm": 2.208763144324675,
      "language_loss": 0.6308552,
      "learning_rate": 9.388112365937608e-07,
      "loss": 0.65273535,
      "num_input_tokens_seen": 122996875,
      "step": 5719,
      "time_per_iteration": 2.650710344314575
    },
    {
      "auxiliary_loss_clip": 0.01127517,
      "auxiliary_loss_mlp": 0.01029474,
      "balance_loss_clip": 1.04376447,
      "balance_loss_mlp": 1.02203536,
      "epoch": 0.6877893344556003,
      "flos": 19427996568960.0,
      "grad_norm": 2.102158939882556,
      "language_loss": 0.82977903,
      "learning_rate": 9.381510405870985e-07,
      "loss": 0.851349,
      "num_input_tokens_seen": 123015890,
      "step": 5720,
      "time_per_iteration": 2.6253960132598877
    },
    {
      "auxiliary_loss_clip": 0.01152266,
      "auxiliary_loss_mlp": 0.01026504,
      "balance_loss_clip": 1.04386973,
      "balance_loss_mlp": 1.01880336,
      "epoch": 0.6879095773462394,
      "flos": 18661303745280.0,
      "grad_norm": 2.210083621518969,
      "language_loss": 0.77206844,
      "learning_rate": 9.374910056620791e-07,
      "loss": 0.79385608,
      "num_input_tokens_seen": 123034955,
      "step": 5721,
      "time_per_iteration": 2.596195936203003
    },
    {
      "auxiliary_loss_clip": 0.0115964,
      "auxiliary_loss_mlp": 0.01026307,
      "balance_loss_clip": 1.04889154,
      "balance_loss_mlp": 1.01908612,
      "epoch": 0.6880298202368785,
      "flos": 20883132437760.0,
      "grad_norm": 1.668424449285381,
      "language_loss": 0.81079853,
      "learning_rate": 9.368311319188293e-07,
      "loss": 0.83265799,
      "num_input_tokens_seen": 123052770,
      "step": 5722,
      "time_per_iteration": 2.622199773788452
    },
    {
      "auxiliary_loss_clip": 0.01125939,
      "auxiliary_loss_mlp": 0.01029932,
      "balance_loss_clip": 1.04082859,
      "balance_loss_mlp": 1.02249956,
      "epoch": 0.6881500631275176,
      "flos": 30153292318080.0,
      "grad_norm": 1.7376272257595549,
      "language_loss": 0.79443508,
      "learning_rate": 9.361714194574515e-07,
      "loss": 0.81599379,
      "num_input_tokens_seen": 123075105,
      "step": 5723,
      "time_per_iteration": 2.830580234527588
    },
    {
      "auxiliary_loss_clip": 0.01071431,
      "auxiliary_loss_mlp": 0.01000864,
      "balance_loss_clip": 1.01435232,
      "balance_loss_mlp": 0.99970782,
      "epoch": 0.6882703060181566,
      "flos": 66181537215360.0,
      "grad_norm": 0.7325801815379657,
      "language_loss": 0.58247733,
      "learning_rate": 9.355118683780228e-07,
      "loss": 0.60320026,
      "num_input_tokens_seen": 123145175,
      "step": 5724,
      "time_per_iteration": 3.356541156768799
    },
    {
      "auxiliary_loss_clip": 0.01167453,
      "auxiliary_loss_mlp": 0.01025868,
      "balance_loss_clip": 1.04545915,
      "balance_loss_mlp": 1.01847148,
      "epoch": 0.6883905489087958,
      "flos": 18214646123520.0,
      "grad_norm": 2.069491869916693,
      "language_loss": 0.79184371,
      "learning_rate": 9.348524787805987e-07,
      "loss": 0.81377697,
      "num_input_tokens_seen": 123160365,
      "step": 5725,
      "time_per_iteration": 2.595020055770874
    },
    {
      "auxiliary_loss_clip": 0.01127641,
      "auxiliary_loss_mlp": 0.01024662,
      "balance_loss_clip": 1.03955984,
      "balance_loss_mlp": 1.01712847,
      "epoch": 0.6885107917994349,
      "flos": 14056262553600.0,
      "grad_norm": 2.756202906673922,
      "language_loss": 0.85753345,
      "learning_rate": 9.341932507652053e-07,
      "loss": 0.87905645,
      "num_input_tokens_seen": 123174855,
      "step": 5726,
      "time_per_iteration": 3.6474852561950684
    },
    {
      "auxiliary_loss_clip": 0.01138934,
      "auxiliary_loss_mlp": 0.01028919,
      "balance_loss_clip": 1.04128242,
      "balance_loss_mlp": 1.02103889,
      "epoch": 0.6886310346900739,
      "flos": 28690722334080.0,
      "grad_norm": 1.9642043354668144,
      "language_loss": 0.78523219,
      "learning_rate": 9.335341844318489e-07,
      "loss": 0.80691063,
      "num_input_tokens_seen": 123194995,
      "step": 5727,
      "time_per_iteration": 2.6935832500457764
    },
    {
      "auxiliary_loss_clip": 0.01139595,
      "auxiliary_loss_mlp": 0.01025846,
      "balance_loss_clip": 1.04453015,
      "balance_loss_mlp": 1.01855016,
      "epoch": 0.6887512775807131,
      "flos": 24535319592960.0,
      "grad_norm": 1.768022677084799,
      "language_loss": 0.73250121,
      "learning_rate": 9.328752798805091e-07,
      "loss": 0.75415564,
      "num_input_tokens_seen": 123213465,
      "step": 5728,
      "time_per_iteration": 3.6679186820983887
    },
    {
      "auxiliary_loss_clip": 0.01155508,
      "auxiliary_loss_mlp": 0.01027645,
      "balance_loss_clip": 1.04750586,
      "balance_loss_mlp": 1.01994956,
      "epoch": 0.6888715204713521,
      "flos": 22414363269120.0,
      "grad_norm": 2.086781438727497,
      "language_loss": 0.76339877,
      "learning_rate": 9.322165372111399e-07,
      "loss": 0.78523028,
      "num_input_tokens_seen": 123231610,
      "step": 5729,
      "time_per_iteration": 2.6472225189208984
    },
    {
      "auxiliary_loss_clip": 0.01120042,
      "auxiliary_loss_mlp": 0.010234,
      "balance_loss_clip": 1.04254651,
      "balance_loss_mlp": 1.01642346,
      "epoch": 0.6889917633619912,
      "flos": 22054323294720.0,
      "grad_norm": 2.0533270763093614,
      "language_loss": 0.75606412,
      "learning_rate": 9.315579565236747e-07,
      "loss": 0.7774986,
      "num_input_tokens_seen": 123250715,
      "step": 5730,
      "time_per_iteration": 3.6397438049316406
    },
    {
      "auxiliary_loss_clip": 0.01138428,
      "auxiliary_loss_mlp": 0.01026454,
      "balance_loss_clip": 1.04581404,
      "balance_loss_mlp": 1.0193876,
      "epoch": 0.6891120062526304,
      "flos": 23949724164480.0,
      "grad_norm": 1.7299420356839093,
      "language_loss": 0.74038839,
      "learning_rate": 9.308995379180162e-07,
      "loss": 0.76203722,
      "num_input_tokens_seen": 123270270,
      "step": 5731,
      "time_per_iteration": 2.7352676391601562
    },
    {
      "auxiliary_loss_clip": 0.01061763,
      "auxiliary_loss_mlp": 0.01000649,
      "balance_loss_clip": 1.01373482,
      "balance_loss_mlp": 0.99948663,
      "epoch": 0.6892322491432694,
      "flos": 64117354337280.0,
      "grad_norm": 0.7886640810767812,
      "language_loss": 0.59470785,
      "learning_rate": 9.302412814940488e-07,
      "loss": 0.61533189,
      "num_input_tokens_seen": 123333045,
      "step": 5732,
      "time_per_iteration": 3.2564289569854736
    },
    {
      "auxiliary_loss_clip": 0.01138893,
      "auxiliary_loss_mlp": 0.01030098,
      "balance_loss_clip": 1.04236758,
      "balance_loss_mlp": 1.02175283,
      "epoch": 0.6893524920339085,
      "flos": 23002436736000.0,
      "grad_norm": 4.099707294444931,
      "language_loss": 0.7166239,
      "learning_rate": 9.295831873516276e-07,
      "loss": 0.73831379,
      "num_input_tokens_seen": 123352320,
      "step": 5733,
      "time_per_iteration": 2.7194089889526367
    },
    {
      "auxiliary_loss_clip": 0.01168617,
      "auxiliary_loss_mlp": 0.01024131,
      "balance_loss_clip": 1.04778671,
      "balance_loss_mlp": 1.0172646,
      "epoch": 0.6894727349245476,
      "flos": 21396260177280.0,
      "grad_norm": 1.711387316726599,
      "language_loss": 0.76024401,
      "learning_rate": 9.289252555905873e-07,
      "loss": 0.78217149,
      "num_input_tokens_seen": 123372400,
      "step": 5734,
      "time_per_iteration": 3.679340362548828
    },
    {
      "auxiliary_loss_clip": 0.01158781,
      "auxiliary_loss_mlp": 0.01025338,
      "balance_loss_clip": 1.04971039,
      "balance_loss_mlp": 1.0177052,
      "epoch": 0.6895929778151867,
      "flos": 19865316654720.0,
      "grad_norm": 1.9719576260980314,
      "language_loss": 0.75710112,
      "learning_rate": 9.282674863107334e-07,
      "loss": 0.77894235,
      "num_input_tokens_seen": 123390215,
      "step": 5735,
      "time_per_iteration": 2.7184972763061523
    },
    {
      "auxiliary_loss_clip": 0.0115388,
      "auxiliary_loss_mlp": 0.0103066,
      "balance_loss_clip": 1.04709649,
      "balance_loss_mlp": 1.02316809,
      "epoch": 0.6897132207058257,
      "flos": 18179166464640.0,
      "grad_norm": 2.3308694341364418,
      "language_loss": 0.75719738,
      "learning_rate": 9.276098796118488e-07,
      "loss": 0.77904284,
      "num_input_tokens_seen": 123406700,
      "step": 5736,
      "time_per_iteration": 2.645996570587158
    },
    {
      "auxiliary_loss_clip": 0.01138836,
      "auxiliary_loss_mlp": 0.01027557,
      "balance_loss_clip": 1.044559,
      "balance_loss_mlp": 1.020908,
      "epoch": 0.6898334635964649,
      "flos": 32561641359360.0,
      "grad_norm": 1.9094504487960677,
      "language_loss": 0.66270196,
      "learning_rate": 9.269524355936938e-07,
      "loss": 0.68436587,
      "num_input_tokens_seen": 123429880,
      "step": 5737,
      "time_per_iteration": 2.8442065715789795
    },
    {
      "auxiliary_loss_clip": 0.01133201,
      "auxiliary_loss_mlp": 0.01024225,
      "balance_loss_clip": 1.04010749,
      "balance_loss_mlp": 1.01682222,
      "epoch": 0.689953706487104,
      "flos": 22819004956800.0,
      "grad_norm": 1.736307379372792,
      "language_loss": 0.85124457,
      "learning_rate": 9.262951543560002e-07,
      "loss": 0.87281883,
      "num_input_tokens_seen": 123449105,
      "step": 5738,
      "time_per_iteration": 2.743579626083374
    },
    {
      "auxiliary_loss_clip": 0.01140351,
      "auxiliary_loss_mlp": 0.01026328,
      "balance_loss_clip": 1.047979,
      "balance_loss_mlp": 1.01838219,
      "epoch": 0.690073949377743,
      "flos": 18515362786560.0,
      "grad_norm": 2.237044017953183,
      "language_loss": 0.8631804,
      "learning_rate": 9.256380359984795e-07,
      "loss": 0.88484716,
      "num_input_tokens_seen": 123466215,
      "step": 5739,
      "time_per_iteration": 2.6478874683380127
    },
    {
      "auxiliary_loss_clip": 0.01118548,
      "auxiliary_loss_mlp": 0.01023523,
      "balance_loss_clip": 1.03781796,
      "balance_loss_mlp": 1.01566696,
      "epoch": 0.6901941922683821,
      "flos": 34857194716800.0,
      "grad_norm": 2.0304094891933424,
      "language_loss": 0.75003648,
      "learning_rate": 9.249810806208139e-07,
      "loss": 0.7714572,
      "num_input_tokens_seen": 123485480,
      "step": 5740,
      "time_per_iteration": 2.858214855194092
    },
    {
      "auxiliary_loss_clip": 0.01108194,
      "auxiliary_loss_mlp": 0.00762072,
      "balance_loss_clip": 1.03632069,
      "balance_loss_mlp": 1.00045776,
      "epoch": 0.6903144351590212,
      "flos": 16253672976000.0,
      "grad_norm": 1.933036296954168,
      "language_loss": 0.80328119,
      "learning_rate": 9.243242883226627e-07,
      "loss": 0.82198387,
      "num_input_tokens_seen": 123504575,
      "step": 5741,
      "time_per_iteration": 2.7187411785125732
    },
    {
      "auxiliary_loss_clip": 0.01157477,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 1.04255795,
      "balance_loss_mlp": 1.01975858,
      "epoch": 0.6904346780496603,
      "flos": 28035137255040.0,
      "grad_norm": 1.8161343923149773,
      "language_loss": 0.69554293,
      "learning_rate": 9.236676592036628e-07,
      "loss": 0.71738863,
      "num_input_tokens_seen": 123524250,
      "step": 5742,
      "time_per_iteration": 2.6792664527893066
    },
    {
      "auxiliary_loss_clip": 0.01137212,
      "auxiliary_loss_mlp": 0.01027778,
      "balance_loss_clip": 1.04634643,
      "balance_loss_mlp": 1.02075338,
      "epoch": 0.6905549209402994,
      "flos": 23624266008960.0,
      "grad_norm": 1.8333666241987792,
      "language_loss": 0.73552585,
      "learning_rate": 9.230111933634228e-07,
      "loss": 0.75717568,
      "num_input_tokens_seen": 123545845,
      "step": 5743,
      "time_per_iteration": 2.6660268306732178
    },
    {
      "auxiliary_loss_clip": 0.01157747,
      "auxiliary_loss_mlp": 0.0103212,
      "balance_loss_clip": 1.04768682,
      "balance_loss_mlp": 1.02478838,
      "epoch": 0.6906751638309385,
      "flos": 23114945111040.0,
      "grad_norm": 1.483923879240007,
      "language_loss": 0.80768657,
      "learning_rate": 9.223548909015288e-07,
      "loss": 0.82958525,
      "num_input_tokens_seen": 123567535,
      "step": 5744,
      "time_per_iteration": 2.672898530960083
    },
    {
      "auxiliary_loss_clip": 0.0110788,
      "auxiliary_loss_mlp": 0.0102668,
      "balance_loss_clip": 1.04170871,
      "balance_loss_mlp": 1.01949763,
      "epoch": 0.6907954067215776,
      "flos": 27305468375040.0,
      "grad_norm": 1.8052687363118793,
      "language_loss": 0.72052419,
      "learning_rate": 9.216987519175407e-07,
      "loss": 0.74186981,
      "num_input_tokens_seen": 123587710,
      "step": 5745,
      "time_per_iteration": 2.776141405105591
    },
    {
      "auxiliary_loss_clip": 0.01152627,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.04731107,
      "balance_loss_mlp": 1.01841819,
      "epoch": 0.6909156496122166,
      "flos": 21689399070720.0,
      "grad_norm": 1.5954833753972755,
      "language_loss": 0.68410063,
      "learning_rate": 9.210427765109942e-07,
      "loss": 0.70587909,
      "num_input_tokens_seen": 123607385,
      "step": 5746,
      "time_per_iteration": 2.602827787399292
    },
    {
      "auxiliary_loss_clip": 0.011437,
      "auxiliary_loss_mlp": 0.01025801,
      "balance_loss_clip": 1.04393303,
      "balance_loss_mlp": 1.01824307,
      "epoch": 0.6910358925028558,
      "flos": 22561453463040.0,
      "grad_norm": 2.0213046251384363,
      "language_loss": 0.81125015,
      "learning_rate": 9.20386964781402e-07,
      "loss": 0.83294517,
      "num_input_tokens_seen": 123625405,
      "step": 5747,
      "time_per_iteration": 2.67291522026062
    },
    {
      "auxiliary_loss_clip": 0.01136116,
      "auxiliary_loss_mlp": 0.01024276,
      "balance_loss_clip": 1.04261875,
      "balance_loss_mlp": 1.01647687,
      "epoch": 0.6911561353934949,
      "flos": 22054107813120.0,
      "grad_norm": 2.229739579522858,
      "language_loss": 0.84490085,
      "learning_rate": 9.197313168282472e-07,
      "loss": 0.86650479,
      "num_input_tokens_seen": 123642850,
      "step": 5748,
      "time_per_iteration": 2.6242897510528564
    },
    {
      "auxiliary_loss_clip": 0.01150622,
      "auxiliary_loss_mlp": 0.0102487,
      "balance_loss_clip": 1.04338622,
      "balance_loss_mlp": 1.01778936,
      "epoch": 0.6912763782841339,
      "flos": 24206557386240.0,
      "grad_norm": 2.060458821778296,
      "language_loss": 0.72081953,
      "learning_rate": 9.190758327509935e-07,
      "loss": 0.74257445,
      "num_input_tokens_seen": 123661595,
      "step": 5749,
      "time_per_iteration": 2.631334066390991
    },
    {
      "auxiliary_loss_clip": 0.01031855,
      "auxiliary_loss_mlp": 0.00754296,
      "balance_loss_clip": 1.01249385,
      "balance_loss_mlp": 1.00018489,
      "epoch": 0.6913966211747731,
      "flos": 52329641091840.0,
      "grad_norm": 0.9288124407860343,
      "language_loss": 0.64404178,
      "learning_rate": 9.184205126490767e-07,
      "loss": 0.66190326,
      "num_input_tokens_seen": 123710490,
      "step": 5750,
      "time_per_iteration": 3.3088181018829346
    },
    {
      "auxiliary_loss_clip": 0.01041478,
      "auxiliary_loss_mlp": 0.00754132,
      "balance_loss_clip": 1.01475942,
      "balance_loss_mlp": 1.00024378,
      "epoch": 0.6915168640654121,
      "flos": 66741274851840.0,
      "grad_norm": 1.0944426119398258,
      "language_loss": 0.59633255,
      "learning_rate": 9.177653566219075e-07,
      "loss": 0.61428869,
      "num_input_tokens_seen": 123765215,
      "step": 5751,
      "time_per_iteration": 3.6083736419677734
    },
    {
      "auxiliary_loss_clip": 0.01129467,
      "auxiliary_loss_mlp": 0.01024081,
      "balance_loss_clip": 1.04137802,
      "balance_loss_mlp": 1.01704466,
      "epoch": 0.6916371069560512,
      "flos": 18296523175680.0,
      "grad_norm": 2.2201062898960777,
      "language_loss": 0.76399153,
      "learning_rate": 9.171103647688744e-07,
      "loss": 0.78552705,
      "num_input_tokens_seen": 123783955,
      "step": 5752,
      "time_per_iteration": 3.9796807765960693
    },
    {
      "auxiliary_loss_clip": 0.01078729,
      "auxiliary_loss_mlp": 0.0102313,
      "balance_loss_clip": 1.03839374,
      "balance_loss_mlp": 1.01623702,
      "epoch": 0.6917573498466904,
      "flos": 19645794685440.0,
      "grad_norm": 1.7999977643501113,
      "language_loss": 0.69300592,
      "learning_rate": 9.164555371893367e-07,
      "loss": 0.71402454,
      "num_input_tokens_seen": 123803885,
      "step": 5753,
      "time_per_iteration": 2.8831613063812256
    },
    {
      "auxiliary_loss_clip": 0.01155076,
      "auxiliary_loss_mlp": 0.00762123,
      "balance_loss_clip": 1.04694223,
      "balance_loss_mlp": 1.00048757,
      "epoch": 0.6918775927373294,
      "flos": 14210319985920.0,
      "grad_norm": 1.9860925236047187,
      "language_loss": 0.75482219,
      "learning_rate": 9.158008739826333e-07,
      "loss": 0.77399415,
      "num_input_tokens_seen": 123821485,
      "step": 5754,
      "time_per_iteration": 3.6422476768493652
    },
    {
      "auxiliary_loss_clip": 0.01139921,
      "auxiliary_loss_mlp": 0.0102638,
      "balance_loss_clip": 1.04579949,
      "balance_loss_mlp": 1.01906049,
      "epoch": 0.6919978356279685,
      "flos": 23985455218560.0,
      "grad_norm": 1.7817828261706212,
      "language_loss": 0.86741352,
      "learning_rate": 9.151463752480744e-07,
      "loss": 0.88907647,
      "num_input_tokens_seen": 123840215,
      "step": 5755,
      "time_per_iteration": 2.7939369678497314
    },
    {
      "auxiliary_loss_clip": 0.01116787,
      "auxiliary_loss_mlp": 0.01026402,
      "balance_loss_clip": 1.04035878,
      "balance_loss_mlp": 1.01938081,
      "epoch": 0.6921180785186076,
      "flos": 23622937205760.0,
      "grad_norm": 1.544130993260468,
      "language_loss": 0.80264813,
      "learning_rate": 9.144920410849493e-07,
      "loss": 0.82407999,
      "num_input_tokens_seen": 123861450,
      "step": 5756,
      "time_per_iteration": 3.6896073818206787
    },
    {
      "auxiliary_loss_clip": 0.01149659,
      "auxiliary_loss_mlp": 0.01027824,
      "balance_loss_clip": 1.04700303,
      "balance_loss_mlp": 1.02077281,
      "epoch": 0.6922383214092467,
      "flos": 21142623265920.0,
      "grad_norm": 1.6137093588188487,
      "language_loss": 0.80839121,
      "learning_rate": 9.138378715925176e-07,
      "loss": 0.8301661,
      "num_input_tokens_seen": 123880545,
      "step": 5757,
      "time_per_iteration": 2.644050359725952
    },
    {
      "auxiliary_loss_clip": 0.01131867,
      "auxiliary_loss_mlp": 0.01029659,
      "balance_loss_clip": 1.04092968,
      "balance_loss_mlp": 1.02232206,
      "epoch": 0.6923585642998857,
      "flos": 21470667200640.0,
      "grad_norm": 1.9387615437558192,
      "language_loss": 0.8100943,
      "learning_rate": 9.131838668700167e-07,
      "loss": 0.83170962,
      "num_input_tokens_seen": 123900615,
      "step": 5758,
      "time_per_iteration": 2.6702864170074463
    },
    {
      "auxiliary_loss_clip": 0.01124588,
      "auxiliary_loss_mlp": 0.0102463,
      "balance_loss_clip": 1.03982115,
      "balance_loss_mlp": 1.01696539,
      "epoch": 0.6924788071905249,
      "flos": 21105204272640.0,
      "grad_norm": 1.745645993692574,
      "language_loss": 0.86715937,
      "learning_rate": 9.125300270166598e-07,
      "loss": 0.88865149,
      "num_input_tokens_seen": 123921220,
      "step": 5759,
      "time_per_iteration": 2.7313644886016846
    },
    {
      "auxiliary_loss_clip": 0.01133044,
      "auxiliary_loss_mlp": 0.01022591,
      "balance_loss_clip": 1.04117107,
      "balance_loss_mlp": 1.01561975,
      "epoch": 0.692599050081164,
      "flos": 26250018117120.0,
      "grad_norm": 1.7032117997739717,
      "language_loss": 0.85530782,
      "learning_rate": 9.118763521316324e-07,
      "loss": 0.87686414,
      "num_input_tokens_seen": 123941795,
      "step": 5760,
      "time_per_iteration": 3.6700282096862793
    },
    {
      "auxiliary_loss_clip": 0.01167903,
      "auxiliary_loss_mlp": 0.00762375,
      "balance_loss_clip": 1.04564333,
      "balance_loss_mlp": 1.00046134,
      "epoch": 0.692719292971803,
      "flos": 20885215426560.0,
      "grad_norm": 1.869500513751974,
      "language_loss": 0.76051044,
      "learning_rate": 9.112228423140987e-07,
      "loss": 0.77981323,
      "num_input_tokens_seen": 123960715,
      "step": 5761,
      "time_per_iteration": 2.613426923751831
    },
    {
      "auxiliary_loss_clip": 0.01145784,
      "auxiliary_loss_mlp": 0.01023306,
      "balance_loss_clip": 1.04542041,
      "balance_loss_mlp": 1.01609945,
      "epoch": 0.6928395358624422,
      "flos": 25921938268800.0,
      "grad_norm": 2.3677708123754058,
      "language_loss": 0.86667734,
      "learning_rate": 9.105694976631932e-07,
      "loss": 0.88836825,
      "num_input_tokens_seen": 123978625,
      "step": 5762,
      "time_per_iteration": 2.712115526199341
    },
    {
      "auxiliary_loss_clip": 0.01157614,
      "auxiliary_loss_mlp": 0.01024677,
      "balance_loss_clip": 1.04777443,
      "balance_loss_mlp": 1.01739359,
      "epoch": 0.6929597787530812,
      "flos": 23586559706880.0,
      "grad_norm": 2.246975243028744,
      "language_loss": 0.72583926,
      "learning_rate": 9.099163182780283e-07,
      "loss": 0.74766213,
      "num_input_tokens_seen": 123996780,
      "step": 5763,
      "time_per_iteration": 2.6322591304779053
    },
    {
      "auxiliary_loss_clip": 0.01139989,
      "auxiliary_loss_mlp": 0.01023201,
      "balance_loss_clip": 1.0457052,
      "balance_loss_mlp": 1.01574111,
      "epoch": 0.6930800216437203,
      "flos": 18255656476800.0,
      "grad_norm": 2.7670021771331315,
      "language_loss": 0.48827931,
      "learning_rate": 9.092633042576916e-07,
      "loss": 0.50991124,
      "num_input_tokens_seen": 124014045,
      "step": 5764,
      "time_per_iteration": 2.6630594730377197
    },
    {
      "auxiliary_loss_clip": 0.01136025,
      "auxiliary_loss_mlp": 0.01026825,
      "balance_loss_clip": 1.0439868,
      "balance_loss_mlp": 1.01964223,
      "epoch": 0.6932002645343595,
      "flos": 29168621809920.0,
      "grad_norm": 1.8565396110218157,
      "language_loss": 0.56068003,
      "learning_rate": 9.086104557012446e-07,
      "loss": 0.58230853,
      "num_input_tokens_seen": 124034615,
      "step": 5765,
      "time_per_iteration": 2.8044261932373047
    },
    {
      "auxiliary_loss_clip": 0.01145874,
      "auxiliary_loss_mlp": 0.01019842,
      "balance_loss_clip": 1.0435878,
      "balance_loss_mlp": 1.01321054,
      "epoch": 0.6933205074249985,
      "flos": 23842746483840.0,
      "grad_norm": 3.149643354891842,
      "language_loss": 0.65533042,
      "learning_rate": 9.079577727077239e-07,
      "loss": 0.67698753,
      "num_input_tokens_seen": 124053445,
      "step": 5766,
      "time_per_iteration": 2.6773533821105957
    },
    {
      "auxiliary_loss_clip": 0.01157277,
      "auxiliary_loss_mlp": 0.01023529,
      "balance_loss_clip": 1.04676759,
      "balance_loss_mlp": 1.01597762,
      "epoch": 0.6934407503156376,
      "flos": 24166696268160.0,
      "grad_norm": 2.100180396246994,
      "language_loss": 0.71877599,
      "learning_rate": 9.073052553761404e-07,
      "loss": 0.74058414,
      "num_input_tokens_seen": 124072810,
      "step": 5767,
      "time_per_iteration": 2.747452974319458
    },
    {
      "auxiliary_loss_clip": 0.0111231,
      "auxiliary_loss_mlp": 0.01025,
      "balance_loss_clip": 1.03995883,
      "balance_loss_mlp": 1.01746011,
      "epoch": 0.6935609932062767,
      "flos": 20631327120000.0,
      "grad_norm": 1.6571405595456843,
      "language_loss": 0.78115249,
      "learning_rate": 9.066529038054805e-07,
      "loss": 0.80252564,
      "num_input_tokens_seen": 124092875,
      "step": 5768,
      "time_per_iteration": 2.7450742721557617
    },
    {
      "auxiliary_loss_clip": 0.01137234,
      "auxiliary_loss_mlp": 0.01029295,
      "balance_loss_clip": 1.04200625,
      "balance_loss_mlp": 1.02241623,
      "epoch": 0.6936812360969158,
      "flos": 18254184019200.0,
      "grad_norm": 2.105996261721915,
      "language_loss": 0.74100769,
      "learning_rate": 9.060007180947071e-07,
      "loss": 0.76267296,
      "num_input_tokens_seen": 124110930,
      "step": 5769,
      "time_per_iteration": 2.688722848892212
    },
    {
      "auxiliary_loss_clip": 0.01114959,
      "auxiliary_loss_mlp": 0.01027632,
      "balance_loss_clip": 1.04027319,
      "balance_loss_mlp": 1.01987469,
      "epoch": 0.6938014789875548,
      "flos": 31317336368640.0,
      "grad_norm": 1.9201081908493263,
      "language_loss": 0.73071945,
      "learning_rate": 9.053486983427534e-07,
      "loss": 0.75214535,
      "num_input_tokens_seen": 124132180,
      "step": 5770,
      "time_per_iteration": 2.7647948265075684
    },
    {
      "auxiliary_loss_clip": 0.01142621,
      "auxiliary_loss_mlp": 0.01022127,
      "balance_loss_clip": 1.04125881,
      "balance_loss_mlp": 1.01487267,
      "epoch": 0.6939217218781939,
      "flos": 17528429721600.0,
      "grad_norm": 1.931262579011117,
      "language_loss": 0.70828652,
      "learning_rate": 9.046968446485326e-07,
      "loss": 0.72993404,
      "num_input_tokens_seen": 124150585,
      "step": 5771,
      "time_per_iteration": 2.642146348953247
    },
    {
      "auxiliary_loss_clip": 0.01159254,
      "auxiliary_loss_mlp": 0.01021222,
      "balance_loss_clip": 1.04791248,
      "balance_loss_mlp": 1.01333618,
      "epoch": 0.6940419647688331,
      "flos": 18551776199040.0,
      "grad_norm": 2.185519904401988,
      "language_loss": 0.70548266,
      "learning_rate": 9.040451571109295e-07,
      "loss": 0.72728747,
      "num_input_tokens_seen": 124166205,
      "step": 5772,
      "time_per_iteration": 2.5596513748168945
    },
    {
      "auxiliary_loss_clip": 0.01047245,
      "auxiliary_loss_mlp": 0.01001773,
      "balance_loss_clip": 1.02418566,
      "balance_loss_mlp": 1.00041986,
      "epoch": 0.6941622076594721,
      "flos": 66926286829440.0,
      "grad_norm": 0.8314282134279664,
      "language_loss": 0.60438156,
      "learning_rate": 9.033936358288042e-07,
      "loss": 0.62487173,
      "num_input_tokens_seen": 124219940,
      "step": 5773,
      "time_per_iteration": 3.1799252033233643
    },
    {
      "auxiliary_loss_clip": 0.01169343,
      "auxiliary_loss_mlp": 0.01022792,
      "balance_loss_clip": 1.04666185,
      "balance_loss_mlp": 1.015154,
      "epoch": 0.6942824505501112,
      "flos": 26578062051840.0,
      "grad_norm": 1.7343229219088774,
      "language_loss": 0.82318759,
      "learning_rate": 9.027422809009937e-07,
      "loss": 0.84510899,
      "num_input_tokens_seen": 124239885,
      "step": 5774,
      "time_per_iteration": 2.570260524749756
    },
    {
      "auxiliary_loss_clip": 0.01156408,
      "auxiliary_loss_mlp": 0.01026984,
      "balance_loss_clip": 1.04407632,
      "balance_loss_mlp": 1.01976943,
      "epoch": 0.6944026934407503,
      "flos": 21248308056960.0,
      "grad_norm": 1.8258830462070064,
      "language_loss": 0.83420086,
      "learning_rate": 9.020910924263054e-07,
      "loss": 0.85603476,
      "num_input_tokens_seen": 124258410,
      "step": 5775,
      "time_per_iteration": 2.5962181091308594
    },
    {
      "auxiliary_loss_clip": 0.01046997,
      "auxiliary_loss_mlp": 0.01000831,
      "balance_loss_clip": 1.02507305,
      "balance_loss_mlp": 0.99948424,
      "epoch": 0.6945229363313894,
      "flos": 70677191537280.0,
      "grad_norm": 0.8151225802804196,
      "language_loss": 0.58105022,
      "learning_rate": 9.014400705035261e-07,
      "loss": 0.60152853,
      "num_input_tokens_seen": 124315315,
      "step": 5776,
      "time_per_iteration": 3.3166887760162354
    },
    {
      "auxiliary_loss_clip": 0.01167967,
      "auxiliary_loss_mlp": 0.01026035,
      "balance_loss_clip": 1.04807281,
      "balance_loss_mlp": 1.01904893,
      "epoch": 0.6946431792220285,
      "flos": 18952934267520.0,
      "grad_norm": 2.1288616960230735,
      "language_loss": 0.76860809,
      "learning_rate": 9.00789215231414e-07,
      "loss": 0.79054809,
      "num_input_tokens_seen": 124333710,
      "step": 5777,
      "time_per_iteration": 3.4972164630889893
    },
    {
      "auxiliary_loss_clip": 0.01123309,
      "auxiliary_loss_mlp": 0.00763108,
      "balance_loss_clip": 1.03725123,
      "balance_loss_mlp": 1.00049448,
      "epoch": 0.6947634221126676,
      "flos": 20338834671360.0,
      "grad_norm": 1.7127678941792257,
      "language_loss": 0.8183679,
      "learning_rate": 9.001385267087056e-07,
      "loss": 0.83723205,
      "num_input_tokens_seen": 124352855,
      "step": 5778,
      "time_per_iteration": 2.7845985889434814
    },
    {
      "auxiliary_loss_clip": 0.01158932,
      "auxiliary_loss_mlp": 0.01023716,
      "balance_loss_clip": 1.04672277,
      "balance_loss_mlp": 1.01674795,
      "epoch": 0.6948836650033067,
      "flos": 21833723917440.0,
      "grad_norm": 1.5206827550709692,
      "language_loss": 0.7045756,
      "learning_rate": 8.994880050341072e-07,
      "loss": 0.72640204,
      "num_input_tokens_seen": 124372960,
      "step": 5779,
      "time_per_iteration": 2.554785966873169
    },
    {
      "auxiliary_loss_clip": 0.01136509,
      "auxiliary_loss_mlp": 0.01030478,
      "balance_loss_clip": 1.04558706,
      "balance_loss_mlp": 1.02275348,
      "epoch": 0.6950039078939457,
      "flos": 23657519024640.0,
      "grad_norm": 1.7078173427065748,
      "language_loss": 0.77634847,
      "learning_rate": 8.988376503063026e-07,
      "loss": 0.7980184,
      "num_input_tokens_seen": 124394220,
      "step": 5780,
      "time_per_iteration": 2.724730968475342
    },
    {
      "auxiliary_loss_clip": 0.01119174,
      "auxiliary_loss_mlp": 0.01028581,
      "balance_loss_clip": 1.04084873,
      "balance_loss_mlp": 1.02108824,
      "epoch": 0.6951241507845849,
      "flos": 21792462168960.0,
      "grad_norm": 1.8405415639220446,
      "language_loss": 0.81867099,
      "learning_rate": 8.981874626239521e-07,
      "loss": 0.84014851,
      "num_input_tokens_seen": 124412795,
      "step": 5781,
      "time_per_iteration": 3.7039763927459717
    },
    {
      "auxiliary_loss_clip": 0.01158805,
      "auxiliary_loss_mlp": 0.01032138,
      "balance_loss_clip": 1.04908299,
      "balance_loss_mlp": 1.02483654,
      "epoch": 0.695244393675224,
      "flos": 14647568244480.0,
      "grad_norm": 2.101260795474958,
      "language_loss": 0.87924731,
      "learning_rate": 8.975374420856872e-07,
      "loss": 0.90115666,
      "num_input_tokens_seen": 124429690,
      "step": 5782,
      "time_per_iteration": 3.5646626949310303
    },
    {
      "auxiliary_loss_clip": 0.01117191,
      "auxiliary_loss_mlp": 0.01025405,
      "balance_loss_clip": 1.03887415,
      "balance_loss_mlp": 1.01829386,
      "epoch": 0.695364636565863,
      "flos": 16873203778560.0,
      "grad_norm": 2.299918211702129,
      "language_loss": 0.72771406,
      "learning_rate": 8.968875887901157e-07,
      "loss": 0.74914008,
      "num_input_tokens_seen": 124447070,
      "step": 5783,
      "time_per_iteration": 2.748270034790039
    },
    {
      "auxiliary_loss_clip": 0.01140294,
      "auxiliary_loss_mlp": 0.010281,
      "balance_loss_clip": 1.04169798,
      "balance_loss_mlp": 1.02072954,
      "epoch": 0.6954848794565022,
      "flos": 19354523299200.0,
      "grad_norm": 2.619582447003379,
      "language_loss": 0.63283062,
      "learning_rate": 8.9623790283582e-07,
      "loss": 0.65451455,
      "num_input_tokens_seen": 124464950,
      "step": 5784,
      "time_per_iteration": 2.6429190635681152
    },
    {
      "auxiliary_loss_clip": 0.01127531,
      "auxiliary_loss_mlp": 0.01026895,
      "balance_loss_clip": 1.04311848,
      "balance_loss_mlp": 1.01901495,
      "epoch": 0.6956051223471412,
      "flos": 18990209606400.0,
      "grad_norm": 2.2516763161422837,
      "language_loss": 0.76560092,
      "learning_rate": 8.955883843213561e-07,
      "loss": 0.78714514,
      "num_input_tokens_seen": 124483965,
      "step": 5785,
      "time_per_iteration": 2.7013463973999023
    },
    {
      "auxiliary_loss_clip": 0.01161296,
      "auxiliary_loss_mlp": 0.01030421,
      "balance_loss_clip": 1.04470611,
      "balance_loss_mlp": 1.02289271,
      "epoch": 0.6957253652377803,
      "flos": 16107229226880.0,
      "grad_norm": 1.8124991524136504,
      "language_loss": 0.868613,
      "learning_rate": 8.949390333452569e-07,
      "loss": 0.89053017,
      "num_input_tokens_seen": 124501910,
      "step": 5786,
      "time_per_iteration": 3.457934617996216
    },
    {
      "auxiliary_loss_clip": 0.01170254,
      "auxiliary_loss_mlp": 0.01025783,
      "balance_loss_clip": 1.05022728,
      "balance_loss_mlp": 1.01818347,
      "epoch": 0.6958456081284194,
      "flos": 29388646569600.0,
      "grad_norm": 2.584747929030377,
      "language_loss": 0.67882669,
      "learning_rate": 8.942898500060279e-07,
      "loss": 0.70078707,
      "num_input_tokens_seen": 124521625,
      "step": 5787,
      "time_per_iteration": 2.7019872665405273
    },
    {
      "auxiliary_loss_clip": 0.01123182,
      "auxiliary_loss_mlp": 0.01028444,
      "balance_loss_clip": 1.04434597,
      "balance_loss_mlp": 1.02144313,
      "epoch": 0.6959658510190585,
      "flos": 25154850395520.0,
      "grad_norm": 2.3997787921380493,
      "language_loss": 0.71892726,
      "learning_rate": 8.936408344021493e-07,
      "loss": 0.74044353,
      "num_input_tokens_seen": 124538540,
      "step": 5788,
      "time_per_iteration": 2.7720470428466797
    },
    {
      "auxiliary_loss_clip": 0.0115104,
      "auxiliary_loss_mlp": 0.01025494,
      "balance_loss_clip": 1.047876,
      "balance_loss_mlp": 1.01819885,
      "epoch": 0.6960860939096976,
      "flos": 42814388759040.0,
      "grad_norm": 2.2014817142966763,
      "language_loss": 0.70903707,
      "learning_rate": 8.929919866320765e-07,
      "loss": 0.73080242,
      "num_input_tokens_seen": 124559355,
      "step": 5789,
      "time_per_iteration": 2.882000207901001
    },
    {
      "auxiliary_loss_clip": 0.01134134,
      "auxiliary_loss_mlp": 0.00762251,
      "balance_loss_clip": 1.04371631,
      "balance_loss_mlp": 1.00045705,
      "epoch": 0.6962063368003367,
      "flos": 17566566986880.0,
      "grad_norm": 1.9550747144180705,
      "language_loss": 0.8174687,
      "learning_rate": 8.923433067942385e-07,
      "loss": 0.83643258,
      "num_input_tokens_seen": 124577920,
      "step": 5790,
      "time_per_iteration": 2.7166712284088135
    },
    {
      "auxiliary_loss_clip": 0.01136567,
      "auxiliary_loss_mlp": 0.01026331,
      "balance_loss_clip": 1.04537797,
      "balance_loss_mlp": 1.01868367,
      "epoch": 0.6963265796909758,
      "flos": 21251648021760.0,
      "grad_norm": 2.8045280358435987,
      "language_loss": 0.68638271,
      "learning_rate": 8.916947949870417e-07,
      "loss": 0.70801169,
      "num_input_tokens_seen": 124597585,
      "step": 5791,
      "time_per_iteration": 2.6313536167144775
    },
    {
      "auxiliary_loss_clip": 0.01061905,
      "auxiliary_loss_mlp": 0.01001687,
      "balance_loss_clip": 1.01443815,
      "balance_loss_mlp": 1.00060236,
      "epoch": 0.6964468225816148,
      "flos": 68828295801600.0,
      "grad_norm": 0.7499801687828961,
      "language_loss": 0.58118498,
      "learning_rate": 8.910464513088615e-07,
      "loss": 0.60182095,
      "num_input_tokens_seen": 124661625,
      "step": 5792,
      "time_per_iteration": 3.332893133163452
    },
    {
      "auxiliary_loss_clip": 0.01133585,
      "auxiliary_loss_mlp": 0.01022286,
      "balance_loss_clip": 1.04268146,
      "balance_loss_mlp": 1.01488614,
      "epoch": 0.696567065472254,
      "flos": 18950887192320.0,
      "grad_norm": 1.8145240945004306,
      "language_loss": 0.78503406,
      "learning_rate": 8.903982758580542e-07,
      "loss": 0.80659282,
      "num_input_tokens_seen": 124680565,
      "step": 5793,
      "time_per_iteration": 2.6937830448150635
    },
    {
      "auxiliary_loss_clip": 0.01138484,
      "auxiliary_loss_mlp": 0.01022968,
      "balance_loss_clip": 1.04422808,
      "balance_loss_mlp": 1.01520801,
      "epoch": 0.696687308362893,
      "flos": 22856675345280.0,
      "grad_norm": 1.9727758615387083,
      "language_loss": 0.80280197,
      "learning_rate": 8.897502687329457e-07,
      "loss": 0.82441652,
      "num_input_tokens_seen": 124700365,
      "step": 5794,
      "time_per_iteration": 2.7059104442596436
    },
    {
      "auxiliary_loss_clip": 0.01124038,
      "auxiliary_loss_mlp": 0.01024094,
      "balance_loss_clip": 1.04166651,
      "balance_loss_mlp": 1.01661992,
      "epoch": 0.6968075512535321,
      "flos": 24972926987520.0,
      "grad_norm": 2.2491711429016648,
      "language_loss": 0.80008692,
      "learning_rate": 8.891024300318382e-07,
      "loss": 0.82156825,
      "num_input_tokens_seen": 124718935,
      "step": 5795,
      "time_per_iteration": 2.734971284866333
    },
    {
      "auxiliary_loss_clip": 0.01117987,
      "auxiliary_loss_mlp": 0.01031553,
      "balance_loss_clip": 1.04061222,
      "balance_loss_mlp": 1.02495193,
      "epoch": 0.6969277941441713,
      "flos": 21030438113280.0,
      "grad_norm": 2.15729182418763,
      "language_loss": 0.75928724,
      "learning_rate": 8.884547598530103e-07,
      "loss": 0.78078264,
      "num_input_tokens_seen": 124739505,
      "step": 5796,
      "time_per_iteration": 2.6992619037628174
    },
    {
      "auxiliary_loss_clip": 0.01075505,
      "auxiliary_loss_mlp": 0.01032569,
      "balance_loss_clip": 1.03667617,
      "balance_loss_mlp": 1.02508891,
      "epoch": 0.6970480370348103,
      "flos": 21579404647680.0,
      "grad_norm": 1.7364400788061227,
      "language_loss": 0.75333351,
      "learning_rate": 8.8780725829471e-07,
      "loss": 0.77441424,
      "num_input_tokens_seen": 124757410,
      "step": 5797,
      "time_per_iteration": 2.818135976791382
    },
    {
      "auxiliary_loss_clip": 0.01172984,
      "auxiliary_loss_mlp": 0.01026536,
      "balance_loss_clip": 1.04952741,
      "balance_loss_mlp": 1.01958656,
      "epoch": 0.6971682799254494,
      "flos": 22419175691520.0,
      "grad_norm": 2.1667887720376906,
      "language_loss": 0.78315806,
      "learning_rate": 8.87159925455165e-07,
      "loss": 0.80515325,
      "num_input_tokens_seen": 124777240,
      "step": 5798,
      "time_per_iteration": 2.653728723526001
    },
    {
      "auxiliary_loss_clip": 0.011214,
      "auxiliary_loss_mlp": 0.01031667,
      "balance_loss_clip": 1.04019141,
      "balance_loss_mlp": 1.02456248,
      "epoch": 0.6972885228160886,
      "flos": 20005834659840.0,
      "grad_norm": 2.104885622348916,
      "language_loss": 0.73080021,
      "learning_rate": 8.865127614325738e-07,
      "loss": 0.75233084,
      "num_input_tokens_seen": 124795670,
      "step": 5799,
      "time_per_iteration": 2.666292190551758
    },
    {
      "auxiliary_loss_clip": 0.01135207,
      "auxiliary_loss_mlp": 0.01027601,
      "balance_loss_clip": 1.04159451,
      "balance_loss_mlp": 1.01913667,
      "epoch": 0.6974087657067276,
      "flos": 37853437656960.0,
      "grad_norm": 1.793853051994157,
      "language_loss": 0.66844225,
      "learning_rate": 8.85865766325113e-07,
      "loss": 0.69007039,
      "num_input_tokens_seen": 124819600,
      "step": 5800,
      "time_per_iteration": 2.763054847717285
    },
    {
      "auxiliary_loss_clip": 0.01139781,
      "auxiliary_loss_mlp": 0.01028612,
      "balance_loss_clip": 1.04437852,
      "balance_loss_mlp": 1.02048814,
      "epoch": 0.6975290085973667,
      "flos": 29489267543040.0,
      "grad_norm": 3.073350199315057,
      "language_loss": 0.71708202,
      "learning_rate": 8.852189402309287e-07,
      "loss": 0.73876595,
      "num_input_tokens_seen": 124838785,
      "step": 5801,
      "time_per_iteration": 2.704833745956421
    },
    {
      "auxiliary_loss_clip": 0.01155709,
      "auxiliary_loss_mlp": 0.01022941,
      "balance_loss_clip": 1.04628098,
      "balance_loss_mlp": 1.01528764,
      "epoch": 0.6976492514880057,
      "flos": 12895630295040.0,
      "grad_norm": 2.147682492184672,
      "language_loss": 0.74284005,
      "learning_rate": 8.845722832481441e-07,
      "loss": 0.76462662,
      "num_input_tokens_seen": 124854215,
      "step": 5802,
      "time_per_iteration": 2.5720767974853516
    },
    {
      "auxiliary_loss_clip": 0.01155238,
      "auxiliary_loss_mlp": 0.01031177,
      "balance_loss_clip": 1.04612362,
      "balance_loss_mlp": 1.02344608,
      "epoch": 0.6977694943786449,
      "flos": 24352929308160.0,
      "grad_norm": 2.031152574377187,
      "language_loss": 0.77410322,
      "learning_rate": 8.83925795474858e-07,
      "loss": 0.79596734,
      "num_input_tokens_seen": 124874340,
      "step": 5803,
      "time_per_iteration": 2.6433310508728027
    },
    {
      "auxiliary_loss_clip": 0.01126396,
      "auxiliary_loss_mlp": 0.01027358,
      "balance_loss_clip": 1.04525471,
      "balance_loss_mlp": 1.01951146,
      "epoch": 0.6978897372692839,
      "flos": 29898470257920.0,
      "grad_norm": 2.3711461055461593,
      "language_loss": 0.59320617,
      "learning_rate": 8.832794770091414e-07,
      "loss": 0.61474371,
      "num_input_tokens_seen": 124895175,
      "step": 5804,
      "time_per_iteration": 3.6935064792633057
    },
    {
      "auxiliary_loss_clip": 0.01145509,
      "auxiliary_loss_mlp": 0.01023336,
      "balance_loss_clip": 1.04327941,
      "balance_loss_mlp": 1.0159924,
      "epoch": 0.698009980159923,
      "flos": 21761579450880.0,
      "grad_norm": 2.148732620379686,
      "language_loss": 0.82424408,
      "learning_rate": 8.826333279490401e-07,
      "loss": 0.84593254,
      "num_input_tokens_seen": 124915810,
      "step": 5805,
      "time_per_iteration": 2.6968674659729004
    },
    {
      "auxiliary_loss_clip": 0.01144751,
      "auxiliary_loss_mlp": 0.01024037,
      "balance_loss_clip": 1.04707539,
      "balance_loss_mlp": 1.01693559,
      "epoch": 0.6981302230505622,
      "flos": 19857164267520.0,
      "grad_norm": 2.4500563451727366,
      "language_loss": 0.68242919,
      "learning_rate": 8.819873483925748e-07,
      "loss": 0.70411706,
      "num_input_tokens_seen": 124932930,
      "step": 5806,
      "time_per_iteration": 3.61437726020813
    },
    {
      "auxiliary_loss_clip": 0.01133331,
      "auxiliary_loss_mlp": 0.00762286,
      "balance_loss_clip": 1.04535079,
      "balance_loss_mlp": 1.00045228,
      "epoch": 0.6982504659412012,
      "flos": 22198648141440.0,
      "grad_norm": 2.204926847643738,
      "language_loss": 0.74118114,
      "learning_rate": 8.81341538437739e-07,
      "loss": 0.76013732,
      "num_input_tokens_seen": 124951220,
      "step": 5807,
      "time_per_iteration": 2.7330801486968994
    },
    {
      "auxiliary_loss_clip": 0.01145729,
      "auxiliary_loss_mlp": 0.01029638,
      "balance_loss_clip": 1.04303432,
      "balance_loss_mlp": 1.02218151,
      "epoch": 0.6983707088318403,
      "flos": 35588479708800.0,
      "grad_norm": 2.095883204222148,
      "language_loss": 0.67945111,
      "learning_rate": 8.80695898182503e-07,
      "loss": 0.70120478,
      "num_input_tokens_seen": 124972200,
      "step": 5808,
      "time_per_iteration": 3.657522439956665
    },
    {
      "auxiliary_loss_clip": 0.01059648,
      "auxiliary_loss_mlp": 0.01003513,
      "balance_loss_clip": 1.01885343,
      "balance_loss_mlp": 1.00214815,
      "epoch": 0.6984909517224794,
      "flos": 65440052760960.0,
      "grad_norm": 0.8842228311734813,
      "language_loss": 0.65112889,
      "learning_rate": 8.800504277248093e-07,
      "loss": 0.67176056,
      "num_input_tokens_seen": 125036950,
      "step": 5809,
      "time_per_iteration": 3.259469747543335
    },
    {
      "auxiliary_loss_clip": 0.01125676,
      "auxiliary_loss_mlp": 0.00762294,
      "balance_loss_clip": 1.04773569,
      "balance_loss_mlp": 1.00051689,
      "epoch": 0.6986111946131185,
      "flos": 18546927863040.0,
      "grad_norm": 1.6075793972597947,
      "language_loss": 0.75009966,
      "learning_rate": 8.794051271625753e-07,
      "loss": 0.76897937,
      "num_input_tokens_seen": 125054585,
      "step": 5810,
      "time_per_iteration": 2.6702182292938232
    },
    {
      "auxiliary_loss_clip": 0.0114153,
      "auxiliary_loss_mlp": 0.01026639,
      "balance_loss_clip": 1.04408193,
      "balance_loss_mlp": 1.0198319,
      "epoch": 0.6987314375037575,
      "flos": 23039173370880.0,
      "grad_norm": 1.645219443420088,
      "language_loss": 0.83282971,
      "learning_rate": 8.787599965936925e-07,
      "loss": 0.85451138,
      "num_input_tokens_seen": 125075515,
      "step": 5811,
      "time_per_iteration": 2.6780834197998047
    },
    {
      "auxiliary_loss_clip": 0.01119823,
      "auxiliary_loss_mlp": 0.0102696,
      "balance_loss_clip": 1.04290998,
      "balance_loss_mlp": 1.01970637,
      "epoch": 0.6988516803943967,
      "flos": 38400393029760.0,
      "grad_norm": 1.6484467305671195,
      "language_loss": 0.71975595,
      "learning_rate": 8.781150361160261e-07,
      "loss": 0.74122369,
      "num_input_tokens_seen": 125097425,
      "step": 5812,
      "time_per_iteration": 3.7405972480773926
    },
    {
      "auxiliary_loss_clip": 0.01131614,
      "auxiliary_loss_mlp": 0.01028753,
      "balance_loss_clip": 1.04408669,
      "balance_loss_mlp": 1.02078998,
      "epoch": 0.6989719232850358,
      "flos": 24096993926400.0,
      "grad_norm": 1.6260899064992103,
      "language_loss": 0.73579001,
      "learning_rate": 8.774702458274181e-07,
      "loss": 0.75739366,
      "num_input_tokens_seen": 125117830,
      "step": 5813,
      "time_per_iteration": 2.682807683944702
    },
    {
      "auxiliary_loss_clip": 0.01154348,
      "auxiliary_loss_mlp": 0.01026396,
      "balance_loss_clip": 1.04568839,
      "balance_loss_mlp": 1.01941347,
      "epoch": 0.6990921661756748,
      "flos": 14866838818560.0,
      "grad_norm": 2.7274135448381904,
      "language_loss": 0.70809257,
      "learning_rate": 8.768256258256799e-07,
      "loss": 0.72989994,
      "num_input_tokens_seen": 125134455,
      "step": 5814,
      "time_per_iteration": 2.652221202850342
    },
    {
      "auxiliary_loss_clip": 0.01161643,
      "auxiliary_loss_mlp": 0.01026496,
      "balance_loss_clip": 1.04809618,
      "balance_loss_mlp": 1.01898623,
      "epoch": 0.699212409066314,
      "flos": 20193719725440.0,
      "grad_norm": 1.8023459818206529,
      "language_loss": 0.74065483,
      "learning_rate": 8.76181176208602e-07,
      "loss": 0.76253617,
      "num_input_tokens_seen": 125152555,
      "step": 5815,
      "time_per_iteration": 2.5946483612060547
    },
    {
      "auxiliary_loss_clip": 0.01101589,
      "auxiliary_loss_mlp": 0.01028263,
      "balance_loss_clip": 1.03863621,
      "balance_loss_mlp": 1.02058625,
      "epoch": 0.699332651956953,
      "flos": 19427888828160.0,
      "grad_norm": 1.7875823160052262,
      "language_loss": 0.73591673,
      "learning_rate": 8.755368970739461e-07,
      "loss": 0.75721526,
      "num_input_tokens_seen": 125171915,
      "step": 5816,
      "time_per_iteration": 2.7624330520629883
    },
    {
      "auxiliary_loss_clip": 0.01132999,
      "auxiliary_loss_mlp": 0.01025391,
      "balance_loss_clip": 1.04224825,
      "balance_loss_mlp": 1.01757669,
      "epoch": 0.6994528948475921,
      "flos": 16143714466560.0,
      "grad_norm": 2.3064717999899624,
      "language_loss": 0.61137187,
      "learning_rate": 8.748927885194479e-07,
      "loss": 0.63295579,
      "num_input_tokens_seen": 125190220,
      "step": 5817,
      "time_per_iteration": 2.706987142562866
    },
    {
      "auxiliary_loss_clip": 0.01033998,
      "auxiliary_loss_mlp": 0.01003455,
      "balance_loss_clip": 1.017295,
      "balance_loss_mlp": 1.00226295,
      "epoch": 0.6995731377382313,
      "flos": 64952420699520.0,
      "grad_norm": 0.8273315282411723,
      "language_loss": 0.57402635,
      "learning_rate": 8.742488506428209e-07,
      "loss": 0.59440088,
      "num_input_tokens_seen": 125249310,
      "step": 5818,
      "time_per_iteration": 3.253621816635132
    },
    {
      "auxiliary_loss_clip": 0.01143751,
      "auxiliary_loss_mlp": 0.00762243,
      "balance_loss_clip": 1.04386973,
      "balance_loss_mlp": 1.00041771,
      "epoch": 0.6996933806288703,
      "flos": 24900136076160.0,
      "grad_norm": 2.0264930989992496,
      "language_loss": 0.78336942,
      "learning_rate": 8.736050835417466e-07,
      "loss": 0.80242932,
      "num_input_tokens_seen": 125269350,
      "step": 5819,
      "time_per_iteration": 2.794365882873535
    },
    {
      "auxiliary_loss_clip": 0.01159429,
      "auxiliary_loss_mlp": 0.01025237,
      "balance_loss_clip": 1.04593635,
      "balance_loss_mlp": 1.01777172,
      "epoch": 0.6998136235195094,
      "flos": 20777806782720.0,
      "grad_norm": 2.0566667762465807,
      "language_loss": 0.61141658,
      "learning_rate": 8.729614873138862e-07,
      "loss": 0.63326323,
      "num_input_tokens_seen": 125286985,
      "step": 5820,
      "time_per_iteration": 2.759895086288452
    },
    {
      "auxiliary_loss_clip": 0.0111997,
      "auxiliary_loss_mlp": 0.01027525,
      "balance_loss_clip": 1.04433441,
      "balance_loss_mlp": 1.01971698,
      "epoch": 0.6999338664101485,
      "flos": 23733470332800.0,
      "grad_norm": 1.9537224207431503,
      "language_loss": 0.77839947,
      "learning_rate": 8.723180620568716e-07,
      "loss": 0.79987442,
      "num_input_tokens_seen": 125306240,
      "step": 5821,
      "time_per_iteration": 2.769481658935547
    },
    {
      "auxiliary_loss_clip": 0.01145161,
      "auxiliary_loss_mlp": 0.01027505,
      "balance_loss_clip": 1.04412174,
      "balance_loss_mlp": 1.02004838,
      "epoch": 0.7000541093007876,
      "flos": 19864598382720.0,
      "grad_norm": 1.9832436344437712,
      "language_loss": 0.85252964,
      "learning_rate": 8.716748078683116e-07,
      "loss": 0.87425625,
      "num_input_tokens_seen": 125323015,
      "step": 5822,
      "time_per_iteration": 2.6799557209014893
    },
    {
      "auxiliary_loss_clip": 0.01079107,
      "auxiliary_loss_mlp": 0.01030416,
      "balance_loss_clip": 1.03883004,
      "balance_loss_mlp": 1.02248263,
      "epoch": 0.7001743521914267,
      "flos": 29679056029440.0,
      "grad_norm": 2.0555862421934017,
      "language_loss": 0.6878438,
      "learning_rate": 8.710317248457855e-07,
      "loss": 0.70893896,
      "num_input_tokens_seen": 125342630,
      "step": 5823,
      "time_per_iteration": 2.8831534385681152
    },
    {
      "auxiliary_loss_clip": 0.01137922,
      "auxiliary_loss_mlp": 0.01024302,
      "balance_loss_clip": 1.04607344,
      "balance_loss_mlp": 1.01666617,
      "epoch": 0.7002945950820658,
      "flos": 27489762080640.0,
      "grad_norm": 1.918262758777171,
      "language_loss": 0.7247088,
      "learning_rate": 8.703888130868482e-07,
      "loss": 0.74633098,
      "num_input_tokens_seen": 125364480,
      "step": 5824,
      "time_per_iteration": 2.8230650424957275
    },
    {
      "auxiliary_loss_clip": 0.01128457,
      "auxiliary_loss_mlp": 0.0102731,
      "balance_loss_clip": 1.04390609,
      "balance_loss_mlp": 1.01982605,
      "epoch": 0.7004148379727049,
      "flos": 22158463800960.0,
      "grad_norm": 3.0668078060857367,
      "language_loss": 0.82067657,
      "learning_rate": 8.697460726890307e-07,
      "loss": 0.84223425,
      "num_input_tokens_seen": 125381625,
      "step": 5825,
      "time_per_iteration": 2.7540905475616455
    },
    {
      "auxiliary_loss_clip": 0.01125604,
      "auxiliary_loss_mlp": 0.00762455,
      "balance_loss_clip": 1.03998923,
      "balance_loss_mlp": 1.000525,
      "epoch": 0.7005350808633439,
      "flos": 19423758764160.0,
      "grad_norm": 2.0669687758409263,
      "language_loss": 0.90771532,
      "learning_rate": 8.691035037498354e-07,
      "loss": 0.92659587,
      "num_input_tokens_seen": 125397615,
      "step": 5826,
      "time_per_iteration": 2.7043471336364746
    },
    {
      "auxiliary_loss_clip": 0.01135502,
      "auxiliary_loss_mlp": 0.01023972,
      "balance_loss_clip": 1.04112029,
      "balance_loss_mlp": 1.01689982,
      "epoch": 0.7006553237539831,
      "flos": 23476708938240.0,
      "grad_norm": 1.6305655583176488,
      "language_loss": 0.72444487,
      "learning_rate": 8.684611063667391e-07,
      "loss": 0.74603957,
      "num_input_tokens_seen": 125418080,
      "step": 5827,
      "time_per_iteration": 2.680734395980835
    },
    {
      "auxiliary_loss_clip": 0.0115397,
      "auxiliary_loss_mlp": 0.01026331,
      "balance_loss_clip": 1.04471183,
      "balance_loss_mlp": 1.01904762,
      "epoch": 0.7007755666446221,
      "flos": 31212872640000.0,
      "grad_norm": 2.5689016829073945,
      "language_loss": 0.76891118,
      "learning_rate": 8.678188806371935e-07,
      "loss": 0.7907142,
      "num_input_tokens_seen": 125440115,
      "step": 5828,
      "time_per_iteration": 2.721036911010742
    },
    {
      "auxiliary_loss_clip": 0.01153677,
      "auxiliary_loss_mlp": 0.01022113,
      "balance_loss_clip": 1.04471982,
      "balance_loss_mlp": 1.01578295,
      "epoch": 0.7008958095352612,
      "flos": 18149899858560.0,
      "grad_norm": 1.7028636290233696,
      "language_loss": 0.85258543,
      "learning_rate": 8.671768266586228e-07,
      "loss": 0.87434328,
      "num_input_tokens_seen": 125458240,
      "step": 5829,
      "time_per_iteration": 3.572343587875366
    },
    {
      "auxiliary_loss_clip": 0.01122814,
      "auxiliary_loss_mlp": 0.01025639,
      "balance_loss_clip": 1.04252052,
      "balance_loss_mlp": 1.01788199,
      "epoch": 0.7010160524259004,
      "flos": 27452307173760.0,
      "grad_norm": 4.915593312020079,
      "language_loss": 0.7816819,
      "learning_rate": 8.665349445284275e-07,
      "loss": 0.80316639,
      "num_input_tokens_seen": 125477980,
      "step": 5830,
      "time_per_iteration": 2.7273168563842773
    },
    {
      "auxiliary_loss_clip": 0.01124698,
      "auxiliary_loss_mlp": 0.01023495,
      "balance_loss_clip": 1.04285645,
      "balance_loss_mlp": 1.01634812,
      "epoch": 0.7011362953165394,
      "flos": 23842064125440.0,
      "grad_norm": 1.4805349447164358,
      "language_loss": 0.80934608,
      "learning_rate": 8.658932343439799e-07,
      "loss": 0.83082795,
      "num_input_tokens_seen": 125497765,
      "step": 5831,
      "time_per_iteration": 2.785181999206543
    },
    {
      "auxiliary_loss_clip": 0.01170172,
      "auxiliary_loss_mlp": 0.0102438,
      "balance_loss_clip": 1.04803717,
      "balance_loss_mlp": 1.01657152,
      "epoch": 0.7012565382071785,
      "flos": 24823430582400.0,
      "grad_norm": 2.2613592535809577,
      "language_loss": 0.77758253,
      "learning_rate": 8.65251696202627e-07,
      "loss": 0.79952812,
      "num_input_tokens_seen": 125514145,
      "step": 5832,
      "time_per_iteration": 3.566073179244995
    },
    {
      "auxiliary_loss_clip": 0.01130014,
      "auxiliary_loss_mlp": 0.01023245,
      "balance_loss_clip": 1.04446745,
      "balance_loss_mlp": 1.01593781,
      "epoch": 0.7013767810978175,
      "flos": 21397445326080.0,
      "grad_norm": 2.131440642551645,
      "language_loss": 0.87507391,
      "learning_rate": 8.646103302016896e-07,
      "loss": 0.89660656,
      "num_input_tokens_seen": 125533115,
      "step": 5833,
      "time_per_iteration": 3.627340316772461
    },
    {
      "auxiliary_loss_clip": 0.01123225,
      "auxiliary_loss_mlp": 0.01025969,
      "balance_loss_clip": 1.04126978,
      "balance_loss_mlp": 1.01868534,
      "epoch": 0.7014970239884567,
      "flos": 16687150306560.0,
      "grad_norm": 1.7278180318166494,
      "language_loss": 0.88248926,
      "learning_rate": 8.639691364384614e-07,
      "loss": 0.90398121,
      "num_input_tokens_seen": 125550740,
      "step": 5834,
      "time_per_iteration": 2.683443784713745
    },
    {
      "auxiliary_loss_clip": 0.01144281,
      "auxiliary_loss_mlp": 0.01024027,
      "balance_loss_clip": 1.04627287,
      "balance_loss_mlp": 1.01663637,
      "epoch": 0.7016172668790958,
      "flos": 12568268718720.0,
      "grad_norm": 2.014697372318599,
      "language_loss": 0.7288866,
      "learning_rate": 8.633281150102136e-07,
      "loss": 0.7505697,
      "num_input_tokens_seen": 125567590,
      "step": 5835,
      "time_per_iteration": 2.693934917449951
    },
    {
      "auxiliary_loss_clip": 0.01142431,
      "auxiliary_loss_mlp": 0.01028516,
      "balance_loss_clip": 1.04527497,
      "balance_loss_mlp": 1.02094293,
      "epoch": 0.7017375097697348,
      "flos": 17452729808640.0,
      "grad_norm": 2.2699714556555444,
      "language_loss": 0.6794824,
      "learning_rate": 8.626872660141855e-07,
      "loss": 0.70119178,
      "num_input_tokens_seen": 125585500,
      "step": 5836,
      "time_per_iteration": 2.721418857574463
    },
    {
      "auxiliary_loss_clip": 0.01117231,
      "auxiliary_loss_mlp": 0.01024436,
      "balance_loss_clip": 1.04483414,
      "balance_loss_mlp": 1.01737857,
      "epoch": 0.701857752660374,
      "flos": 18513028402560.0,
      "grad_norm": 1.6542522655245253,
      "language_loss": 0.74649477,
      "learning_rate": 8.620465895475957e-07,
      "loss": 0.76791137,
      "num_input_tokens_seen": 125603720,
      "step": 5837,
      "time_per_iteration": 3.7478787899017334
    },
    {
      "auxiliary_loss_clip": 0.01109476,
      "auxiliary_loss_mlp": 0.01026408,
      "balance_loss_clip": 1.04361677,
      "balance_loss_mlp": 1.01927304,
      "epoch": 0.701977995551013,
      "flos": 24425971614720.0,
      "grad_norm": 1.5190096587592972,
      "language_loss": 0.75205833,
      "learning_rate": 8.614060857076333e-07,
      "loss": 0.77341712,
      "num_input_tokens_seen": 125624390,
      "step": 5838,
      "time_per_iteration": 2.7823619842529297
    },
    {
      "auxiliary_loss_clip": 0.01138322,
      "auxiliary_loss_mlp": 0.01029501,
      "balance_loss_clip": 1.04507732,
      "balance_loss_mlp": 1.02212763,
      "epoch": 0.7020982384416521,
      "flos": 23002759958400.0,
      "grad_norm": 1.767495094189689,
      "language_loss": 0.74609023,
      "learning_rate": 8.60765754591462e-07,
      "loss": 0.76776844,
      "num_input_tokens_seen": 125644085,
      "step": 5839,
      "time_per_iteration": 2.6580851078033447
    },
    {
      "auxiliary_loss_clip": 0.01168747,
      "auxiliary_loss_mlp": 0.01029235,
      "balance_loss_clip": 1.04737425,
      "balance_loss_mlp": 1.02177274,
      "epoch": 0.7022184813322913,
      "flos": 20449080489600.0,
      "grad_norm": 1.9687074430062752,
      "language_loss": 0.72819912,
      "learning_rate": 8.601255962962211e-07,
      "loss": 0.75017893,
      "num_input_tokens_seen": 125663095,
      "step": 5840,
      "time_per_iteration": 2.6684701442718506
    },
    {
      "auxiliary_loss_clip": 0.01163309,
      "auxiliary_loss_mlp": 0.01028036,
      "balance_loss_clip": 1.04874396,
      "balance_loss_mlp": 1.01892257,
      "epoch": 0.7023387242229303,
      "flos": 19790514581760.0,
      "grad_norm": 2.5812301742349635,
      "language_loss": 0.7231406,
      "learning_rate": 8.594856109190194e-07,
      "loss": 0.74505407,
      "num_input_tokens_seen": 125680125,
      "step": 5841,
      "time_per_iteration": 2.6279587745666504
    },
    {
      "auxiliary_loss_clip": 0.01170666,
      "auxiliary_loss_mlp": 0.01024908,
      "balance_loss_clip": 1.04720187,
      "balance_loss_mlp": 1.01757622,
      "epoch": 0.7024589671135694,
      "flos": 33259278286080.0,
      "grad_norm": 1.7449815744445807,
      "language_loss": 0.69390082,
      "learning_rate": 8.588457985569446e-07,
      "loss": 0.71585649,
      "num_input_tokens_seen": 125703035,
      "step": 5842,
      "time_per_iteration": 2.800624132156372
    },
    {
      "auxiliary_loss_clip": 0.01172195,
      "auxiliary_loss_mlp": 0.01032321,
      "balance_loss_clip": 1.04844499,
      "balance_loss_mlp": 1.02478993,
      "epoch": 0.7025792100042085,
      "flos": 19098982967040.0,
      "grad_norm": 2.0849110612225608,
      "language_loss": 0.7188251,
      "learning_rate": 8.582061593070542e-07,
      "loss": 0.74087024,
      "num_input_tokens_seen": 125723765,
      "step": 5843,
      "time_per_iteration": 2.6414637565612793
    },
    {
      "auxiliary_loss_clip": 0.01171935,
      "auxiliary_loss_mlp": 0.00762542,
      "balance_loss_clip": 1.04913831,
      "balance_loss_mlp": 1.00046539,
      "epoch": 0.7026994528948476,
      "flos": 18952611045120.0,
      "grad_norm": 2.412183104473722,
      "language_loss": 0.77347505,
      "learning_rate": 8.57566693266383e-07,
      "loss": 0.79281974,
      "num_input_tokens_seen": 125741455,
      "step": 5844,
      "time_per_iteration": 2.5886712074279785
    },
    {
      "auxiliary_loss_clip": 0.01147597,
      "auxiliary_loss_mlp": 0.00762934,
      "balance_loss_clip": 1.0435127,
      "balance_loss_mlp": 1.0004884,
      "epoch": 0.7028196957854866,
      "flos": 19536662188800.0,
      "grad_norm": 1.9245577172126884,
      "language_loss": 0.69245851,
      "learning_rate": 8.569274005319354e-07,
      "loss": 0.71156383,
      "num_input_tokens_seen": 125759855,
      "step": 5845,
      "time_per_iteration": 2.6440470218658447
    },
    {
      "auxiliary_loss_clip": 0.01152886,
      "auxiliary_loss_mlp": 0.01027834,
      "balance_loss_clip": 1.04707456,
      "balance_loss_mlp": 1.02061629,
      "epoch": 0.7029399386761258,
      "flos": 20845318394880.0,
      "grad_norm": 1.750962480784786,
      "language_loss": 0.79891837,
      "learning_rate": 8.562882812006913e-07,
      "loss": 0.82072562,
      "num_input_tokens_seen": 125777345,
      "step": 5846,
      "time_per_iteration": 2.6221730709075928
    },
    {
      "auxiliary_loss_clip": 0.01169619,
      "auxiliary_loss_mlp": 0.01023082,
      "balance_loss_clip": 1.04833746,
      "balance_loss_mlp": 1.0158664,
      "epoch": 0.7030601815667649,
      "flos": 22055005653120.0,
      "grad_norm": 1.645634564858331,
      "language_loss": 0.77131832,
      "learning_rate": 8.556493353696066e-07,
      "loss": 0.79324532,
      "num_input_tokens_seen": 125796345,
      "step": 5847,
      "time_per_iteration": 2.5985822677612305
    },
    {
      "auxiliary_loss_clip": 0.01161449,
      "auxiliary_loss_mlp": 0.0076296,
      "balance_loss_clip": 1.04849291,
      "balance_loss_mlp": 1.00050879,
      "epoch": 0.7031804244574039,
      "flos": 27198742089600.0,
      "grad_norm": 6.164350309087768,
      "language_loss": 0.68148714,
      "learning_rate": 8.550105631356077e-07,
      "loss": 0.70073122,
      "num_input_tokens_seen": 125816070,
      "step": 5848,
      "time_per_iteration": 2.688530206680298
    },
    {
      "auxiliary_loss_clip": 0.01124416,
      "auxiliary_loss_mlp": 0.0102287,
      "balance_loss_clip": 1.04174078,
      "balance_loss_mlp": 1.01488328,
      "epoch": 0.7033006673480431,
      "flos": 22379853277440.0,
      "grad_norm": 1.9301445314710028,
      "language_loss": 0.77105927,
      "learning_rate": 8.543719645955961e-07,
      "loss": 0.79253209,
      "num_input_tokens_seen": 125834400,
      "step": 5849,
      "time_per_iteration": 2.7539658546447754
    },
    {
      "auxiliary_loss_clip": 0.01141308,
      "auxiliary_loss_mlp": 0.01022211,
      "balance_loss_clip": 1.04370952,
      "balance_loss_mlp": 1.01492453,
      "epoch": 0.7034209102386821,
      "flos": 24715986024960.0,
      "grad_norm": 1.6525937127209396,
      "language_loss": 0.7470383,
      "learning_rate": 8.537335398464467e-07,
      "loss": 0.76867348,
      "num_input_tokens_seen": 125854720,
      "step": 5850,
      "time_per_iteration": 2.7098097801208496
    },
    {
      "auxiliary_loss_clip": 0.01141042,
      "auxiliary_loss_mlp": 0.01026274,
      "balance_loss_clip": 1.04165411,
      "balance_loss_mlp": 1.01792979,
      "epoch": 0.7035411531293212,
      "flos": 22556174163840.0,
      "grad_norm": 4.396330511606085,
      "language_loss": 0.85512853,
      "learning_rate": 8.53095288985007e-07,
      "loss": 0.87680161,
      "num_input_tokens_seen": 125868455,
      "step": 5851,
      "time_per_iteration": 2.7180612087249756
    },
    {
      "auxiliary_loss_clip": 0.01165986,
      "auxiliary_loss_mlp": 0.01022896,
      "balance_loss_clip": 1.04614854,
      "balance_loss_mlp": 1.01604724,
      "epoch": 0.7036613960199604,
      "flos": 22674967418880.0,
      "grad_norm": 1.7456832123705772,
      "language_loss": 0.82543087,
      "learning_rate": 8.524572121081009e-07,
      "loss": 0.84731972,
      "num_input_tokens_seen": 125888555,
      "step": 5852,
      "time_per_iteration": 2.6090939044952393
    },
    {
      "auxiliary_loss_clip": 0.01158943,
      "auxiliary_loss_mlp": 0.01026124,
      "balance_loss_clip": 1.04509211,
      "balance_loss_mlp": 1.01829195,
      "epoch": 0.7037816389105994,
      "flos": 22492146170880.0,
      "grad_norm": 2.013494390773777,
      "language_loss": 0.62418544,
      "learning_rate": 8.518193093125232e-07,
      "loss": 0.64603609,
      "num_input_tokens_seen": 125907610,
      "step": 5853,
      "time_per_iteration": 2.684634208679199
    },
    {
      "auxiliary_loss_clip": 0.01144724,
      "auxiliary_loss_mlp": 0.01027009,
      "balance_loss_clip": 1.04435921,
      "balance_loss_mlp": 1.01996326,
      "epoch": 0.7039018818012385,
      "flos": 27087490690560.0,
      "grad_norm": 1.7444981213270008,
      "language_loss": 0.81068814,
      "learning_rate": 8.511815806950436e-07,
      "loss": 0.83240545,
      "num_input_tokens_seen": 125928640,
      "step": 5854,
      "time_per_iteration": 2.7879602909088135
    },
    {
      "auxiliary_loss_clip": 0.01150203,
      "auxiliary_loss_mlp": 0.01021506,
      "balance_loss_clip": 1.04173636,
      "balance_loss_mlp": 1.01486897,
      "epoch": 0.7040221246918776,
      "flos": 17749819198080.0,
      "grad_norm": 1.6934481351204795,
      "language_loss": 0.78014159,
      "learning_rate": 8.505440263524044e-07,
      "loss": 0.80185866,
      "num_input_tokens_seen": 125947485,
      "step": 5855,
      "time_per_iteration": 3.5706064701080322
    },
    {
      "auxiliary_loss_clip": 0.01159561,
      "auxiliary_loss_mlp": 0.01027849,
      "balance_loss_clip": 1.04597998,
      "balance_loss_mlp": 1.01992774,
      "epoch": 0.7041423675825167,
      "flos": 16279851012480.0,
      "grad_norm": 2.653077124902142,
      "language_loss": 0.88090879,
      "learning_rate": 8.49906646381322e-07,
      "loss": 0.90278292,
      "num_input_tokens_seen": 125960320,
      "step": 5856,
      "time_per_iteration": 2.6576919555664062
    },
    {
      "auxiliary_loss_clip": 0.01129434,
      "auxiliary_loss_mlp": 0.01024553,
      "balance_loss_clip": 1.04429448,
      "balance_loss_mlp": 1.01781225,
      "epoch": 0.7042626104731557,
      "flos": 25483181639040.0,
      "grad_norm": 1.7837947977575717,
      "language_loss": 0.7218529,
      "learning_rate": 8.492694408784884e-07,
      "loss": 0.74339271,
      "num_input_tokens_seen": 125980575,
      "step": 5857,
      "time_per_iteration": 2.720989465713501
    },
    {
      "auxiliary_loss_clip": 0.01160323,
      "auxiliary_loss_mlp": 0.01023834,
      "balance_loss_clip": 1.04695225,
      "balance_loss_mlp": 1.01656532,
      "epoch": 0.7043828533637949,
      "flos": 17857622891520.0,
      "grad_norm": 2.292731365337256,
      "language_loss": 0.62001485,
      "learning_rate": 8.486324099405642e-07,
      "loss": 0.64185643,
      "num_input_tokens_seen": 125997420,
      "step": 5858,
      "time_per_iteration": 3.5986831188201904
    },
    {
      "auxiliary_loss_clip": 0.0115294,
      "auxiliary_loss_mlp": 0.01025373,
      "balance_loss_clip": 1.04388487,
      "balance_loss_mlp": 1.01864934,
      "epoch": 0.704503096254434,
      "flos": 29494259533440.0,
      "grad_norm": 1.7486482528287213,
      "language_loss": 0.74658942,
      "learning_rate": 8.479955536641887e-07,
      "loss": 0.76837254,
      "num_input_tokens_seen": 126018915,
      "step": 5859,
      "time_per_iteration": 2.665661573410034
    },
    {
      "auxiliary_loss_clip": 0.01133301,
      "auxiliary_loss_mlp": 0.01026881,
      "balance_loss_clip": 1.040308,
      "balance_loss_mlp": 1.01982427,
      "epoch": 0.704623339145073,
      "flos": 30920739327360.0,
      "grad_norm": 2.8514414345908627,
      "language_loss": 0.66059577,
      "learning_rate": 8.473588721459716e-07,
      "loss": 0.68219757,
      "num_input_tokens_seen": 126038825,
      "step": 5860,
      "time_per_iteration": 3.6199865341186523
    },
    {
      "auxiliary_loss_clip": 0.01160107,
      "auxiliary_loss_mlp": 0.01030342,
      "balance_loss_clip": 1.05123973,
      "balance_loss_mlp": 1.0226028,
      "epoch": 0.7047435820357122,
      "flos": 23914747296000.0,
      "grad_norm": 2.2429486087456247,
      "language_loss": 0.7075417,
      "learning_rate": 8.467223654824967e-07,
      "loss": 0.72944617,
      "num_input_tokens_seen": 126058280,
      "step": 5861,
      "time_per_iteration": 2.6557364463806152
    },
    {
      "auxiliary_loss_clip": 0.01151159,
      "auxiliary_loss_mlp": 0.01023765,
      "balance_loss_clip": 1.04485238,
      "balance_loss_mlp": 1.01655626,
      "epoch": 0.7048638249263512,
      "flos": 46494010926720.0,
      "grad_norm": 1.8243827265668355,
      "language_loss": 0.62648475,
      "learning_rate": 8.460860337703233e-07,
      "loss": 0.64823401,
      "num_input_tokens_seen": 126078885,
      "step": 5862,
      "time_per_iteration": 2.810192584991455
    },
    {
      "auxiliary_loss_clip": 0.01114493,
      "auxiliary_loss_mlp": 0.01024667,
      "balance_loss_clip": 1.04057539,
      "balance_loss_mlp": 1.01715064,
      "epoch": 0.7049840678169903,
      "flos": 21689219502720.0,
      "grad_norm": 1.8414970453165826,
      "language_loss": 0.70628595,
      "learning_rate": 8.454498771059797e-07,
      "loss": 0.72767752,
      "num_input_tokens_seen": 126098260,
      "step": 5863,
      "time_per_iteration": 3.5392227172851562
    },
    {
      "auxiliary_loss_clip": 0.01105133,
      "auxiliary_loss_mlp": 0.010249,
      "balance_loss_clip": 1.04005861,
      "balance_loss_mlp": 1.01737177,
      "epoch": 0.7051043107076294,
      "flos": 18405081054720.0,
      "grad_norm": 2.0514093426084763,
      "language_loss": 0.83088732,
      "learning_rate": 8.448138955859725e-07,
      "loss": 0.85218763,
      "num_input_tokens_seen": 126114845,
      "step": 5864,
      "time_per_iteration": 2.888139009475708
    },
    {
      "auxiliary_loss_clip": 0.01142535,
      "auxiliary_loss_mlp": 0.01030837,
      "balance_loss_clip": 1.04514623,
      "balance_loss_mlp": 1.02293313,
      "epoch": 0.7052245535982685,
      "flos": 19319043640320.0,
      "grad_norm": 1.780424236066068,
      "language_loss": 0.90012884,
      "learning_rate": 8.44178089306778e-07,
      "loss": 0.9218626,
      "num_input_tokens_seen": 126132780,
      "step": 5865,
      "time_per_iteration": 2.635498523712158
    },
    {
      "auxiliary_loss_clip": 0.01170885,
      "auxiliary_loss_mlp": 0.01027926,
      "balance_loss_clip": 1.04911113,
      "balance_loss_mlp": 1.02055311,
      "epoch": 0.7053447964889076,
      "flos": 19062138591360.0,
      "grad_norm": 1.883842355512516,
      "language_loss": 0.77147472,
      "learning_rate": 8.4354245836485e-07,
      "loss": 0.79346281,
      "num_input_tokens_seen": 126151225,
      "step": 5866,
      "time_per_iteration": 2.5778565406799316
    },
    {
      "auxiliary_loss_clip": 0.01128326,
      "auxiliary_loss_mlp": 0.01026636,
      "balance_loss_clip": 1.04216361,
      "balance_loss_mlp": 1.01914394,
      "epoch": 0.7054650393795466,
      "flos": 27379228953600.0,
      "grad_norm": 1.5852951430768631,
      "language_loss": 0.72805226,
      "learning_rate": 8.429070028566108e-07,
      "loss": 0.74960184,
      "num_input_tokens_seen": 126172535,
      "step": 5867,
      "time_per_iteration": 2.744642496109009
    },
    {
      "auxiliary_loss_clip": 0.01153808,
      "auxiliary_loss_mlp": 0.01026631,
      "balance_loss_clip": 1.04501021,
      "balance_loss_mlp": 1.01926708,
      "epoch": 0.7055852822701858,
      "flos": 16102201322880.0,
      "grad_norm": 3.1888303874861523,
      "language_loss": 0.7478019,
      "learning_rate": 8.422717228784586e-07,
      "loss": 0.76960635,
      "num_input_tokens_seen": 126189410,
      "step": 5868,
      "time_per_iteration": 2.6816253662109375
    },
    {
      "auxiliary_loss_clip": 0.01115717,
      "auxiliary_loss_mlp": 0.01030291,
      "balance_loss_clip": 1.04673147,
      "balance_loss_mlp": 1.02254224,
      "epoch": 0.7057055251608249,
      "flos": 11692299744000.0,
      "grad_norm": 1.847081147716386,
      "language_loss": 0.69143838,
      "learning_rate": 8.416366185267663e-07,
      "loss": 0.71289849,
      "num_input_tokens_seen": 126206910,
      "step": 5869,
      "time_per_iteration": 2.6830389499664307
    },
    {
      "auxiliary_loss_clip": 0.01154564,
      "auxiliary_loss_mlp": 0.01026024,
      "balance_loss_clip": 1.04328108,
      "balance_loss_mlp": 1.01870191,
      "epoch": 0.7058257680514639,
      "flos": 22711560399360.0,
      "grad_norm": 1.630554820429009,
      "language_loss": 0.77729976,
      "learning_rate": 8.410016898978778e-07,
      "loss": 0.79910564,
      "num_input_tokens_seen": 126224385,
      "step": 5870,
      "time_per_iteration": 2.6357765197753906
    },
    {
      "auxiliary_loss_clip": 0.01110684,
      "auxiliary_loss_mlp": 0.01026743,
      "balance_loss_clip": 1.04265964,
      "balance_loss_mlp": 1.01990581,
      "epoch": 0.7059460109421031,
      "flos": 17529543043200.0,
      "grad_norm": 1.838974209451696,
      "language_loss": 0.79101813,
      "learning_rate": 8.403669370881115e-07,
      "loss": 0.81239235,
      "num_input_tokens_seen": 126243120,
      "step": 5871,
      "time_per_iteration": 2.837751865386963
    },
    {
      "auxiliary_loss_clip": 0.01170428,
      "auxiliary_loss_mlp": 0.01022,
      "balance_loss_clip": 1.04834175,
      "balance_loss_mlp": 1.01480556,
      "epoch": 0.7060662538327421,
      "flos": 23544687427200.0,
      "grad_norm": 1.6223582976913804,
      "language_loss": 0.78661394,
      "learning_rate": 8.397323601937587e-07,
      "loss": 0.8085382,
      "num_input_tokens_seen": 126263020,
      "step": 5872,
      "time_per_iteration": 2.6784138679504395
    },
    {
      "auxiliary_loss_clip": 0.01123155,
      "auxiliary_loss_mlp": 0.0102482,
      "balance_loss_clip": 1.04221499,
      "balance_loss_mlp": 1.01781607,
      "epoch": 0.7061864967233812,
      "flos": 30260736875520.0,
      "grad_norm": 1.8830647329523706,
      "language_loss": 0.77254868,
      "learning_rate": 8.390979593110838e-07,
      "loss": 0.7940284,
      "num_input_tokens_seen": 126285150,
      "step": 5873,
      "time_per_iteration": 2.7504963874816895
    },
    {
      "auxiliary_loss_clip": 0.0114814,
      "auxiliary_loss_mlp": 0.0102484,
      "balance_loss_clip": 1.04814827,
      "balance_loss_mlp": 1.01747894,
      "epoch": 0.7063067396140204,
      "flos": 20701460424960.0,
      "grad_norm": 1.7717292050157025,
      "language_loss": 0.81932139,
      "learning_rate": 8.384637345363262e-07,
      "loss": 0.84105116,
      "num_input_tokens_seen": 126304340,
      "step": 5874,
      "time_per_iteration": 2.678549289703369
    },
    {
      "auxiliary_loss_clip": 0.0113391,
      "auxiliary_loss_mlp": 0.01022322,
      "balance_loss_clip": 1.04022741,
      "balance_loss_mlp": 1.01479959,
      "epoch": 0.7064269825046594,
      "flos": 32266168081920.0,
      "grad_norm": 1.788069109942293,
      "language_loss": 0.76440012,
      "learning_rate": 8.378296859656964e-07,
      "loss": 0.7859624,
      "num_input_tokens_seen": 126325495,
      "step": 5875,
      "time_per_iteration": 2.8381645679473877
    },
    {
      "auxiliary_loss_clip": 0.01142851,
      "auxiliary_loss_mlp": 0.01021242,
      "balance_loss_clip": 1.04486203,
      "balance_loss_mlp": 1.01406264,
      "epoch": 0.7065472253952985,
      "flos": 30227124723840.0,
      "grad_norm": 6.416503624459208,
      "language_loss": 0.68454492,
      "learning_rate": 8.371958136953792e-07,
      "loss": 0.70618582,
      "num_input_tokens_seen": 126345525,
      "step": 5876,
      "time_per_iteration": 2.7895352840423584
    },
    {
      "auxiliary_loss_clip": 0.01132372,
      "auxiliary_loss_mlp": 0.01025826,
      "balance_loss_clip": 1.04298449,
      "balance_loss_mlp": 1.01791048,
      "epoch": 0.7066674682859376,
      "flos": 16216720859520.0,
      "grad_norm": 2.3377377694281836,
      "language_loss": 0.66504216,
      "learning_rate": 8.365621178215326e-07,
      "loss": 0.68662411,
      "num_input_tokens_seen": 126361995,
      "step": 5877,
      "time_per_iteration": 2.676149368286133
    },
    {
      "auxiliary_loss_clip": 0.01150474,
      "auxiliary_loss_mlp": 0.01026495,
      "balance_loss_clip": 1.0455296,
      "balance_loss_mlp": 1.01901507,
      "epoch": 0.7067877111765767,
      "flos": 14830461319680.0,
      "grad_norm": 2.410498897204674,
      "language_loss": 0.74916649,
      "learning_rate": 8.359285984402871e-07,
      "loss": 0.77093613,
      "num_input_tokens_seen": 126379260,
      "step": 5878,
      "time_per_iteration": 2.6173806190490723
    },
    {
      "auxiliary_loss_clip": 0.01140349,
      "auxiliary_loss_mlp": 0.01022377,
      "balance_loss_clip": 1.04710162,
      "balance_loss_mlp": 1.01585007,
      "epoch": 0.7069079540672157,
      "flos": 25440196037760.0,
      "grad_norm": 1.8166273628178795,
      "language_loss": 0.73939347,
      "learning_rate": 8.352952556477489e-07,
      "loss": 0.76102066,
      "num_input_tokens_seen": 126397170,
      "step": 5879,
      "time_per_iteration": 2.7111949920654297
    },
    {
      "auxiliary_loss_clip": 0.01155551,
      "auxiliary_loss_mlp": 0.01023513,
      "balance_loss_clip": 1.04740596,
      "balance_loss_mlp": 1.01584828,
      "epoch": 0.7070281969578549,
      "flos": 24607751368320.0,
      "grad_norm": 1.7274105825957071,
      "language_loss": 0.76615769,
      "learning_rate": 8.34662089539993e-07,
      "loss": 0.78794837,
      "num_input_tokens_seen": 126416680,
      "step": 5880,
      "time_per_iteration": 2.671912431716919
    },
    {
      "auxiliary_loss_clip": 0.0116726,
      "auxiliary_loss_mlp": 0.01029458,
      "balance_loss_clip": 1.0473212,
      "balance_loss_mlp": 1.02271068,
      "epoch": 0.707148439848494,
      "flos": 26724469887360.0,
      "grad_norm": 1.901765313120718,
      "language_loss": 0.79131633,
      "learning_rate": 8.340291002130722e-07,
      "loss": 0.8132835,
      "num_input_tokens_seen": 126435870,
      "step": 5881,
      "time_per_iteration": 3.5224192142486572
    },
    {
      "auxiliary_loss_clip": 0.01174361,
      "auxiliary_loss_mlp": 0.0102695,
      "balance_loss_clip": 1.04982519,
      "balance_loss_mlp": 1.01905274,
      "epoch": 0.707268682739133,
      "flos": 15085750256640.0,
      "grad_norm": 2.5281513926367496,
      "language_loss": 0.79943204,
      "learning_rate": 8.3339628776301e-07,
      "loss": 0.82144511,
      "num_input_tokens_seen": 126454010,
      "step": 5882,
      "time_per_iteration": 2.604020357131958
    },
    {
      "auxiliary_loss_clip": 0.01167813,
      "auxiliary_loss_mlp": 0.010232,
      "balance_loss_clip": 1.04697537,
      "balance_loss_mlp": 1.016083,
      "epoch": 0.7073889256297722,
      "flos": 34313148345600.0,
      "grad_norm": 2.002215409114501,
      "language_loss": 0.56875318,
      "learning_rate": 8.327636522858033e-07,
      "loss": 0.59066325,
      "num_input_tokens_seen": 126473615,
      "step": 5883,
      "time_per_iteration": 2.711421012878418
    },
    {
      "auxiliary_loss_clip": 0.01116433,
      "auxiliary_loss_mlp": 0.0102401,
      "balance_loss_clip": 1.04488218,
      "balance_loss_mlp": 1.01615429,
      "epoch": 0.7075091685204112,
      "flos": 20083940784000.0,
      "grad_norm": 2.8733439692292078,
      "language_loss": 0.76916456,
      "learning_rate": 8.321311938774225e-07,
      "loss": 0.79056895,
      "num_input_tokens_seen": 126492705,
      "step": 5884,
      "time_per_iteration": 3.744283676147461
    },
    {
      "auxiliary_loss_clip": 0.01169831,
      "auxiliary_loss_mlp": 0.01026067,
      "balance_loss_clip": 1.04607952,
      "balance_loss_mlp": 1.0192064,
      "epoch": 0.7076294114110503,
      "flos": 20777124424320.0,
      "grad_norm": 2.23620261437103,
      "language_loss": 0.79201716,
      "learning_rate": 8.314989126338104e-07,
      "loss": 0.81397617,
      "num_input_tokens_seen": 126512715,
      "step": 5885,
      "time_per_iteration": 2.61051607131958
    },
    {
      "auxiliary_loss_clip": 0.01157908,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 1.04575491,
      "balance_loss_mlp": 1.01774216,
      "epoch": 0.7077496543016895,
      "flos": 17967689141760.0,
      "grad_norm": 1.6573300311817207,
      "language_loss": 0.84372574,
      "learning_rate": 8.308668086508847e-07,
      "loss": 0.86555099,
      "num_input_tokens_seen": 126530795,
      "step": 5886,
      "time_per_iteration": 3.5022826194763184
    },
    {
      "auxiliary_loss_clip": 0.01129151,
      "auxiliary_loss_mlp": 0.01020909,
      "balance_loss_clip": 1.04029107,
      "balance_loss_mlp": 1.0135597,
      "epoch": 0.7078698971923285,
      "flos": 45478098564480.0,
      "grad_norm": 1.8759900335847886,
      "language_loss": 0.73636878,
      "learning_rate": 8.302348820245342e-07,
      "loss": 0.75786942,
      "num_input_tokens_seen": 126553360,
      "step": 5887,
      "time_per_iteration": 2.885112762451172
    },
    {
      "auxiliary_loss_clip": 0.01128825,
      "auxiliary_loss_mlp": 0.01025446,
      "balance_loss_clip": 1.04377675,
      "balance_loss_mlp": 1.01729167,
      "epoch": 0.7079901400829676,
      "flos": 26943704547840.0,
      "grad_norm": 3.1953111897607425,
      "language_loss": 0.69725049,
      "learning_rate": 8.296031328506232e-07,
      "loss": 0.71879321,
      "num_input_tokens_seen": 126573110,
      "step": 5888,
      "time_per_iteration": 2.7507333755493164
    },
    {
      "auxiliary_loss_clip": 0.01144247,
      "auxiliary_loss_mlp": 0.01022047,
      "balance_loss_clip": 1.04594433,
      "balance_loss_mlp": 1.01459336,
      "epoch": 0.7081103829736067,
      "flos": 24423206267520.0,
      "grad_norm": 2.0967192093696205,
      "language_loss": 0.75842762,
      "learning_rate": 8.289715612249857e-07,
      "loss": 0.78009057,
      "num_input_tokens_seen": 126593725,
      "step": 5889,
      "time_per_iteration": 3.623176097869873
    },
    {
      "auxiliary_loss_clip": 0.01139824,
      "auxiliary_loss_mlp": 0.01033783,
      "balance_loss_clip": 1.04530489,
      "balance_loss_mlp": 1.02636218,
      "epoch": 0.7082306258642458,
      "flos": 18543300589440.0,
      "grad_norm": 2.5743534432996515,
      "language_loss": 0.77849144,
      "learning_rate": 8.283401672434305e-07,
      "loss": 0.80022752,
      "num_input_tokens_seen": 126608950,
      "step": 5890,
      "time_per_iteration": 2.613532543182373
    },
    {
      "auxiliary_loss_clip": 0.01140755,
      "auxiliary_loss_mlp": 0.01023956,
      "balance_loss_clip": 1.04707265,
      "balance_loss_mlp": 1.01660085,
      "epoch": 0.7083508687548848,
      "flos": 23477534951040.0,
      "grad_norm": 2.0606732773709595,
      "language_loss": 0.70235968,
      "learning_rate": 8.277089510017412e-07,
      "loss": 0.72400677,
      "num_input_tokens_seen": 126629755,
      "step": 5891,
      "time_per_iteration": 2.7131686210632324
    },
    {
      "auxiliary_loss_clip": 0.01137579,
      "auxiliary_loss_mlp": 0.0102925,
      "balance_loss_clip": 1.04487658,
      "balance_loss_mlp": 1.02234232,
      "epoch": 0.708471111645524,
      "flos": 22419463000320.0,
      "grad_norm": 1.839939401594107,
      "language_loss": 0.82301378,
      "learning_rate": 8.270779125956719e-07,
      "loss": 0.84468204,
      "num_input_tokens_seen": 126650135,
      "step": 5892,
      "time_per_iteration": 2.6619436740875244
    },
    {
      "auxiliary_loss_clip": 0.011066,
      "auxiliary_loss_mlp": 0.01023856,
      "balance_loss_clip": 1.0414902,
      "balance_loss_mlp": 1.01694822,
      "epoch": 0.7085913545361631,
      "flos": 20922885815040.0,
      "grad_norm": 2.284891804788153,
      "language_loss": 0.8008424,
      "learning_rate": 8.264470521209505e-07,
      "loss": 0.82214701,
      "num_input_tokens_seen": 126668500,
      "step": 5893,
      "time_per_iteration": 2.7178423404693604
    },
    {
      "auxiliary_loss_clip": 0.01145722,
      "auxiliary_loss_mlp": 0.01029266,
      "balance_loss_clip": 1.04220271,
      "balance_loss_mlp": 1.02188158,
      "epoch": 0.7087115974268021,
      "flos": 15012384727680.0,
      "grad_norm": 2.5784025350627253,
      "language_loss": 0.76536942,
      "learning_rate": 8.258163696732785e-07,
      "loss": 0.78711927,
      "num_input_tokens_seen": 126686090,
      "step": 5894,
      "time_per_iteration": 2.6174371242523193
    },
    {
      "auxiliary_loss_clip": 0.01149106,
      "auxiliary_loss_mlp": 0.01026819,
      "balance_loss_clip": 1.04422677,
      "balance_loss_mlp": 1.01995504,
      "epoch": 0.7088318403174413,
      "flos": 21539040739200.0,
      "grad_norm": 1.9449283666471007,
      "language_loss": 0.76940322,
      "learning_rate": 8.251858653483288e-07,
      "loss": 0.79116249,
      "num_input_tokens_seen": 126704255,
      "step": 5895,
      "time_per_iteration": 2.702460289001465
    },
    {
      "auxiliary_loss_clip": 0.01157895,
      "auxiliary_loss_mlp": 0.0102679,
      "balance_loss_clip": 1.04820848,
      "balance_loss_mlp": 1.01955128,
      "epoch": 0.7089520832080803,
      "flos": 15516785462400.0,
      "grad_norm": 2.0950787089405587,
      "language_loss": 0.85921597,
      "learning_rate": 8.245555392417501e-07,
      "loss": 0.88106275,
      "num_input_tokens_seen": 126718910,
      "step": 5896,
      "time_per_iteration": 2.6082305908203125
    },
    {
      "auxiliary_loss_clip": 0.01096675,
      "auxiliary_loss_mlp": 0.01027654,
      "balance_loss_clip": 1.03717232,
      "balance_loss_mlp": 1.02061129,
      "epoch": 0.7090723260987194,
      "flos": 20412667077120.0,
      "grad_norm": 1.898690114962306,
      "language_loss": 0.78892267,
      "learning_rate": 8.239253914491613e-07,
      "loss": 0.810166,
      "num_input_tokens_seen": 126737235,
      "step": 5897,
      "time_per_iteration": 2.7123208045959473
    },
    {
      "auxiliary_loss_clip": 0.01123127,
      "auxiliary_loss_mlp": 0.01025608,
      "balance_loss_clip": 1.04294729,
      "balance_loss_mlp": 1.01774585,
      "epoch": 0.7091925689893585,
      "flos": 25668337271040.0,
      "grad_norm": 1.7255764156074929,
      "language_loss": 0.74975342,
      "learning_rate": 8.232954220661556e-07,
      "loss": 0.77124071,
      "num_input_tokens_seen": 126759970,
      "step": 5898,
      "time_per_iteration": 2.8093488216400146
    },
    {
      "auxiliary_loss_clip": 0.01170355,
      "auxiliary_loss_mlp": 0.01025523,
      "balance_loss_clip": 1.05083692,
      "balance_loss_mlp": 1.01870155,
      "epoch": 0.7093128118799976,
      "flos": 24206629213440.0,
      "grad_norm": 2.9105456312600353,
      "language_loss": 0.70237136,
      "learning_rate": 8.226656311882989e-07,
      "loss": 0.72433013,
      "num_input_tokens_seen": 126779280,
      "step": 5899,
      "time_per_iteration": 2.584559679031372
    },
    {
      "auxiliary_loss_clip": 0.01153614,
      "auxiliary_loss_mlp": 0.01025209,
      "balance_loss_clip": 1.04812288,
      "balance_loss_mlp": 1.01822639,
      "epoch": 0.7094330547706367,
      "flos": 16646786398080.0,
      "grad_norm": 2.202027468615871,
      "language_loss": 0.77109796,
      "learning_rate": 8.22036018911129e-07,
      "loss": 0.7928862,
      "num_input_tokens_seen": 126797310,
      "step": 5900,
      "time_per_iteration": 2.6311819553375244
    },
    {
      "auxiliary_loss_clip": 0.01173507,
      "auxiliary_loss_mlp": 0.01028562,
      "balance_loss_clip": 1.04822671,
      "balance_loss_mlp": 1.02049768,
      "epoch": 0.7095532976612757,
      "flos": 16283370545280.0,
      "grad_norm": 2.6782141234475025,
      "language_loss": 0.80760813,
      "learning_rate": 8.214065853301599e-07,
      "loss": 0.82962883,
      "num_input_tokens_seen": 126812840,
      "step": 5901,
      "time_per_iteration": 2.542280912399292
    },
    {
      "auxiliary_loss_clip": 0.01061038,
      "auxiliary_loss_mlp": 0.01002907,
      "balance_loss_clip": 1.01320195,
      "balance_loss_mlp": 1.00169706,
      "epoch": 0.7096735405519149,
      "flos": 70722080559360.0,
      "grad_norm": 0.8082763908702347,
      "language_loss": 0.58198375,
      "learning_rate": 8.207773305408734e-07,
      "loss": 0.60262328,
      "num_input_tokens_seen": 126880060,
      "step": 5902,
      "time_per_iteration": 3.3593337535858154
    },
    {
      "auxiliary_loss_clip": 0.01123063,
      "auxiliary_loss_mlp": 0.01026826,
      "balance_loss_clip": 1.04185033,
      "balance_loss_mlp": 1.01918447,
      "epoch": 0.709793783442554,
      "flos": 23621500661760.0,
      "grad_norm": 2.5054197816682438,
      "language_loss": 0.80036533,
      "learning_rate": 8.201482546387288e-07,
      "loss": 0.82186425,
      "num_input_tokens_seen": 126899535,
      "step": 5903,
      "time_per_iteration": 2.7340495586395264
    },
    {
      "auxiliary_loss_clip": 0.01154949,
      "auxiliary_loss_mlp": 0.01025754,
      "balance_loss_clip": 1.04708004,
      "balance_loss_mlp": 1.01866138,
      "epoch": 0.709914026333193,
      "flos": 25993472204160.0,
      "grad_norm": 1.6933203642417367,
      "language_loss": 0.91702288,
      "learning_rate": 8.195193577191553e-07,
      "loss": 0.9388299,
      "num_input_tokens_seen": 126921365,
      "step": 5904,
      "time_per_iteration": 2.6866002082824707
    },
    {
      "auxiliary_loss_clip": 0.01147093,
      "auxiliary_loss_mlp": 0.00762469,
      "balance_loss_clip": 1.04371846,
      "balance_loss_mlp": 1.00045228,
      "epoch": 0.7100342692238322,
      "flos": 24861531934080.0,
      "grad_norm": 1.6314562037770444,
      "language_loss": 0.84326196,
      "learning_rate": 8.188906398775579e-07,
      "loss": 0.86235756,
      "num_input_tokens_seen": 126941910,
      "step": 5905,
      "time_per_iteration": 2.761972188949585
    },
    {
      "auxiliary_loss_clip": 0.01169285,
      "auxiliary_loss_mlp": 0.0076245,
      "balance_loss_clip": 1.04590249,
      "balance_loss_mlp": 1.00046694,
      "epoch": 0.7101545121144712,
      "flos": 24932203943040.0,
      "grad_norm": 1.710471975434766,
      "language_loss": 0.68513906,
      "learning_rate": 8.18262101209311e-07,
      "loss": 0.70445639,
      "num_input_tokens_seen": 126961120,
      "step": 5906,
      "time_per_iteration": 2.630396842956543
    },
    {
      "auxiliary_loss_clip": 0.01157425,
      "auxiliary_loss_mlp": 0.01029922,
      "balance_loss_clip": 1.0448904,
      "balance_loss_mlp": 1.02279902,
      "epoch": 0.7102747550051103,
      "flos": 23768842250880.0,
      "grad_norm": 1.8503137651901902,
      "language_loss": 0.70115912,
      "learning_rate": 8.176337418097626e-07,
      "loss": 0.72303259,
      "num_input_tokens_seen": 126981590,
      "step": 5907,
      "time_per_iteration": 3.6312475204467773
    },
    {
      "auxiliary_loss_clip": 0.01156656,
      "auxiliary_loss_mlp": 0.00761924,
      "balance_loss_clip": 1.0475893,
      "balance_loss_mlp": 1.00051224,
      "epoch": 0.7103949978957494,
      "flos": 15303907509120.0,
      "grad_norm": 2.757161381079919,
      "language_loss": 0.79969442,
      "learning_rate": 8.170055617742364e-07,
      "loss": 0.8188802,
      "num_input_tokens_seen": 126998870,
      "step": 5908,
      "time_per_iteration": 2.679163694381714
    },
    {
      "auxiliary_loss_clip": 0.01136317,
      "auxiliary_loss_mlp": 0.01025491,
      "balance_loss_clip": 1.04399431,
      "balance_loss_mlp": 1.01798713,
      "epoch": 0.7105152407863885,
      "flos": 22638805401600.0,
      "grad_norm": 2.103260247446368,
      "language_loss": 0.7078436,
      "learning_rate": 8.163775611980252e-07,
      "loss": 0.72946167,
      "num_input_tokens_seen": 127017980,
      "step": 5909,
      "time_per_iteration": 2.7439260482788086
    },
    {
      "auxiliary_loss_clip": 0.01139283,
      "auxiliary_loss_mlp": 0.01022215,
      "balance_loss_clip": 1.04500115,
      "balance_loss_mlp": 1.01541734,
      "epoch": 0.7106354836770276,
      "flos": 17238594879360.0,
      "grad_norm": 1.8070948103970246,
      "language_loss": 0.78807652,
      "learning_rate": 8.157497401763982e-07,
      "loss": 0.80969149,
      "num_input_tokens_seen": 127035645,
      "step": 5910,
      "time_per_iteration": 3.60172438621521
    },
    {
      "auxiliary_loss_clip": 0.01154529,
      "auxiliary_loss_mlp": 0.01022287,
      "balance_loss_clip": 1.04695928,
      "balance_loss_mlp": 1.01539624,
      "epoch": 0.7107557265676667,
      "flos": 20193647898240.0,
      "grad_norm": 1.7164480080624704,
      "language_loss": 0.77956212,
      "learning_rate": 8.151220988045935e-07,
      "loss": 0.80133027,
      "num_input_tokens_seen": 127054900,
      "step": 5911,
      "time_per_iteration": 2.6132876873016357
    },
    {
      "auxiliary_loss_clip": 0.01153708,
      "auxiliary_loss_mlp": 0.01025681,
      "balance_loss_clip": 1.04624891,
      "balance_loss_mlp": 1.01847458,
      "epoch": 0.7108759694583058,
      "flos": 21507080613120.0,
      "grad_norm": 1.7352923060080183,
      "language_loss": 0.82478666,
      "learning_rate": 8.144946371778234e-07,
      "loss": 0.84658056,
      "num_input_tokens_seen": 127075010,
      "step": 5912,
      "time_per_iteration": 3.590484380722046
    },
    {
      "auxiliary_loss_clip": 0.01141794,
      "auxiliary_loss_mlp": 0.00762773,
      "balance_loss_clip": 1.04644072,
      "balance_loss_mlp": 1.00052118,
      "epoch": 0.7109962123489448,
      "flos": 24061909317120.0,
      "grad_norm": 2.047124359657653,
      "language_loss": 0.78447074,
      "learning_rate": 8.138673553912751e-07,
      "loss": 0.80351645,
      "num_input_tokens_seen": 127095570,
      "step": 5913,
      "time_per_iteration": 2.673729658126831
    },
    {
      "auxiliary_loss_clip": 0.01112955,
      "auxiliary_loss_mlp": 0.01027695,
      "balance_loss_clip": 1.04174948,
      "balance_loss_mlp": 1.02010727,
      "epoch": 0.711116455239584,
      "flos": 30480474326400.0,
      "grad_norm": 2.851097068313711,
      "language_loss": 0.57223094,
      "learning_rate": 8.132402535401059e-07,
      "loss": 0.59363741,
      "num_input_tokens_seen": 127116825,
      "step": 5914,
      "time_per_iteration": 2.793269157409668
    },
    {
      "auxiliary_loss_clip": 0.01152043,
      "auxiliary_loss_mlp": 0.01023294,
      "balance_loss_clip": 1.04684234,
      "balance_loss_mlp": 1.01674974,
      "epoch": 0.711236698130223,
      "flos": 25045610158080.0,
      "grad_norm": 1.6829144585458093,
      "language_loss": 0.74032915,
      "learning_rate": 8.126133317194465e-07,
      "loss": 0.76208258,
      "num_input_tokens_seen": 127137015,
      "step": 5915,
      "time_per_iteration": 3.5337934494018555
    },
    {
      "auxiliary_loss_clip": 0.01108594,
      "auxiliary_loss_mlp": 0.01027977,
      "balance_loss_clip": 1.03973198,
      "balance_loss_mlp": 1.0197103,
      "epoch": 0.7113569410208621,
      "flos": 24206701040640.0,
      "grad_norm": 2.8184517218380045,
      "language_loss": 0.74649179,
      "learning_rate": 8.11986590024401e-07,
      "loss": 0.76785749,
      "num_input_tokens_seen": 127156755,
      "step": 5916,
      "time_per_iteration": 2.6757571697235107
    },
    {
      "auxiliary_loss_clip": 0.01144506,
      "auxiliary_loss_mlp": 0.01027204,
      "balance_loss_clip": 1.04645419,
      "balance_loss_mlp": 1.01927018,
      "epoch": 0.7114771839115013,
      "flos": 35439306526080.0,
      "grad_norm": 1.7827660910154974,
      "language_loss": 0.68887722,
      "learning_rate": 8.113600285500442e-07,
      "loss": 0.71059442,
      "num_input_tokens_seen": 127176965,
      "step": 5917,
      "time_per_iteration": 2.7090091705322266
    },
    {
      "auxiliary_loss_clip": 0.01169557,
      "auxiliary_loss_mlp": 0.01024143,
      "balance_loss_clip": 1.04748249,
      "balance_loss_mlp": 1.0172441,
      "epoch": 0.7115974268021403,
      "flos": 21099458096640.0,
      "grad_norm": 2.0218899686022174,
      "language_loss": 0.74298537,
      "learning_rate": 8.107336473914268e-07,
      "loss": 0.76492238,
      "num_input_tokens_seen": 127195595,
      "step": 5918,
      "time_per_iteration": 2.459580659866333
    },
    {
      "auxiliary_loss_clip": 0.01047059,
      "auxiliary_loss_mlp": 0.01001011,
      "balance_loss_clip": 1.01185524,
      "balance_loss_mlp": 0.99990267,
      "epoch": 0.7117176696927794,
      "flos": 56752866616320.0,
      "grad_norm": 0.7679232797553799,
      "language_loss": 0.55732477,
      "learning_rate": 8.101074466435694e-07,
      "loss": 0.57780546,
      "num_input_tokens_seen": 127255070,
      "step": 5919,
      "time_per_iteration": 3.1003365516662598
    },
    {
      "auxiliary_loss_clip": 0.01152486,
      "auxiliary_loss_mlp": 0.01022812,
      "balance_loss_clip": 1.04656494,
      "balance_loss_mlp": 1.01515889,
      "epoch": 0.7118379125834186,
      "flos": 15925269905280.0,
      "grad_norm": 1.9849101782874348,
      "language_loss": 0.68176591,
      "learning_rate": 8.094814264014662e-07,
      "loss": 0.70351887,
      "num_input_tokens_seen": 127273825,
      "step": 5920,
      "time_per_iteration": 2.6907846927642822
    },
    {
      "auxiliary_loss_clip": 0.01171273,
      "auxiliary_loss_mlp": 0.01022419,
      "balance_loss_clip": 1.04736602,
      "balance_loss_mlp": 1.01513565,
      "epoch": 0.7119581554740576,
      "flos": 20193360589440.0,
      "grad_norm": 3.2857617322561294,
      "language_loss": 0.81235427,
      "learning_rate": 8.088555867600844e-07,
      "loss": 0.83429122,
      "num_input_tokens_seen": 127289990,
      "step": 5921,
      "time_per_iteration": 2.5715837478637695
    },
    {
      "auxiliary_loss_clip": 0.01122988,
      "auxiliary_loss_mlp": 0.0102521,
      "balance_loss_clip": 1.04130816,
      "balance_loss_mlp": 1.01872778,
      "epoch": 0.7120783983646967,
      "flos": 34715383822080.0,
      "grad_norm": 1.8127407553111445,
      "language_loss": 0.60598958,
      "learning_rate": 8.08229927814362e-07,
      "loss": 0.62747151,
      "num_input_tokens_seen": 127312880,
      "step": 5922,
      "time_per_iteration": 2.7809062004089355
    },
    {
      "auxiliary_loss_clip": 0.0112293,
      "auxiliary_loss_mlp": 0.01022849,
      "balance_loss_clip": 1.04056847,
      "balance_loss_mlp": 1.01609588,
      "epoch": 0.7121986412553358,
      "flos": 26359114700160.0,
      "grad_norm": 1.8650590095781927,
      "language_loss": 0.64797169,
      "learning_rate": 8.076044496592134e-07,
      "loss": 0.66942954,
      "num_input_tokens_seen": 127334730,
      "step": 5923,
      "time_per_iteration": 2.7348692417144775
    },
    {
      "auxiliary_loss_clip": 0.01139529,
      "auxiliary_loss_mlp": 0.01021883,
      "balance_loss_clip": 1.04551697,
      "balance_loss_mlp": 1.01435792,
      "epoch": 0.7123188841459749,
      "flos": 11145344371200.0,
      "grad_norm": 2.524536439488622,
      "language_loss": 0.78285807,
      "learning_rate": 8.069791523895204e-07,
      "loss": 0.80447221,
      "num_input_tokens_seen": 127351180,
      "step": 5924,
      "time_per_iteration": 2.5905065536499023
    },
    {
      "auxiliary_loss_clip": 0.01114704,
      "auxiliary_loss_mlp": 0.01023471,
      "balance_loss_clip": 1.03866816,
      "balance_loss_mlp": 1.01600814,
      "epoch": 0.7124391270366139,
      "flos": 20811670329600.0,
      "grad_norm": 6.841786142866126,
      "language_loss": 0.77432728,
      "learning_rate": 8.063540361001422e-07,
      "loss": 0.79570901,
      "num_input_tokens_seen": 127369750,
      "step": 5925,
      "time_per_iteration": 2.6896307468414307
    },
    {
      "auxiliary_loss_clip": 0.01121089,
      "auxiliary_loss_mlp": 0.01032219,
      "balance_loss_clip": 1.04248571,
      "balance_loss_mlp": 1.02471137,
      "epoch": 0.7125593699272531,
      "flos": 17603734584960.0,
      "grad_norm": 2.465766419307217,
      "language_loss": 0.7939074,
      "learning_rate": 8.057291008859069e-07,
      "loss": 0.81544048,
      "num_input_tokens_seen": 127387910,
      "step": 5926,
      "time_per_iteration": 2.664895534515381
    },
    {
      "auxiliary_loss_clip": 0.01151792,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.04357004,
      "balance_loss_mlp": 1.02069759,
      "epoch": 0.7126796128178922,
      "flos": 28654057526400.0,
      "grad_norm": 1.8188092895304209,
      "language_loss": 0.682046,
      "learning_rate": 8.051043468416187e-07,
      "loss": 0.70384383,
      "num_input_tokens_seen": 127409160,
      "step": 5927,
      "time_per_iteration": 2.7000789642333984
    },
    {
      "auxiliary_loss_clip": 0.01166078,
      "auxiliary_loss_mlp": 0.01024723,
      "balance_loss_clip": 1.04744303,
      "balance_loss_mlp": 1.0174036,
      "epoch": 0.7127998557085312,
      "flos": 16034438315520.0,
      "grad_norm": 1.8396817355264647,
      "language_loss": 0.8211208,
      "learning_rate": 8.044797740620506e-07,
      "loss": 0.84302884,
      "num_input_tokens_seen": 127427765,
      "step": 5928,
      "time_per_iteration": 2.573573589324951
    },
    {
      "auxiliary_loss_clip": 0.01107988,
      "auxiliary_loss_mlp": 0.01025481,
      "balance_loss_clip": 1.04360652,
      "balance_loss_mlp": 1.01835251,
      "epoch": 0.7129200985991703,
      "flos": 23403271582080.0,
      "grad_norm": 2.1656322590607457,
      "language_loss": 0.7888298,
      "learning_rate": 8.038553826419494e-07,
      "loss": 0.81016451,
      "num_input_tokens_seen": 127446475,
      "step": 5929,
      "time_per_iteration": 2.7168281078338623
    },
    {
      "auxiliary_loss_clip": 0.01165737,
      "auxiliary_loss_mlp": 0.01023815,
      "balance_loss_clip": 1.0450201,
      "balance_loss_mlp": 1.01649308,
      "epoch": 0.7130403414898094,
      "flos": 21397445326080.0,
      "grad_norm": 1.6187897262602156,
      "language_loss": 0.80862772,
      "learning_rate": 8.032311726760364e-07,
      "loss": 0.83052325,
      "num_input_tokens_seen": 127467695,
      "step": 5930,
      "time_per_iteration": 2.6366443634033203
    },
    {
      "auxiliary_loss_clip": 0.01119721,
      "auxiliary_loss_mlp": 0.01023537,
      "balance_loss_clip": 1.04200113,
      "balance_loss_mlp": 1.01580071,
      "epoch": 0.7131605843804485,
      "flos": 74739045306240.0,
      "grad_norm": 1.7620532007737026,
      "language_loss": 0.68559384,
      "learning_rate": 8.026071442590022e-07,
      "loss": 0.70702642,
      "num_input_tokens_seen": 127494590,
      "step": 5931,
      "time_per_iteration": 3.1424460411071777
    },
    {
      "auxiliary_loss_clip": 0.01156318,
      "auxiliary_loss_mlp": 0.01022984,
      "balance_loss_clip": 1.04968476,
      "balance_loss_mlp": 1.01577759,
      "epoch": 0.7132808272710875,
      "flos": 18368739469440.0,
      "grad_norm": 2.174486020010103,
      "language_loss": 0.80873203,
      "learning_rate": 8.019832974855134e-07,
      "loss": 0.83052504,
      "num_input_tokens_seen": 127512550,
      "step": 5932,
      "time_per_iteration": 2.598475933074951
    },
    {
      "auxiliary_loss_clip": 0.01127881,
      "auxiliary_loss_mlp": 0.01029972,
      "balance_loss_clip": 1.04424143,
      "balance_loss_mlp": 1.02288795,
      "epoch": 0.7134010701617267,
      "flos": 23253380127360.0,
      "grad_norm": 2.459369121151865,
      "language_loss": 0.8229093,
      "learning_rate": 8.013596324502052e-07,
      "loss": 0.84448785,
      "num_input_tokens_seen": 127531015,
      "step": 5933,
      "time_per_iteration": 3.603365659713745
    },
    {
      "auxiliary_loss_clip": 0.01149981,
      "auxiliary_loss_mlp": 0.0102105,
      "balance_loss_clip": 1.04728651,
      "balance_loss_mlp": 1.01408529,
      "epoch": 0.7135213130523658,
      "flos": 23653137565440.0,
      "grad_norm": 1.8042318350612188,
      "language_loss": 0.79032642,
      "learning_rate": 8.007361492476872e-07,
      "loss": 0.81203675,
      "num_input_tokens_seen": 127550340,
      "step": 5934,
      "time_per_iteration": 2.6694788932800293
    },
    {
      "auxiliary_loss_clip": 0.0113213,
      "auxiliary_loss_mlp": 0.01028817,
      "balance_loss_clip": 1.04175377,
      "balance_loss_mlp": 1.02148545,
      "epoch": 0.7136415559430048,
      "flos": 24790644443520.0,
      "grad_norm": 4.442838448246524,
      "language_loss": 0.793989,
      "learning_rate": 8.001128479725426e-07,
      "loss": 0.81559849,
      "num_input_tokens_seen": 127572245,
      "step": 5935,
      "time_per_iteration": 2.7471015453338623
    },
    {
      "auxiliary_loss_clip": 0.01106274,
      "auxiliary_loss_mlp": 0.01028584,
      "balance_loss_clip": 1.03985548,
      "balance_loss_mlp": 1.02197993,
      "epoch": 0.713761798833644,
      "flos": 18296954138880.0,
      "grad_norm": 1.9841931128431745,
      "language_loss": 0.81385779,
      "learning_rate": 7.994897287193248e-07,
      "loss": 0.83520633,
      "num_input_tokens_seen": 127591625,
      "step": 5936,
      "time_per_iteration": 3.6976704597473145
    },
    {
      "auxiliary_loss_clip": 0.01157485,
      "auxiliary_loss_mlp": 0.01024478,
      "balance_loss_clip": 1.04483795,
      "balance_loss_mlp": 1.01725376,
      "epoch": 0.713882041724283,
      "flos": 15558262692480.0,
      "grad_norm": 2.409587358933706,
      "language_loss": 0.83759689,
      "learning_rate": 7.988667915825605e-07,
      "loss": 0.85941648,
      "num_input_tokens_seen": 127608690,
      "step": 5937,
      "time_per_iteration": 2.660447835922241
    },
    {
      "auxiliary_loss_clip": 0.01144788,
      "auxiliary_loss_mlp": 0.01029405,
      "balance_loss_clip": 1.04724205,
      "balance_loss_mlp": 1.02207708,
      "epoch": 0.7140022846149221,
      "flos": 24061011477120.0,
      "grad_norm": 2.14795705158136,
      "language_loss": 0.75779063,
      "learning_rate": 7.982440366567491e-07,
      "loss": 0.77953255,
      "num_input_tokens_seen": 127627180,
      "step": 5938,
      "time_per_iteration": 3.5851340293884277
    },
    {
      "auxiliary_loss_clip": 0.01146946,
      "auxiliary_loss_mlp": 0.01023199,
      "balance_loss_clip": 1.043504,
      "balance_loss_mlp": 1.01619244,
      "epoch": 0.7141225275055613,
      "flos": 27891710248320.0,
      "grad_norm": 1.7926098099859056,
      "language_loss": 0.75308728,
      "learning_rate": 7.97621464036361e-07,
      "loss": 0.77478874,
      "num_input_tokens_seen": 127648940,
      "step": 5939,
      "time_per_iteration": 2.6826679706573486
    },
    {
      "auxiliary_loss_clip": 0.011573,
      "auxiliary_loss_mlp": 0.01022263,
      "balance_loss_clip": 1.04564404,
      "balance_loss_mlp": 1.01489615,
      "epoch": 0.7142427703962003,
      "flos": 19682603147520.0,
      "grad_norm": 1.6951551781152054,
      "language_loss": 0.67913485,
      "learning_rate": 7.969990738158417e-07,
      "loss": 0.70093048,
      "num_input_tokens_seen": 127667350,
      "step": 5940,
      "time_per_iteration": 3.506030321121216
    },
    {
      "auxiliary_loss_clip": 0.01158231,
      "auxiliary_loss_mlp": 0.01032495,
      "balance_loss_clip": 1.04922748,
      "balance_loss_mlp": 1.02481198,
      "epoch": 0.7143630132868394,
      "flos": 21032377447680.0,
      "grad_norm": 1.9661892906724783,
      "language_loss": 0.85181099,
      "learning_rate": 7.963768660896062e-07,
      "loss": 0.87371826,
      "num_input_tokens_seen": 127685760,
      "step": 5941,
      "time_per_iteration": 2.630401134490967
    },
    {
      "auxiliary_loss_clip": 0.01157426,
      "auxiliary_loss_mlp": 0.01024367,
      "balance_loss_clip": 1.04584062,
      "balance_loss_mlp": 1.01707125,
      "epoch": 0.7144832561774785,
      "flos": 24129923719680.0,
      "grad_norm": 6.840594556584783,
      "language_loss": 0.82687521,
      "learning_rate": 7.957548409520432e-07,
      "loss": 0.84869313,
      "num_input_tokens_seen": 127704985,
      "step": 5942,
      "time_per_iteration": 2.680835008621216
    },
    {
      "auxiliary_loss_clip": 0.01123067,
      "auxiliary_loss_mlp": 0.0102747,
      "balance_loss_clip": 1.04060268,
      "balance_loss_mlp": 1.02066875,
      "epoch": 0.7146034990681176,
      "flos": 16325817442560.0,
      "grad_norm": 1.9461005764905417,
      "language_loss": 0.84059095,
      "learning_rate": 7.951329984975135e-07,
      "loss": 0.86209631,
      "num_input_tokens_seen": 127721925,
      "step": 5943,
      "time_per_iteration": 2.6393728256225586
    },
    {
      "auxiliary_loss_clip": 0.01042877,
      "auxiliary_loss_mlp": 0.01002024,
      "balance_loss_clip": 1.01289892,
      "balance_loss_mlp": 1.00084364,
      "epoch": 0.7147237419587567,
      "flos": 69627164232960.0,
      "grad_norm": 0.7124811579652193,
      "language_loss": 0.54289871,
      "learning_rate": 7.94511338820349e-07,
      "loss": 0.56334776,
      "num_input_tokens_seen": 127784230,
      "step": 5944,
      "time_per_iteration": 3.2774412631988525
    },
    {
      "auxiliary_loss_clip": 0.0114109,
      "auxiliary_loss_mlp": 0.00762722,
      "balance_loss_clip": 1.04525328,
      "balance_loss_mlp": 1.00051308,
      "epoch": 0.7148439848493958,
      "flos": 22266806198400.0,
      "grad_norm": 1.8904318763387613,
      "language_loss": 0.78501368,
      "learning_rate": 7.938898620148575e-07,
      "loss": 0.80405176,
      "num_input_tokens_seen": 127801990,
      "step": 5945,
      "time_per_iteration": 2.7025394439697266
    },
    {
      "auxiliary_loss_clip": 0.01140603,
      "auxiliary_loss_mlp": 0.01020908,
      "balance_loss_clip": 1.0446744,
      "balance_loss_mlp": 1.01346302,
      "epoch": 0.7149642277400349,
      "flos": 17931383470080.0,
      "grad_norm": 2.130988484450793,
      "language_loss": 0.70861125,
      "learning_rate": 7.932685681753135e-07,
      "loss": 0.73022634,
      "num_input_tokens_seen": 127819270,
      "step": 5946,
      "time_per_iteration": 2.6213414669036865
    },
    {
      "auxiliary_loss_clip": 0.01166013,
      "auxiliary_loss_mlp": 0.01023166,
      "balance_loss_clip": 1.04735911,
      "balance_loss_mlp": 1.01633286,
      "epoch": 0.7150844706306739,
      "flos": 31681937370240.0,
      "grad_norm": 1.933412929800732,
      "language_loss": 0.62316501,
      "learning_rate": 7.92647457395969e-07,
      "loss": 0.64505678,
      "num_input_tokens_seen": 127841095,
      "step": 5947,
      "time_per_iteration": 2.6395583152770996
    },
    {
      "auxiliary_loss_clip": 0.01102934,
      "auxiliary_loss_mlp": 0.01029663,
      "balance_loss_clip": 1.03632379,
      "balance_loss_mlp": 1.02180099,
      "epoch": 0.7152047135213131,
      "flos": 10926217451520.0,
      "grad_norm": 2.3880072057433552,
      "language_loss": 0.73875451,
      "learning_rate": 7.920265297710444e-07,
      "loss": 0.76008046,
      "num_input_tokens_seen": 127858485,
      "step": 5948,
      "time_per_iteration": 2.6950740814208984
    },
    {
      "auxiliary_loss_clip": 0.01157048,
      "auxiliary_loss_mlp": 0.01020167,
      "balance_loss_clip": 1.04737473,
      "balance_loss_mlp": 1.01358962,
      "epoch": 0.7153249564119522,
      "flos": 20995640812800.0,
      "grad_norm": 1.8280224720273721,
      "language_loss": 0.73403251,
      "learning_rate": 7.914057853947363e-07,
      "loss": 0.75580466,
      "num_input_tokens_seen": 127877665,
      "step": 5949,
      "time_per_iteration": 2.614824056625366
    },
    {
      "auxiliary_loss_clip": 0.01121788,
      "auxiliary_loss_mlp": 0.01025135,
      "balance_loss_clip": 1.040748,
      "balance_loss_mlp": 1.01800048,
      "epoch": 0.7154451993025912,
      "flos": 24243114453120.0,
      "grad_norm": 1.8863226953057652,
      "language_loss": 0.62808287,
      "learning_rate": 7.907852243612089e-07,
      "loss": 0.64955211,
      "num_input_tokens_seen": 127898070,
      "step": 5950,
      "time_per_iteration": 2.749410629272461
    },
    {
      "auxiliary_loss_clip": 0.0113834,
      "auxiliary_loss_mlp": 0.01027133,
      "balance_loss_clip": 1.04387915,
      "balance_loss_mlp": 1.01988506,
      "epoch": 0.7155654421932304,
      "flos": 23330947547520.0,
      "grad_norm": 1.8682166625379448,
      "language_loss": 0.72102207,
      "learning_rate": 7.901648467646009e-07,
      "loss": 0.74267679,
      "num_input_tokens_seen": 127917010,
      "step": 5951,
      "time_per_iteration": 2.702850103378296
    },
    {
      "auxiliary_loss_clip": 0.01170848,
      "auxiliary_loss_mlp": 0.01022821,
      "balance_loss_clip": 1.0481391,
      "balance_loss_mlp": 1.01568353,
      "epoch": 0.7156856850838694,
      "flos": 22711883621760.0,
      "grad_norm": 1.5563958634665287,
      "language_loss": 0.72497296,
      "learning_rate": 7.895446526990244e-07,
      "loss": 0.74690974,
      "num_input_tokens_seen": 127937025,
      "step": 5952,
      "time_per_iteration": 2.5904288291931152
    },
    {
      "auxiliary_loss_clip": 0.01120202,
      "auxiliary_loss_mlp": 0.01035995,
      "balance_loss_clip": 1.04080248,
      "balance_loss_mlp": 1.02877665,
      "epoch": 0.7158059279745085,
      "flos": 19865424395520.0,
      "grad_norm": 1.6972202369004763,
      "language_loss": 0.75927198,
      "learning_rate": 7.889246422585609e-07,
      "loss": 0.78083396,
      "num_input_tokens_seen": 127956410,
      "step": 5953,
      "time_per_iteration": 2.7835564613342285
    },
    {
      "auxiliary_loss_clip": 0.01167948,
      "auxiliary_loss_mlp": 0.01026997,
      "balance_loss_clip": 1.04694104,
      "balance_loss_mlp": 1.01972568,
      "epoch": 0.7159261708651476,
      "flos": 24134772055680.0,
      "grad_norm": 2.0041077298020995,
      "language_loss": 0.73456115,
      "learning_rate": 7.883048155372675e-07,
      "loss": 0.75651062,
      "num_input_tokens_seen": 127974925,
      "step": 5954,
      "time_per_iteration": 2.6042799949645996
    },
    {
      "auxiliary_loss_clip": 0.01145187,
      "auxiliary_loss_mlp": 0.01022115,
      "balance_loss_clip": 1.04430723,
      "balance_loss_mlp": 1.01536453,
      "epoch": 0.7160464137557867,
      "flos": 16983198201600.0,
      "grad_norm": 2.1614894877926605,
      "language_loss": 0.71660602,
      "learning_rate": 7.876851726291698e-07,
      "loss": 0.73827899,
      "num_input_tokens_seen": 127993225,
      "step": 5955,
      "time_per_iteration": 2.639533042907715
    },
    {
      "auxiliary_loss_clip": 0.01129228,
      "auxiliary_loss_mlp": 0.01029034,
      "balance_loss_clip": 1.03995728,
      "balance_loss_mlp": 1.02219403,
      "epoch": 0.7161666566464258,
      "flos": 25228251838080.0,
      "grad_norm": 1.9059783187946302,
      "language_loss": 0.78292072,
      "learning_rate": 7.870657136282666e-07,
      "loss": 0.80450332,
      "num_input_tokens_seen": 128012085,
      "step": 5956,
      "time_per_iteration": 2.703115701675415
    },
    {
      "auxiliary_loss_clip": 0.01149507,
      "auxiliary_loss_mlp": 0.01024269,
      "balance_loss_clip": 1.04472268,
      "balance_loss_mlp": 1.01684833,
      "epoch": 0.7162868995370649,
      "flos": 26468390851200.0,
      "grad_norm": 1.6729137686104236,
      "language_loss": 0.82032263,
      "learning_rate": 7.86446438628531e-07,
      "loss": 0.84206039,
      "num_input_tokens_seen": 128033155,
      "step": 5957,
      "time_per_iteration": 2.6468446254730225
    },
    {
      "auxiliary_loss_clip": 0.01069234,
      "auxiliary_loss_mlp": 0.01001364,
      "balance_loss_clip": 1.0134325,
      "balance_loss_mlp": 1.00027299,
      "epoch": 0.716407142427704,
      "flos": 69998912040960.0,
      "grad_norm": 0.7641018382065691,
      "language_loss": 0.56902033,
      "learning_rate": 7.858273477239059e-07,
      "loss": 0.58972633,
      "num_input_tokens_seen": 128101575,
      "step": 5958,
      "time_per_iteration": 3.215191602706909
    },
    {
      "auxiliary_loss_clip": 0.01100212,
      "auxiliary_loss_mlp": 0.01023783,
      "balance_loss_clip": 1.03949308,
      "balance_loss_mlp": 1.01655865,
      "epoch": 0.716527385318343,
      "flos": 20740459616640.0,
      "grad_norm": 1.9060084375664466,
      "language_loss": 0.71255577,
      "learning_rate": 7.852084410083067e-07,
      "loss": 0.73379564,
      "num_input_tokens_seen": 128120395,
      "step": 5959,
      "time_per_iteration": 3.6476917266845703
    },
    {
      "auxiliary_loss_clip": 0.01134645,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 1.04458988,
      "balance_loss_mlp": 1.01947355,
      "epoch": 0.7166476282089821,
      "flos": 25371966153600.0,
      "grad_norm": 1.6330414503290143,
      "language_loss": 0.63672316,
      "learning_rate": 7.84589718575621e-07,
      "loss": 0.65832567,
      "num_input_tokens_seen": 128140840,
      "step": 5960,
      "time_per_iteration": 2.732532262802124
    },
    {
      "auxiliary_loss_clip": 0.01140272,
      "auxiliary_loss_mlp": 0.01021925,
      "balance_loss_clip": 1.03991461,
      "balance_loss_mlp": 1.0148083,
      "epoch": 0.7167678710996213,
      "flos": 24133730561280.0,
      "grad_norm": 1.9200712352608669,
      "language_loss": 0.68690026,
      "learning_rate": 7.83971180519708e-07,
      "loss": 0.7085222,
      "num_input_tokens_seen": 128159695,
      "step": 5961,
      "time_per_iteration": 2.6491539478302
    },
    {
      "auxiliary_loss_clip": 0.01171959,
      "auxiliary_loss_mlp": 0.0102436,
      "balance_loss_clip": 1.04853952,
      "balance_loss_mlp": 1.01650441,
      "epoch": 0.7168881139902603,
      "flos": 30226586019840.0,
      "grad_norm": 2.7216692236929183,
      "language_loss": 0.76240253,
      "learning_rate": 7.833528269344008e-07,
      "loss": 0.78436577,
      "num_input_tokens_seen": 128179600,
      "step": 5962,
      "time_per_iteration": 3.6375930309295654
    },
    {
      "auxiliary_loss_clip": 0.01122943,
      "auxiliary_loss_mlp": 0.01033116,
      "balance_loss_clip": 1.0431627,
      "balance_loss_mlp": 1.02523637,
      "epoch": 0.7170083568808994,
      "flos": 14606414236800.0,
      "grad_norm": 2.884749995217395,
      "language_loss": 0.77378017,
      "learning_rate": 7.827346579135023e-07,
      "loss": 0.79534072,
      "num_input_tokens_seen": 128196940,
      "step": 5963,
      "time_per_iteration": 2.7571957111358643
    },
    {
      "auxiliary_loss_clip": 0.01136467,
      "auxiliary_loss_mlp": 0.01025166,
      "balance_loss_clip": 1.04239571,
      "balance_loss_mlp": 1.01776636,
      "epoch": 0.7171285997715385,
      "flos": 23331091201920.0,
      "grad_norm": 2.231754411048932,
      "language_loss": 0.83027744,
      "learning_rate": 7.821166735507885e-07,
      "loss": 0.85189378,
      "num_input_tokens_seen": 128215970,
      "step": 5964,
      "time_per_iteration": 3.611799478530884
    },
    {
      "auxiliary_loss_clip": 0.01167917,
      "auxiliary_loss_mlp": 0.01026795,
      "balance_loss_clip": 1.0479157,
      "balance_loss_mlp": 1.01935649,
      "epoch": 0.7172488426621776,
      "flos": 16543543731840.0,
      "grad_norm": 1.6049486425254005,
      "language_loss": 0.68293786,
      "learning_rate": 7.81498873940007e-07,
      "loss": 0.70488495,
      "num_input_tokens_seen": 128233185,
      "step": 5965,
      "time_per_iteration": 2.648043632507324
    },
    {
      "auxiliary_loss_clip": 0.01160138,
      "auxiliary_loss_mlp": 0.01021966,
      "balance_loss_clip": 1.04485679,
      "balance_loss_mlp": 1.0142622,
      "epoch": 0.7173690855528166,
      "flos": 26541612725760.0,
      "grad_norm": 2.229137068378478,
      "language_loss": 0.7772367,
      "learning_rate": 7.808812591748768e-07,
      "loss": 0.79905772,
      "num_input_tokens_seen": 128253565,
      "step": 5966,
      "time_per_iteration": 3.5203135013580322
    },
    {
      "auxiliary_loss_clip": 0.01122912,
      "auxiliary_loss_mlp": 0.01025146,
      "balance_loss_clip": 1.04213619,
      "balance_loss_mlp": 1.01720667,
      "epoch": 0.7174893284434558,
      "flos": 22784099915520.0,
      "grad_norm": 2.4930044474866864,
      "language_loss": 0.65307075,
      "learning_rate": 7.802638293490915e-07,
      "loss": 0.67455125,
      "num_input_tokens_seen": 128273210,
      "step": 5967,
      "time_per_iteration": 2.6806979179382324
    },
    {
      "auxiliary_loss_clip": 0.01144379,
      "auxiliary_loss_mlp": 0.01024641,
      "balance_loss_clip": 1.0452112,
      "balance_loss_mlp": 1.01801848,
      "epoch": 0.7176095713340949,
      "flos": 23293564467840.0,
      "grad_norm": 1.614967961281152,
      "language_loss": 0.7705856,
      "learning_rate": 7.796465845563123e-07,
      "loss": 0.79227579,
      "num_input_tokens_seen": 128292085,
      "step": 5968,
      "time_per_iteration": 2.6985793113708496
    },
    {
      "auxiliary_loss_clip": 0.01136139,
      "auxiliary_loss_mlp": 0.00762136,
      "balance_loss_clip": 1.04323161,
      "balance_loss_mlp": 1.00048971,
      "epoch": 0.7177298142247339,
      "flos": 25591631777280.0,
      "grad_norm": 3.128429018381726,
      "language_loss": 0.79634362,
      "learning_rate": 7.790295248901766e-07,
      "loss": 0.81532633,
      "num_input_tokens_seen": 128313215,
      "step": 5969,
      "time_per_iteration": 2.6772968769073486
    },
    {
      "auxiliary_loss_clip": 0.01154983,
      "auxiliary_loss_mlp": 0.01023092,
      "balance_loss_clip": 1.04515159,
      "balance_loss_mlp": 1.0153079,
      "epoch": 0.7178500571153731,
      "flos": 31652778504960.0,
      "grad_norm": 1.8376642967099126,
      "language_loss": 0.62145644,
      "learning_rate": 7.784126504442902e-07,
      "loss": 0.64323717,
      "num_input_tokens_seen": 128336445,
      "step": 5970,
      "time_per_iteration": 2.7576663494110107
    },
    {
      "auxiliary_loss_clip": 0.0111631,
      "auxiliary_loss_mlp": 0.01021658,
      "balance_loss_clip": 1.04042506,
      "balance_loss_mlp": 1.01477635,
      "epoch": 0.7179703000060121,
      "flos": 19427242383360.0,
      "grad_norm": 1.3786158693642658,
      "language_loss": 0.67538601,
      "learning_rate": 7.777959613122351e-07,
      "loss": 0.69676572,
      "num_input_tokens_seen": 128356270,
      "step": 5971,
      "time_per_iteration": 2.6522915363311768
    },
    {
      "auxiliary_loss_clip": 0.01136941,
      "auxiliary_loss_mlp": 0.01022972,
      "balance_loss_clip": 1.04647112,
      "balance_loss_mlp": 1.01570654,
      "epoch": 0.7180905428966512,
      "flos": 28839249072000.0,
      "grad_norm": 1.6306388324022445,
      "language_loss": 0.77911288,
      "learning_rate": 7.771794575875604e-07,
      "loss": 0.80071199,
      "num_input_tokens_seen": 128378140,
      "step": 5972,
      "time_per_iteration": 2.6966331005096436
    },
    {
      "auxiliary_loss_clip": 0.01153122,
      "auxiliary_loss_mlp": 0.01027631,
      "balance_loss_clip": 1.04605675,
      "balance_loss_mlp": 1.02009678,
      "epoch": 0.7182107857872904,
      "flos": 20047563285120.0,
      "grad_norm": 2.395006155782625,
      "language_loss": 0.77460933,
      "learning_rate": 7.765631393637888e-07,
      "loss": 0.79641682,
      "num_input_tokens_seen": 128396335,
      "step": 5973,
      "time_per_iteration": 2.661557197570801
    },
    {
      "auxiliary_loss_clip": 0.01147998,
      "auxiliary_loss_mlp": 0.0102218,
      "balance_loss_clip": 1.04140627,
      "balance_loss_mlp": 1.01478052,
      "epoch": 0.7183310286779294,
      "flos": 22747686503040.0,
      "grad_norm": 3.005066889552014,
      "language_loss": 0.48119491,
      "learning_rate": 7.75947006734417e-07,
      "loss": 0.50289673,
      "num_input_tokens_seen": 128414115,
      "step": 5974,
      "time_per_iteration": 2.746615171432495
    },
    {
      "auxiliary_loss_clip": 0.01169315,
      "auxiliary_loss_mlp": 0.01027779,
      "balance_loss_clip": 1.04731607,
      "balance_loss_mlp": 1.020015,
      "epoch": 0.7184512715685685,
      "flos": 17158262112000.0,
      "grad_norm": 2.0763838475784233,
      "language_loss": 0.8271364,
      "learning_rate": 7.753310597929101e-07,
      "loss": 0.84910727,
      "num_input_tokens_seen": 128430755,
      "step": 5975,
      "time_per_iteration": 2.6496829986572266
    },
    {
      "auxiliary_loss_clip": 0.01068966,
      "auxiliary_loss_mlp": 0.01002303,
      "balance_loss_clip": 1.01317215,
      "balance_loss_mlp": 1.00120008,
      "epoch": 0.7185715144592076,
      "flos": 65509611448320.0,
      "grad_norm": 0.7620349630687402,
      "language_loss": 0.55517817,
      "learning_rate": 7.747152986327095e-07,
      "loss": 0.57589078,
      "num_input_tokens_seen": 128491300,
      "step": 5976,
      "time_per_iteration": 3.1072442531585693
    },
    {
      "auxiliary_loss_clip": 0.01117358,
      "auxiliary_loss_mlp": 0.01022807,
      "balance_loss_clip": 1.0421828,
      "balance_loss_mlp": 1.01572585,
      "epoch": 0.7186917573498467,
      "flos": 16180522928640.0,
      "grad_norm": 5.933660246758854,
      "language_loss": 0.6797325,
      "learning_rate": 7.740997233472228e-07,
      "loss": 0.70113415,
      "num_input_tokens_seen": 128508920,
      "step": 5977,
      "time_per_iteration": 2.7217118740081787
    },
    {
      "auxiliary_loss_clip": 0.01139399,
      "auxiliary_loss_mlp": 0.01028809,
      "balance_loss_clip": 1.04246533,
      "balance_loss_mlp": 1.0219903,
      "epoch": 0.7188120002404857,
      "flos": 29242274647680.0,
      "grad_norm": 2.1643604171114887,
      "language_loss": 0.70706111,
      "learning_rate": 7.734843340298329e-07,
      "loss": 0.72874314,
      "num_input_tokens_seen": 128528745,
      "step": 5978,
      "time_per_iteration": 2.7117276191711426
    },
    {
      "auxiliary_loss_clip": 0.01144831,
      "auxiliary_loss_mlp": 0.01027248,
      "balance_loss_clip": 1.04254472,
      "balance_loss_mlp": 1.01989901,
      "epoch": 0.7189322431311249,
      "flos": 33401161008000.0,
      "grad_norm": 2.150431130949243,
      "language_loss": 0.75061655,
      "learning_rate": 7.72869130773895e-07,
      "loss": 0.77233732,
      "num_input_tokens_seen": 128549345,
      "step": 5979,
      "time_per_iteration": 2.737966537475586
    },
    {
      "auxiliary_loss_clip": 0.01060179,
      "auxiliary_loss_mlp": 0.01003283,
      "balance_loss_clip": 1.01319802,
      "balance_loss_mlp": 1.00218678,
      "epoch": 0.719052486021764,
      "flos": 61351263792000.0,
      "grad_norm": 0.7859458574163565,
      "language_loss": 0.59318113,
      "learning_rate": 7.722541136727343e-07,
      "loss": 0.61381572,
      "num_input_tokens_seen": 128605360,
      "step": 5980,
      "time_per_iteration": 3.03460431098938
    },
    {
      "auxiliary_loss_clip": 0.01153985,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.04569125,
      "balance_loss_mlp": 1.01803398,
      "epoch": 0.719172728912403,
      "flos": 15596795007360.0,
      "grad_norm": 1.949938011943916,
      "language_loss": 0.80462837,
      "learning_rate": 7.716392828196483e-07,
      "loss": 0.82642007,
      "num_input_tokens_seen": 128623160,
      "step": 5981,
      "time_per_iteration": 2.6276376247406006
    },
    {
      "auxiliary_loss_clip": 0.01154966,
      "auxiliary_loss_mlp": 0.01026633,
      "balance_loss_clip": 1.04589462,
      "balance_loss_mlp": 1.01897979,
      "epoch": 0.7192929718030422,
      "flos": 15553162961280.0,
      "grad_norm": 2.629618249900072,
      "language_loss": 0.77091718,
      "learning_rate": 7.710246383079064e-07,
      "loss": 0.79273319,
      "num_input_tokens_seen": 128638545,
      "step": 5982,
      "time_per_iteration": 2.550541877746582
    },
    {
      "auxiliary_loss_clip": 0.01139318,
      "auxiliary_loss_mlp": 0.01026251,
      "balance_loss_clip": 1.0399313,
      "balance_loss_mlp": 1.0192709,
      "epoch": 0.7194132146936812,
      "flos": 21862487733120.0,
      "grad_norm": 2.360604047941329,
      "language_loss": 0.91996747,
      "learning_rate": 7.704101802307492e-07,
      "loss": 0.94162315,
      "num_input_tokens_seen": 128650845,
      "step": 5983,
      "time_per_iteration": 2.6126937866210938
    },
    {
      "auxiliary_loss_clip": 0.01119285,
      "auxiliary_loss_mlp": 0.01023099,
      "balance_loss_clip": 1.04180646,
      "balance_loss_mlp": 1.01589525,
      "epoch": 0.7195334575843203,
      "flos": 27338900958720.0,
      "grad_norm": 2.0484589717305823,
      "language_loss": 0.86824656,
      "learning_rate": 7.697959086813912e-07,
      "loss": 0.88967043,
      "num_input_tokens_seen": 128667010,
      "step": 5984,
      "time_per_iteration": 2.715475559234619
    },
    {
      "auxiliary_loss_clip": 0.01118543,
      "auxiliary_loss_mlp": 0.01028881,
      "balance_loss_clip": 1.03972197,
      "balance_loss_mlp": 1.02132273,
      "epoch": 0.7196537004749595,
      "flos": 18770615809920.0,
      "grad_norm": 1.6055033597454507,
      "language_loss": 0.80264378,
      "learning_rate": 7.691818237530145e-07,
      "loss": 0.82411796,
      "num_input_tokens_seen": 128685870,
      "step": 5985,
      "time_per_iteration": 3.571729898452759
    },
    {
      "auxiliary_loss_clip": 0.01125495,
      "auxiliary_loss_mlp": 0.01025537,
      "balance_loss_clip": 1.04267645,
      "balance_loss_mlp": 1.01821208,
      "epoch": 0.7197739433655985,
      "flos": 24531009960960.0,
      "grad_norm": 2.33406383515282,
      "language_loss": 0.77671707,
      "learning_rate": 7.685679255387774e-07,
      "loss": 0.79822737,
      "num_input_tokens_seen": 128704185,
      "step": 5986,
      "time_per_iteration": 2.8646109104156494
    },
    {
      "auxiliary_loss_clip": 0.01139178,
      "auxiliary_loss_mlp": 0.01024973,
      "balance_loss_clip": 1.04455137,
      "balance_loss_mlp": 1.0180831,
      "epoch": 0.7198941862562376,
      "flos": 18040587793920.0,
      "grad_norm": 2.1132427464102292,
      "language_loss": 0.77299112,
      "learning_rate": 7.679542141318065e-07,
      "loss": 0.79463261,
      "num_input_tokens_seen": 128721290,
      "step": 5987,
      "time_per_iteration": 2.621979236602783
    },
    {
      "auxiliary_loss_clip": 0.01126683,
      "auxiliary_loss_mlp": 0.01030077,
      "balance_loss_clip": 1.04073429,
      "balance_loss_mlp": 1.02268577,
      "epoch": 0.7200144291468767,
      "flos": 29022393542400.0,
      "grad_norm": 1.7292269968013068,
      "language_loss": 0.75879514,
      "learning_rate": 7.673406896252013e-07,
      "loss": 0.78036273,
      "num_input_tokens_seen": 128742665,
      "step": 5988,
      "time_per_iteration": 3.7614095211029053
    },
    {
      "auxiliary_loss_clip": 0.01121265,
      "auxiliary_loss_mlp": 0.01027625,
      "balance_loss_clip": 1.03819084,
      "balance_loss_mlp": 1.02014494,
      "epoch": 0.7201346720375158,
      "flos": 25374264624000.0,
      "grad_norm": 1.5961703625695083,
      "language_loss": 0.78503239,
      "learning_rate": 7.667273521120347e-07,
      "loss": 0.8065213,
      "num_input_tokens_seen": 128762225,
      "step": 5989,
      "time_per_iteration": 3.6551685333251953
    },
    {
      "auxiliary_loss_clip": 0.01124122,
      "auxiliary_loss_mlp": 0.01027139,
      "balance_loss_clip": 1.03928769,
      "balance_loss_mlp": 1.02028739,
      "epoch": 0.7202549149281549,
      "flos": 14355614499840.0,
      "grad_norm": 2.941981551291609,
      "language_loss": 0.79646784,
      "learning_rate": 7.661142016853468e-07,
      "loss": 0.81798047,
      "num_input_tokens_seen": 128779585,
      "step": 5990,
      "time_per_iteration": 2.676557779312134
    },
    {
      "auxiliary_loss_clip": 0.01110355,
      "auxiliary_loss_mlp": 0.01023767,
      "balance_loss_clip": 1.04020119,
      "balance_loss_mlp": 1.01605415,
      "epoch": 0.7203751578187939,
      "flos": 23001682550400.0,
      "grad_norm": 1.6814273215552205,
      "language_loss": 0.75202954,
      "learning_rate": 7.655012384381543e-07,
      "loss": 0.77337074,
      "num_input_tokens_seen": 128799070,
      "step": 5991,
      "time_per_iteration": 2.7704620361328125
    },
    {
      "auxiliary_loss_clip": 0.01139558,
      "auxiliary_loss_mlp": 0.01027916,
      "balance_loss_clip": 1.04746127,
      "balance_loss_mlp": 1.02062047,
      "epoch": 0.7204954007094331,
      "flos": 23692424065920.0,
      "grad_norm": 1.7507496862778682,
      "language_loss": 0.81863075,
      "learning_rate": 7.648884624634415e-07,
      "loss": 0.84030545,
      "num_input_tokens_seen": 128817620,
      "step": 5992,
      "time_per_iteration": 3.6613101959228516
    },
    {
      "auxiliary_loss_clip": 0.01155953,
      "auxiliary_loss_mlp": 0.01024278,
      "balance_loss_clip": 1.04830217,
      "balance_loss_mlp": 1.01714325,
      "epoch": 0.7206156436000721,
      "flos": 16253026531200.0,
      "grad_norm": 2.3642409651405725,
      "language_loss": 0.88711119,
      "learning_rate": 7.642758738541683e-07,
      "loss": 0.90891349,
      "num_input_tokens_seen": 128834200,
      "step": 5993,
      "time_per_iteration": 2.5870108604431152
    },
    {
      "auxiliary_loss_clip": 0.01059155,
      "auxiliary_loss_mlp": 0.01003118,
      "balance_loss_clip": 1.01307678,
      "balance_loss_mlp": 1.00200355,
      "epoch": 0.7207358864907112,
      "flos": 54377806504320.0,
      "grad_norm": 0.88760861742911,
      "language_loss": 0.6075542,
      "learning_rate": 7.636634727032621e-07,
      "loss": 0.62817693,
      "num_input_tokens_seen": 128891305,
      "step": 5994,
      "time_per_iteration": 3.098778009414673
    },
    {
      "auxiliary_loss_clip": 0.01125478,
      "auxiliary_loss_mlp": 0.01028337,
      "balance_loss_clip": 1.03822815,
      "balance_loss_mlp": 1.02041817,
      "epoch": 0.7208561293813504,
      "flos": 19135540033920.0,
      "grad_norm": 1.8366225370559612,
      "language_loss": 0.78431016,
      "learning_rate": 7.630512591036231e-07,
      "loss": 0.8058483,
      "num_input_tokens_seen": 128910615,
      "step": 5995,
      "time_per_iteration": 2.6821060180664062
    },
    {
      "auxiliary_loss_clip": 0.0115794,
      "auxiliary_loss_mlp": 0.01027515,
      "balance_loss_clip": 1.04851389,
      "balance_loss_mlp": 1.02022219,
      "epoch": 0.7209763722719894,
      "flos": 17748526308480.0,
      "grad_norm": 2.128604026119056,
      "language_loss": 0.6495558,
      "learning_rate": 7.624392331481255e-07,
      "loss": 0.67141032,
      "num_input_tokens_seen": 128928270,
      "step": 5996,
      "time_per_iteration": 2.692699670791626
    },
    {
      "auxiliary_loss_clip": 0.0105718,
      "auxiliary_loss_mlp": 0.01002805,
      "balance_loss_clip": 1.01301646,
      "balance_loss_mlp": 1.00168443,
      "epoch": 0.7210966151626285,
      "flos": 66819488716800.0,
      "grad_norm": 0.7379893830840482,
      "language_loss": 0.51799291,
      "learning_rate": 7.618273949296115e-07,
      "loss": 0.53859276,
      "num_input_tokens_seen": 128987780,
      "step": 5997,
      "time_per_iteration": 3.124922275543213
    },
    {
      "auxiliary_loss_clip": 0.01132661,
      "auxiliary_loss_mlp": 0.01032407,
      "balance_loss_clip": 1.04056931,
      "balance_loss_mlp": 1.02466774,
      "epoch": 0.7212168580532676,
      "flos": 21141869080320.0,
      "grad_norm": 1.9962954915761884,
      "language_loss": 0.68708301,
      "learning_rate": 7.612157445408987e-07,
      "loss": 0.70873368,
      "num_input_tokens_seen": 129005590,
      "step": 5998,
      "time_per_iteration": 2.689570665359497
    },
    {
      "auxiliary_loss_clip": 0.01145737,
      "auxiliary_loss_mlp": 0.01024191,
      "balance_loss_clip": 1.04835844,
      "balance_loss_mlp": 1.01675189,
      "epoch": 0.7213371009439067,
      "flos": 22345738335360.0,
      "grad_norm": 2.09908757744875,
      "language_loss": 0.74243224,
      "learning_rate": 7.606042820747716e-07,
      "loss": 0.76413143,
      "num_input_tokens_seen": 129021995,
      "step": 5999,
      "time_per_iteration": 2.696467638015747
    },
    {
      "auxiliary_loss_clip": 0.01149592,
      "auxiliary_loss_mlp": 0.01024061,
      "balance_loss_clip": 1.0489608,
      "balance_loss_mlp": 1.01709664,
      "epoch": 0.7214573438345457,
      "flos": 18515901490560.0,
      "grad_norm": 1.8101811894730397,
      "language_loss": 0.85186398,
      "learning_rate": 7.599930076239889e-07,
      "loss": 0.87360048,
      "num_input_tokens_seen": 129039280,
      "step": 6000,
      "time_per_iteration": 2.61403226852417
    },
    {
      "auxiliary_loss_clip": 0.01119007,
      "auxiliary_loss_mlp": 0.00762218,
      "balance_loss_clip": 1.04427481,
      "balance_loss_mlp": 1.00042057,
      "epoch": 0.7215775867251849,
      "flos": 35736108606720.0,
      "grad_norm": 2.2754610261186605,
      "language_loss": 0.70793509,
      "learning_rate": 7.593819212812818e-07,
      "loss": 0.72674739,
      "num_input_tokens_seen": 129060860,
      "step": 6001,
      "time_per_iteration": 2.886016845703125
    },
    {
      "auxiliary_loss_clip": 0.01152659,
      "auxiliary_loss_mlp": 0.01025235,
      "balance_loss_clip": 1.0462153,
      "balance_loss_mlp": 1.01822603,
      "epoch": 0.721697829615824,
      "flos": 20372410909440.0,
      "grad_norm": 1.921434475019056,
      "language_loss": 0.71434057,
      "learning_rate": 7.587710231393508e-07,
      "loss": 0.73611951,
      "num_input_tokens_seen": 129079215,
      "step": 6002,
      "time_per_iteration": 2.6018102169036865
    },
    {
      "auxiliary_loss_clip": 0.01072769,
      "auxiliary_loss_mlp": 0.01027814,
      "balance_loss_clip": 1.03419578,
      "balance_loss_mlp": 1.0205121,
      "epoch": 0.721818072506463,
      "flos": 20229809915520.0,
      "grad_norm": 1.9565145724889768,
      "language_loss": 0.83627021,
      "learning_rate": 7.581603132908685e-07,
      "loss": 0.85727608,
      "num_input_tokens_seen": 129097185,
      "step": 6003,
      "time_per_iteration": 2.805802822113037
    },
    {
      "auxiliary_loss_clip": 0.01119991,
      "auxiliary_loss_mlp": 0.01025964,
      "balance_loss_clip": 1.04248905,
      "balance_loss_mlp": 1.01846004,
      "epoch": 0.7219383153971022,
      "flos": 18186887888640.0,
      "grad_norm": 1.9264464349986414,
      "language_loss": 0.78781664,
      "learning_rate": 7.575497918284795e-07,
      "loss": 0.80927622,
      "num_input_tokens_seen": 129114730,
      "step": 6004,
      "time_per_iteration": 2.6486427783966064
    },
    {
      "auxiliary_loss_clip": 0.01171138,
      "auxiliary_loss_mlp": 0.01038083,
      "balance_loss_clip": 1.04656744,
      "balance_loss_mlp": 1.03038168,
      "epoch": 0.7220585582877412,
      "flos": 17342124854400.0,
      "grad_norm": 1.9777521904200235,
      "language_loss": 0.74841464,
      "learning_rate": 7.569394588447984e-07,
      "loss": 0.77050686,
      "num_input_tokens_seen": 129131745,
      "step": 6005,
      "time_per_iteration": 2.611745595932007
    },
    {
      "auxiliary_loss_clip": 0.01146459,
      "auxiliary_loss_mlp": 0.01020795,
      "balance_loss_clip": 1.04309511,
      "balance_loss_mlp": 1.01388347,
      "epoch": 0.7221788011783803,
      "flos": 16976338704000.0,
      "grad_norm": 5.127573321283999,
      "language_loss": 0.7802,
      "learning_rate": 7.563293144324146e-07,
      "loss": 0.80187249,
      "num_input_tokens_seen": 129147295,
      "step": 6006,
      "time_per_iteration": 2.614257335662842
    },
    {
      "auxiliary_loss_clip": 0.01167719,
      "auxiliary_loss_mlp": 0.0102589,
      "balance_loss_clip": 1.04902935,
      "balance_loss_mlp": 1.01943755,
      "epoch": 0.7222990440690195,
      "flos": 26286359702400.0,
      "grad_norm": 2.0033094540193073,
      "language_loss": 0.80186415,
      "learning_rate": 7.557193586838834e-07,
      "loss": 0.82380021,
      "num_input_tokens_seen": 129162660,
      "step": 6007,
      "time_per_iteration": 2.6874029636383057
    },
    {
      "auxiliary_loss_clip": 0.01140828,
      "auxiliary_loss_mlp": 0.01025999,
      "balance_loss_clip": 1.04188979,
      "balance_loss_mlp": 1.01919782,
      "epoch": 0.7224192869596585,
      "flos": 17601687509760.0,
      "grad_norm": 2.2259188666445033,
      "language_loss": 0.71017319,
      "learning_rate": 7.551095916917371e-07,
      "loss": 0.73184144,
      "num_input_tokens_seen": 129179990,
      "step": 6008,
      "time_per_iteration": 2.6946029663085938
    },
    {
      "auxiliary_loss_clip": 0.01138253,
      "auxiliary_loss_mlp": 0.01037242,
      "balance_loss_clip": 1.04296994,
      "balance_loss_mlp": 1.02892089,
      "epoch": 0.7225395298502976,
      "flos": 12932331016320.0,
      "grad_norm": 2.4980442300563808,
      "language_loss": 0.66816783,
      "learning_rate": 7.545000135484758e-07,
      "loss": 0.68992281,
      "num_input_tokens_seen": 129197425,
      "step": 6009,
      "time_per_iteration": 2.7120201587677
    },
    {
      "auxiliary_loss_clip": 0.01169552,
      "auxiliary_loss_mlp": 0.00762618,
      "balance_loss_clip": 1.04785275,
      "balance_loss_mlp": 1.00051141,
      "epoch": 0.7226597727409367,
      "flos": 29643899592960.0,
      "grad_norm": 1.88189438697995,
      "language_loss": 0.62682831,
      "learning_rate": 7.538906243465714e-07,
      "loss": 0.64614999,
      "num_input_tokens_seen": 129217560,
      "step": 6010,
      "time_per_iteration": 2.6536030769348145
    },
    {
      "auxiliary_loss_clip": 0.01170332,
      "auxiliary_loss_mlp": 0.01029676,
      "balance_loss_clip": 1.04746675,
      "balance_loss_mlp": 1.02223754,
      "epoch": 0.7227800156315758,
      "flos": 13771635183360.0,
      "grad_norm": 2.0807136818669845,
      "language_loss": 0.78703099,
      "learning_rate": 7.5328142417847e-07,
      "loss": 0.80903101,
      "num_input_tokens_seen": 129234325,
      "step": 6011,
      "time_per_iteration": 3.54526686668396
    },
    {
      "auxiliary_loss_clip": 0.01150614,
      "auxiliary_loss_mlp": 0.01024945,
      "balance_loss_clip": 1.0427314,
      "balance_loss_mlp": 1.01826382,
      "epoch": 0.7229002585222148,
      "flos": 20301882554880.0,
      "grad_norm": 1.7612563004156065,
      "language_loss": 0.69359243,
      "learning_rate": 7.526724131365838e-07,
      "loss": 0.71534801,
      "num_input_tokens_seen": 129255280,
      "step": 6012,
      "time_per_iteration": 2.636235237121582
    },
    {
      "auxiliary_loss_clip": 0.01142525,
      "auxiliary_loss_mlp": 0.01026281,
      "balance_loss_clip": 1.04771602,
      "balance_loss_mlp": 1.01917934,
      "epoch": 0.723020501412854,
      "flos": 16581250033920.0,
      "grad_norm": 1.8679819088804706,
      "language_loss": 0.70845002,
      "learning_rate": 7.520635913133017e-07,
      "loss": 0.73013806,
      "num_input_tokens_seen": 129273910,
      "step": 6013,
      "time_per_iteration": 2.629301071166992
    },
    {
      "auxiliary_loss_clip": 0.01160074,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.0458169,
      "balance_loss_mlp": 1.02210498,
      "epoch": 0.7231407443034931,
      "flos": 28548300908160.0,
      "grad_norm": 1.7134138836915547,
      "language_loss": 0.82348061,
      "learning_rate": 7.514549588009798e-07,
      "loss": 0.84538114,
      "num_input_tokens_seen": 129294785,
      "step": 6014,
      "time_per_iteration": 3.6462883949279785
    },
    {
      "auxiliary_loss_clip": 0.01143421,
      "auxiliary_loss_mlp": 0.01028461,
      "balance_loss_clip": 1.04566479,
      "balance_loss_mlp": 1.02111149,
      "epoch": 0.7232609871941321,
      "flos": 30008536508160.0,
      "grad_norm": 2.879571924697623,
      "language_loss": 0.70631444,
      "learning_rate": 7.508465156919492e-07,
      "loss": 0.7280333,
      "num_input_tokens_seen": 129318295,
      "step": 6015,
      "time_per_iteration": 2.7295796871185303
    },
    {
      "auxiliary_loss_clip": 0.01140205,
      "auxiliary_loss_mlp": 0.01023636,
      "balance_loss_clip": 1.04287624,
      "balance_loss_mlp": 1.01607239,
      "epoch": 0.7233812300847713,
      "flos": 16654005031680.0,
      "grad_norm": 2.7950851854591066,
      "language_loss": 0.6126312,
      "learning_rate": 7.502382620785083e-07,
      "loss": 0.63426971,
      "num_input_tokens_seen": 129334845,
      "step": 6016,
      "time_per_iteration": 3.4593629837036133
    },
    {
      "auxiliary_loss_clip": 0.01034885,
      "auxiliary_loss_mlp": 0.01002343,
      "balance_loss_clip": 1.01612592,
      "balance_loss_mlp": 1.00126982,
      "epoch": 0.7235014729754103,
      "flos": 67258784050560.0,
      "grad_norm": 0.8042331994800976,
      "language_loss": 0.62557673,
      "learning_rate": 7.496301980529289e-07,
      "loss": 0.64594901,
      "num_input_tokens_seen": 129398055,
      "step": 6017,
      "time_per_iteration": 3.342642307281494
    },
    {
      "auxiliary_loss_clip": 0.01169143,
      "auxiliary_loss_mlp": 0.01029261,
      "balance_loss_clip": 1.04766893,
      "balance_loss_mlp": 1.02265358,
      "epoch": 0.7236217158660494,
      "flos": 26943237671040.0,
      "grad_norm": 2.0255911370787887,
      "language_loss": 0.74674225,
      "learning_rate": 7.490223237074547e-07,
      "loss": 0.76872623,
      "num_input_tokens_seen": 129417765,
      "step": 6018,
      "time_per_iteration": 3.6600136756896973
    },
    {
      "auxiliary_loss_clip": 0.01127142,
      "auxiliary_loss_mlp": 0.01024706,
      "balance_loss_clip": 1.04300821,
      "balance_loss_mlp": 1.01763678,
      "epoch": 0.7237419587566886,
      "flos": 29423372042880.0,
      "grad_norm": 2.063045654665973,
      "language_loss": 0.65972131,
      "learning_rate": 7.484146391342989e-07,
      "loss": 0.68123984,
      "num_input_tokens_seen": 129437560,
      "step": 6019,
      "time_per_iteration": 2.792160749435425
    },
    {
      "auxiliary_loss_clip": 0.01132549,
      "auxiliary_loss_mlp": 0.01032046,
      "balance_loss_clip": 1.04165912,
      "balance_loss_mlp": 1.0248816,
      "epoch": 0.7238622016473276,
      "flos": 17821496787840.0,
      "grad_norm": 2.4288829873466016,
      "language_loss": 0.56292099,
      "learning_rate": 7.478071444256484e-07,
      "loss": 0.58456689,
      "num_input_tokens_seen": 129455320,
      "step": 6020,
      "time_per_iteration": 2.7437868118286133
    },
    {
      "auxiliary_loss_clip": 0.01137956,
      "auxiliary_loss_mlp": 0.01025351,
      "balance_loss_clip": 1.04479313,
      "balance_loss_mlp": 1.01754868,
      "epoch": 0.7239824445379667,
      "flos": 25739117020800.0,
      "grad_norm": 2.6669095041473376,
      "language_loss": 0.79238588,
      "learning_rate": 7.471998396736579e-07,
      "loss": 0.81401896,
      "num_input_tokens_seen": 129475700,
      "step": 6021,
      "time_per_iteration": 2.7686469554901123
    },
    {
      "auxiliary_loss_clip": 0.01126182,
      "auxiliary_loss_mlp": 0.01027909,
      "balance_loss_clip": 1.04309702,
      "balance_loss_mlp": 1.02080393,
      "epoch": 0.7241026874286057,
      "flos": 23148916398720.0,
      "grad_norm": 1.710602996967692,
      "language_loss": 0.75883532,
      "learning_rate": 7.465927249704549e-07,
      "loss": 0.78037626,
      "num_input_tokens_seen": 129493585,
      "step": 6022,
      "time_per_iteration": 2.721277952194214
    },
    {
      "auxiliary_loss_clip": 0.01151056,
      "auxiliary_loss_mlp": 0.0102044,
      "balance_loss_clip": 1.04369354,
      "balance_loss_mlp": 1.01328743,
      "epoch": 0.7242229303192449,
      "flos": 20266905686400.0,
      "grad_norm": 1.8652088212433782,
      "language_loss": 0.77326041,
      "learning_rate": 7.459858004081398e-07,
      "loss": 0.7949754,
      "num_input_tokens_seen": 129511555,
      "step": 6023,
      "time_per_iteration": 2.623974084854126
    },
    {
      "auxiliary_loss_clip": 0.0102785,
      "auxiliary_loss_mlp": 0.01002039,
      "balance_loss_clip": 1.01117229,
      "balance_loss_mlp": 1.00091255,
      "epoch": 0.724343173209884,
      "flos": 62311659684480.0,
      "grad_norm": 0.6622173124468023,
      "language_loss": 0.58018768,
      "learning_rate": 7.453790660787815e-07,
      "loss": 0.60048652,
      "num_input_tokens_seen": 129579650,
      "step": 6024,
      "time_per_iteration": 3.3654680252075195
    },
    {
      "auxiliary_loss_clip": 0.01145507,
      "auxiliary_loss_mlp": 0.01027942,
      "balance_loss_clip": 1.04642725,
      "balance_loss_mlp": 1.01996636,
      "epoch": 0.724463416100523,
      "flos": 35006403813120.0,
      "grad_norm": 2.6191525802008027,
      "language_loss": 0.63469702,
      "learning_rate": 7.447725220744214e-07,
      "loss": 0.65643156,
      "num_input_tokens_seen": 129601895,
      "step": 6025,
      "time_per_iteration": 2.885371446609497
    },
    {
      "auxiliary_loss_clip": 0.01168177,
      "auxiliary_loss_mlp": 0.01027003,
      "balance_loss_clip": 1.04523504,
      "balance_loss_mlp": 1.01948071,
      "epoch": 0.7245836589911622,
      "flos": 21871968923520.0,
      "grad_norm": 2.2900594143764916,
      "language_loss": 0.77175581,
      "learning_rate": 7.441661684870717e-07,
      "loss": 0.79370761,
      "num_input_tokens_seen": 129622150,
      "step": 6026,
      "time_per_iteration": 2.6330676078796387
    },
    {
      "auxiliary_loss_clip": 0.01168215,
      "auxiliary_loss_mlp": 0.01023662,
      "balance_loss_clip": 1.04620934,
      "balance_loss_mlp": 1.01617575,
      "epoch": 0.7247039018818012,
      "flos": 23006494972800.0,
      "grad_norm": 2.0148088341013555,
      "language_loss": 0.82252127,
      "learning_rate": 7.435600054087152e-07,
      "loss": 0.8444401,
      "num_input_tokens_seen": 129644315,
      "step": 6027,
      "time_per_iteration": 2.6369972229003906
    },
    {
      "auxiliary_loss_clip": 0.01171351,
      "auxiliary_loss_mlp": 0.01027919,
      "balance_loss_clip": 1.04793859,
      "balance_loss_mlp": 1.02015853,
      "epoch": 0.7248241447724403,
      "flos": 31722588587520.0,
      "grad_norm": 1.9326139594386633,
      "language_loss": 0.74602067,
      "learning_rate": 7.42954032931308e-07,
      "loss": 0.76801336,
      "num_input_tokens_seen": 129665355,
      "step": 6028,
      "time_per_iteration": 2.6036477088928223
    },
    {
      "auxiliary_loss_clip": 0.01143585,
      "auxiliary_loss_mlp": 0.01023361,
      "balance_loss_clip": 1.04557502,
      "balance_loss_mlp": 1.01659262,
      "epoch": 0.7249443876630794,
      "flos": 34896984007680.0,
      "grad_norm": 1.8247142438225092,
      "language_loss": 0.74691665,
      "learning_rate": 7.423482511467733e-07,
      "loss": 0.76858616,
      "num_input_tokens_seen": 129686125,
      "step": 6029,
      "time_per_iteration": 2.7858493328094482
    },
    {
      "auxiliary_loss_clip": 0.01087207,
      "auxiliary_loss_mlp": 0.01027572,
      "balance_loss_clip": 1.03850937,
      "balance_loss_mlp": 1.02010369,
      "epoch": 0.7250646305537185,
      "flos": 26359294268160.0,
      "grad_norm": 2.939089179775525,
      "language_loss": 0.64896369,
      "learning_rate": 7.417426601470099e-07,
      "loss": 0.67011142,
      "num_input_tokens_seen": 129706485,
      "step": 6030,
      "time_per_iteration": 2.846094846725464
    },
    {
      "auxiliary_loss_clip": 0.01158455,
      "auxiliary_loss_mlp": 0.01025203,
      "balance_loss_clip": 1.04695034,
      "balance_loss_mlp": 1.01698935,
      "epoch": 0.7251848734443576,
      "flos": 30081614728320.0,
      "grad_norm": 2.364294980806612,
      "language_loss": 0.78911912,
      "learning_rate": 7.411372600238841e-07,
      "loss": 0.81095576,
      "num_input_tokens_seen": 129727100,
      "step": 6031,
      "time_per_iteration": 2.6934359073638916
    },
    {
      "auxiliary_loss_clip": 0.01166927,
      "auxiliary_loss_mlp": 0.0102453,
      "balance_loss_clip": 1.04487956,
      "balance_loss_mlp": 1.01766908,
      "epoch": 0.7253051163349967,
      "flos": 17785262943360.0,
      "grad_norm": 1.9534667195942927,
      "language_loss": 0.74072063,
      "learning_rate": 7.405320508692346e-07,
      "loss": 0.76263523,
      "num_input_tokens_seen": 129745840,
      "step": 6032,
      "time_per_iteration": 2.565553665161133
    },
    {
      "auxiliary_loss_clip": 0.01164753,
      "auxiliary_loss_mlp": 0.01019831,
      "balance_loss_clip": 1.04707718,
      "balance_loss_mlp": 1.01326585,
      "epoch": 0.7254253592256358,
      "flos": 12641346938880.0,
      "grad_norm": 1.883765977510127,
      "language_loss": 0.75518143,
      "learning_rate": 7.399270327748727e-07,
      "loss": 0.77702725,
      "num_input_tokens_seen": 129763500,
      "step": 6033,
      "time_per_iteration": 2.5950703620910645
    },
    {
      "auxiliary_loss_clip": 0.0112698,
      "auxiliary_loss_mlp": 0.00762197,
      "balance_loss_clip": 1.04131293,
      "balance_loss_mlp": 1.00050497,
      "epoch": 0.7255456021162748,
      "flos": 27199208966400.0,
      "grad_norm": 1.6951026405374912,
      "language_loss": 0.73996353,
      "learning_rate": 7.39322205832577e-07,
      "loss": 0.75885528,
      "num_input_tokens_seen": 129784390,
      "step": 6034,
      "time_per_iteration": 2.7417588233947754
    },
    {
      "auxiliary_loss_clip": 0.01135778,
      "auxiliary_loss_mlp": 0.01031809,
      "balance_loss_clip": 1.0432688,
      "balance_loss_mlp": 1.02460909,
      "epoch": 0.725665845006914,
      "flos": 21288205088640.0,
      "grad_norm": 2.1729259436346937,
      "language_loss": 0.80984604,
      "learning_rate": 7.387175701341009e-07,
      "loss": 0.83152193,
      "num_input_tokens_seen": 129803060,
      "step": 6035,
      "time_per_iteration": 2.682269811630249
    },
    {
      "auxiliary_loss_clip": 0.01155293,
      "auxiliary_loss_mlp": 0.0102648,
      "balance_loss_clip": 1.04534841,
      "balance_loss_mlp": 1.01899385,
      "epoch": 0.7257860878975531,
      "flos": 16033684129920.0,
      "grad_norm": 2.2499513702812965,
      "language_loss": 0.7225036,
      "learning_rate": 7.381131257711659e-07,
      "loss": 0.74432135,
      "num_input_tokens_seen": 129820165,
      "step": 6036,
      "time_per_iteration": 2.5395188331604004
    },
    {
      "auxiliary_loss_clip": 0.01137122,
      "auxiliary_loss_mlp": 0.01025946,
      "balance_loss_clip": 1.04631877,
      "balance_loss_mlp": 1.01896274,
      "epoch": 0.7259063307881921,
      "flos": 12129943052160.0,
      "grad_norm": 1.9243247290005863,
      "language_loss": 0.8354001,
      "learning_rate": 7.375088728354677e-07,
      "loss": 0.85703075,
      "num_input_tokens_seen": 129835195,
      "step": 6037,
      "time_per_iteration": 2.6259469985961914
    },
    {
      "auxiliary_loss_clip": 0.01130574,
      "auxiliary_loss_mlp": 0.01027683,
      "balance_loss_clip": 1.04218817,
      "balance_loss_mlp": 1.01977324,
      "epoch": 0.7260265736788313,
      "flos": 30443845432320.0,
      "grad_norm": 1.611608127342657,
      "language_loss": 0.67345154,
      "learning_rate": 7.369048114186691e-07,
      "loss": 0.69503409,
      "num_input_tokens_seen": 129856240,
      "step": 6038,
      "time_per_iteration": 3.6943089962005615
    },
    {
      "auxiliary_loss_clip": 0.0113442,
      "auxiliary_loss_mlp": 0.00761784,
      "balance_loss_clip": 1.04366779,
      "balance_loss_mlp": 1.00048518,
      "epoch": 0.7261468165694703,
      "flos": 21142264129920.0,
      "grad_norm": 1.7252588338763846,
      "language_loss": 0.83240414,
      "learning_rate": 7.363009416124055e-07,
      "loss": 0.85136622,
      "num_input_tokens_seen": 129875565,
      "step": 6039,
      "time_per_iteration": 2.644244909286499
    },
    {
      "auxiliary_loss_clip": 0.01129696,
      "auxiliary_loss_mlp": 0.01021987,
      "balance_loss_clip": 1.04413962,
      "balance_loss_mlp": 1.01435137,
      "epoch": 0.7262670594601094,
      "flos": 22306308180480.0,
      "grad_norm": 2.505559585286645,
      "language_loss": 0.62613881,
      "learning_rate": 7.356972635082852e-07,
      "loss": 0.64765573,
      "num_input_tokens_seen": 129894420,
      "step": 6040,
      "time_per_iteration": 3.661090850830078
    },
    {
      "auxiliary_loss_clip": 0.01110882,
      "auxiliary_loss_mlp": 0.01030952,
      "balance_loss_clip": 1.04597783,
      "balance_loss_mlp": 1.02310228,
      "epoch": 0.7263873023507486,
      "flos": 25335049950720.0,
      "grad_norm": 1.9506326564264742,
      "language_loss": 0.75420082,
      "learning_rate": 7.35093777197884e-07,
      "loss": 0.77561915,
      "num_input_tokens_seen": 129914490,
      "step": 6041,
      "time_per_iteration": 2.7841336727142334
    },
    {
      "auxiliary_loss_clip": 0.01140613,
      "auxiliary_loss_mlp": 0.010281,
      "balance_loss_clip": 1.04636586,
      "balance_loss_mlp": 1.02081966,
      "epoch": 0.7265075452413876,
      "flos": 23878621192320.0,
      "grad_norm": 2.535210887871698,
      "language_loss": 0.85829937,
      "learning_rate": 7.344904827727525e-07,
      "loss": 0.87998652,
      "num_input_tokens_seen": 129931670,
      "step": 6042,
      "time_per_iteration": 3.5182671546936035
    },
    {
      "auxiliary_loss_clip": 0.0112412,
      "auxiliary_loss_mlp": 0.0102983,
      "balance_loss_clip": 1.04004908,
      "balance_loss_mlp": 1.02277851,
      "epoch": 0.7266277881320267,
      "flos": 28724549967360.0,
      "grad_norm": 3.9676776459536964,
      "language_loss": 0.73717988,
      "learning_rate": 7.338873803244076e-07,
      "loss": 0.75871938,
      "num_input_tokens_seen": 129946905,
      "step": 6043,
      "time_per_iteration": 3.7689168453216553
    },
    {
      "auxiliary_loss_clip": 0.01137069,
      "auxiliary_loss_mlp": 0.01025962,
      "balance_loss_clip": 1.04402995,
      "balance_loss_mlp": 1.01877069,
      "epoch": 0.7267480310226658,
      "flos": 24863507182080.0,
      "grad_norm": 1.855519757133444,
      "language_loss": 0.80739188,
      "learning_rate": 7.332844699443401e-07,
      "loss": 0.82902217,
      "num_input_tokens_seen": 129965505,
      "step": 6044,
      "time_per_iteration": 2.7695343494415283
    },
    {
      "auxiliary_loss_clip": 0.01104064,
      "auxiliary_loss_mlp": 0.01020023,
      "balance_loss_clip": 1.04072523,
      "balance_loss_mlp": 1.0127933,
      "epoch": 0.7268682739133049,
      "flos": 27198490694400.0,
      "grad_norm": 1.8265371910283956,
      "language_loss": 0.75400984,
      "learning_rate": 7.326817517240121e-07,
      "loss": 0.77525067,
      "num_input_tokens_seen": 129987210,
      "step": 6045,
      "time_per_iteration": 2.774348020553589
    },
    {
      "auxiliary_loss_clip": 0.01153826,
      "auxiliary_loss_mlp": 0.00762187,
      "balance_loss_clip": 1.0442549,
      "balance_loss_mlp": 1.0003624,
      "epoch": 0.7269885168039439,
      "flos": 33508138688640.0,
      "grad_norm": 1.6360963827040123,
      "language_loss": 0.8330602,
      "learning_rate": 7.320792257548545e-07,
      "loss": 0.85222042,
      "num_input_tokens_seen": 130008385,
      "step": 6046,
      "time_per_iteration": 2.843472957611084
    },
    {
      "auxiliary_loss_clip": 0.01145398,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 1.04477799,
      "balance_loss_mlp": 1.01686335,
      "epoch": 0.7271087596945831,
      "flos": 24313750548480.0,
      "grad_norm": 2.6496434780557263,
      "language_loss": 0.76142198,
      "learning_rate": 7.314768921282704e-07,
      "loss": 0.78312212,
      "num_input_tokens_seen": 130029040,
      "step": 6047,
      "time_per_iteration": 2.704103708267212
    },
    {
      "auxiliary_loss_clip": 0.01155891,
      "auxiliary_loss_mlp": 0.01029164,
      "balance_loss_clip": 1.04494977,
      "balance_loss_mlp": 1.0222199,
      "epoch": 0.7272290025852222,
      "flos": 23805147922560.0,
      "grad_norm": 2.548517646583692,
      "language_loss": 0.71948904,
      "learning_rate": 7.30874750935633e-07,
      "loss": 0.74133956,
      "num_input_tokens_seen": 130048725,
      "step": 6048,
      "time_per_iteration": 2.7366344928741455
    },
    {
      "auxiliary_loss_clip": 0.0112576,
      "auxiliary_loss_mlp": 0.01026212,
      "balance_loss_clip": 1.04628301,
      "balance_loss_mlp": 1.01914907,
      "epoch": 0.7273492454758612,
      "flos": 16720367408640.0,
      "grad_norm": 1.8976280166605914,
      "language_loss": 0.79147887,
      "learning_rate": 7.30272802268286e-07,
      "loss": 0.81299853,
      "num_input_tokens_seen": 130065720,
      "step": 6049,
      "time_per_iteration": 2.766047716140747
    },
    {
      "auxiliary_loss_clip": 0.01074721,
      "auxiliary_loss_mlp": 0.01021175,
      "balance_loss_clip": 1.03736532,
      "balance_loss_mlp": 1.01480305,
      "epoch": 0.7274694883665004,
      "flos": 28031330413440.0,
      "grad_norm": 1.8463501990940934,
      "language_loss": 0.76124406,
      "learning_rate": 7.29671046217547e-07,
      "loss": 0.78220302,
      "num_input_tokens_seen": 130084830,
      "step": 6050,
      "time_per_iteration": 2.8344318866729736
    },
    {
      "auxiliary_loss_clip": 0.01126552,
      "auxiliary_loss_mlp": 0.01026808,
      "balance_loss_clip": 1.0425483,
      "balance_loss_mlp": 1.01989079,
      "epoch": 0.7275897312571394,
      "flos": 30372706546560.0,
      "grad_norm": 1.8469447149965865,
      "language_loss": 0.82000464,
      "learning_rate": 7.290694828746988e-07,
      "loss": 0.84153825,
      "num_input_tokens_seen": 130104495,
      "step": 6051,
      "time_per_iteration": 2.8092734813690186
    },
    {
      "auxiliary_loss_clip": 0.01131375,
      "auxiliary_loss_mlp": 0.01020357,
      "balance_loss_clip": 1.04233015,
      "balance_loss_mlp": 1.01362443,
      "epoch": 0.7277099741477785,
      "flos": 19204775498880.0,
      "grad_norm": 3.2988396453663342,
      "language_loss": 0.8555494,
      "learning_rate": 7.284681123310004e-07,
      "loss": 0.87706673,
      "num_input_tokens_seen": 130123210,
      "step": 6052,
      "time_per_iteration": 2.805621385574341
    },
    {
      "auxiliary_loss_clip": 0.0115578,
      "auxiliary_loss_mlp": 0.01027655,
      "balance_loss_clip": 1.04701483,
      "balance_loss_mlp": 1.02013302,
      "epoch": 0.7278302170384175,
      "flos": 20667884186880.0,
      "grad_norm": 2.0558144532474434,
      "language_loss": 0.79413271,
      "learning_rate": 7.27866934677678e-07,
      "loss": 0.81596702,
      "num_input_tokens_seen": 130142880,
      "step": 6053,
      "time_per_iteration": 2.6840944290161133
    },
    {
      "auxiliary_loss_clip": 0.01111307,
      "auxiliary_loss_mlp": 0.01030418,
      "balance_loss_clip": 1.04242575,
      "balance_loss_mlp": 1.02375078,
      "epoch": 0.7279504599290567,
      "flos": 19093200877440.0,
      "grad_norm": 2.6842628763117418,
      "language_loss": 0.78485179,
      "learning_rate": 7.272659500059297e-07,
      "loss": 0.80626905,
      "num_input_tokens_seen": 130160220,
      "step": 6054,
      "time_per_iteration": 2.719508171081543
    },
    {
      "auxiliary_loss_clip": 0.01150432,
      "auxiliary_loss_mlp": 0.01030212,
      "balance_loss_clip": 1.04569113,
      "balance_loss_mlp": 1.02257097,
      "epoch": 0.7280707028196958,
      "flos": 19062174504960.0,
      "grad_norm": 2.2135263576301636,
      "language_loss": 0.80418754,
      "learning_rate": 7.266651584069264e-07,
      "loss": 0.82599396,
      "num_input_tokens_seen": 130177885,
      "step": 6055,
      "time_per_iteration": 2.7250959873199463
    },
    {
      "auxiliary_loss_clip": 0.0116098,
      "auxiliary_loss_mlp": 0.01024945,
      "balance_loss_clip": 1.04947817,
      "balance_loss_mlp": 1.01773334,
      "epoch": 0.7281909457103348,
      "flos": 37196308293120.0,
      "grad_norm": 2.426394453965791,
      "language_loss": 0.56653541,
      "learning_rate": 7.260645599718045e-07,
      "loss": 0.5883947,
      "num_input_tokens_seen": 130204240,
      "step": 6056,
      "time_per_iteration": 2.746609687805176
    },
    {
      "auxiliary_loss_clip": 0.01141004,
      "auxiliary_loss_mlp": 0.01029972,
      "balance_loss_clip": 1.04423809,
      "balance_loss_mlp": 1.02188075,
      "epoch": 0.728311188600974,
      "flos": 20667094087680.0,
      "grad_norm": 2.4093169748622194,
      "language_loss": 0.67530572,
      "learning_rate": 7.254641547916767e-07,
      "loss": 0.69701546,
      "num_input_tokens_seen": 130221735,
      "step": 6057,
      "time_per_iteration": 2.6372549533843994
    },
    {
      "auxiliary_loss_clip": 0.01169553,
      "auxiliary_loss_mlp": 0.01028681,
      "balance_loss_clip": 1.04964685,
      "balance_loss_mlp": 1.02082479,
      "epoch": 0.728431431491613,
      "flos": 28840685616000.0,
      "grad_norm": 1.7697727457209231,
      "language_loss": 0.69516104,
      "learning_rate": 7.248639429576226e-07,
      "loss": 0.71714342,
      "num_input_tokens_seen": 130241190,
      "step": 6058,
      "time_per_iteration": 2.66440749168396
    },
    {
      "auxiliary_loss_clip": 0.01158098,
      "auxiliary_loss_mlp": 0.01025235,
      "balance_loss_clip": 1.04665709,
      "balance_loss_mlp": 1.01795781,
      "epoch": 0.7285516743822521,
      "flos": 25991856092160.0,
      "grad_norm": 1.6396838993723266,
      "language_loss": 0.72115958,
      "learning_rate": 7.242639245606959e-07,
      "loss": 0.74299294,
      "num_input_tokens_seen": 130260980,
      "step": 6059,
      "time_per_iteration": 2.7847440242767334
    },
    {
      "auxiliary_loss_clip": 0.01147444,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 1.04423583,
      "balance_loss_mlp": 1.01925135,
      "epoch": 0.7286719172728913,
      "flos": 16399721675520.0,
      "grad_norm": 1.8483861491469487,
      "language_loss": 0.82856584,
      "learning_rate": 7.236640996919168e-07,
      "loss": 0.85031039,
      "num_input_tokens_seen": 130280025,
      "step": 6060,
      "time_per_iteration": 2.6948609352111816
    },
    {
      "auxiliary_loss_clip": 0.01155868,
      "auxiliary_loss_mlp": 0.01024727,
      "balance_loss_clip": 1.04640698,
      "balance_loss_mlp": 1.0177207,
      "epoch": 0.7287921601635303,
      "flos": 22018161277440.0,
      "grad_norm": 1.6669281791124535,
      "language_loss": 0.70620751,
      "learning_rate": 7.230644684422782e-07,
      "loss": 0.72801346,
      "num_input_tokens_seen": 130300255,
      "step": 6061,
      "time_per_iteration": 2.6385855674743652
    },
    {
      "auxiliary_loss_clip": 0.01122337,
      "auxiliary_loss_mlp": 0.01026803,
      "balance_loss_clip": 1.04070354,
      "balance_loss_mlp": 1.01956117,
      "epoch": 0.7289124030541694,
      "flos": 24600927784320.0,
      "grad_norm": 1.8373045031556106,
      "language_loss": 0.8153165,
      "learning_rate": 7.224650309027451e-07,
      "loss": 0.83680785,
      "num_input_tokens_seen": 130320005,
      "step": 6062,
      "time_per_iteration": 2.724644422531128
    },
    {
      "auxiliary_loss_clip": 0.01158826,
      "auxiliary_loss_mlp": 0.01024347,
      "balance_loss_clip": 1.04792356,
      "balance_loss_mlp": 1.01711059,
      "epoch": 0.7290326459448085,
      "flos": 21393638484480.0,
      "grad_norm": 1.7879442554754248,
      "language_loss": 0.6885621,
      "learning_rate": 7.218657871642506e-07,
      "loss": 0.71039379,
      "num_input_tokens_seen": 130338810,
      "step": 6063,
      "time_per_iteration": 2.7416863441467285
    },
    {
      "auxiliary_loss_clip": 0.01172451,
      "auxiliary_loss_mlp": 0.01027487,
      "balance_loss_clip": 1.04922533,
      "balance_loss_mlp": 1.02042651,
      "epoch": 0.7291528888354476,
      "flos": 18587686821120.0,
      "grad_norm": 2.206293166427509,
      "language_loss": 0.62533784,
      "learning_rate": 7.212667373177012e-07,
      "loss": 0.6473372,
      "num_input_tokens_seen": 130353805,
      "step": 6064,
      "time_per_iteration": 3.5416882038116455
    },
    {
      "auxiliary_loss_clip": 0.01125117,
      "auxiliary_loss_mlp": 0.01022004,
      "balance_loss_clip": 1.04112911,
      "balance_loss_mlp": 1.01473212,
      "epoch": 0.7292731317260867,
      "flos": 18951066760320.0,
      "grad_norm": 1.8400974973285498,
      "language_loss": 0.75520265,
      "learning_rate": 7.206678814539704e-07,
      "loss": 0.77667385,
      "num_input_tokens_seen": 130372105,
      "step": 6065,
      "time_per_iteration": 2.688159227371216
    },
    {
      "auxiliary_loss_clip": 0.01120521,
      "auxiliary_loss_mlp": 0.01024613,
      "balance_loss_clip": 1.04160213,
      "balance_loss_mlp": 1.01788056,
      "epoch": 0.7293933746167258,
      "flos": 21067569797760.0,
      "grad_norm": 1.7042634643452053,
      "language_loss": 0.72566283,
      "learning_rate": 7.20069219663904e-07,
      "loss": 0.74711418,
      "num_input_tokens_seen": 130391990,
      "step": 6066,
      "time_per_iteration": 3.738140821456909
    },
    {
      "auxiliary_loss_clip": 0.01156822,
      "auxiliary_loss_mlp": 0.01022982,
      "balance_loss_clip": 1.04450428,
      "balance_loss_mlp": 1.01588881,
      "epoch": 0.7295136175073649,
      "flos": 22453326547200.0,
      "grad_norm": 3.4958486570518583,
      "language_loss": 0.79648411,
      "learning_rate": 7.1947075203832e-07,
      "loss": 0.81828213,
      "num_input_tokens_seen": 130411970,
      "step": 6067,
      "time_per_iteration": 3.5604584217071533
    },
    {
      "auxiliary_loss_clip": 0.01067042,
      "auxiliary_loss_mlp": 0.0100148,
      "balance_loss_clip": 1.01161027,
      "balance_loss_mlp": 1.00038326,
      "epoch": 0.7296338603980039,
      "flos": 56125506648960.0,
      "grad_norm": 0.8717250012307006,
      "language_loss": 0.6016326,
      "learning_rate": 7.188724786680049e-07,
      "loss": 0.62231779,
      "num_input_tokens_seen": 130472440,
      "step": 6068,
      "time_per_iteration": 3.1580066680908203
    },
    {
      "auxiliary_loss_clip": 0.01140313,
      "auxiliary_loss_mlp": 0.01024481,
      "balance_loss_clip": 1.04286063,
      "balance_loss_mlp": 1.01746583,
      "epoch": 0.7297541032886431,
      "flos": 25228287751680.0,
      "grad_norm": 1.5251486149729556,
      "language_loss": 0.75822389,
      "learning_rate": 7.182743996437162e-07,
      "loss": 0.77987182,
      "num_input_tokens_seen": 130491975,
      "step": 6069,
      "time_per_iteration": 3.5650408267974854
    },
    {
      "auxiliary_loss_clip": 0.01133764,
      "auxiliary_loss_mlp": 0.01026829,
      "balance_loss_clip": 1.04201341,
      "balance_loss_mlp": 1.01912236,
      "epoch": 0.7298743461792822,
      "flos": 26467600752000.0,
      "grad_norm": 1.803512162646333,
      "language_loss": 0.68999541,
      "learning_rate": 7.176765150561819e-07,
      "loss": 0.71160138,
      "num_input_tokens_seen": 130510580,
      "step": 6070,
      "time_per_iteration": 2.705667018890381
    },
    {
      "auxiliary_loss_clip": 0.01167364,
      "auxiliary_loss_mlp": 0.01030735,
      "balance_loss_clip": 1.04464531,
      "balance_loss_mlp": 1.02363563,
      "epoch": 0.7299945890699212,
      "flos": 19569053278080.0,
      "grad_norm": 2.161320006402385,
      "language_loss": 0.79831499,
      "learning_rate": 7.170788249961002e-07,
      "loss": 0.82029593,
      "num_input_tokens_seen": 130529090,
      "step": 6071,
      "time_per_iteration": 2.548200845718384
    },
    {
      "auxiliary_loss_clip": 0.0116492,
      "auxiliary_loss_mlp": 0.01023026,
      "balance_loss_clip": 1.04524159,
      "balance_loss_mlp": 1.01583171,
      "epoch": 0.7301148319605604,
      "flos": 22928963466240.0,
      "grad_norm": 1.7972045737814695,
      "language_loss": 0.88163012,
      "learning_rate": 7.164813295541418e-07,
      "loss": 0.90350962,
      "num_input_tokens_seen": 130548655,
      "step": 6072,
      "time_per_iteration": 2.6509146690368652
    },
    {
      "auxiliary_loss_clip": 0.01142458,
      "auxiliary_loss_mlp": 0.01026115,
      "balance_loss_clip": 1.04353952,
      "balance_loss_mlp": 1.01864684,
      "epoch": 0.7302350748511994,
      "flos": 25369703596800.0,
      "grad_norm": 1.8014078959940987,
      "language_loss": 0.70234555,
      "learning_rate": 7.15884028820944e-07,
      "loss": 0.72403133,
      "num_input_tokens_seen": 130567710,
      "step": 6073,
      "time_per_iteration": 2.6406314373016357
    },
    {
      "auxiliary_loss_clip": 0.01119481,
      "auxiliary_loss_mlp": 0.01020014,
      "balance_loss_clip": 1.03955066,
      "balance_loss_mlp": 1.01275373,
      "epoch": 0.7303553177418385,
      "flos": 27819170732160.0,
      "grad_norm": 2.0886198458986738,
      "language_loss": 0.60453653,
      "learning_rate": 7.152869228871185e-07,
      "loss": 0.6259315,
      "num_input_tokens_seen": 130590195,
      "step": 6074,
      "time_per_iteration": 2.7987170219421387
    },
    {
      "auxiliary_loss_clip": 0.01137001,
      "auxiliary_loss_mlp": 0.0102862,
      "balance_loss_clip": 1.04436815,
      "balance_loss_mlp": 1.02083564,
      "epoch": 0.7304755606324776,
      "flos": 24426510318720.0,
      "grad_norm": 1.8193269530027092,
      "language_loss": 0.71880651,
      "learning_rate": 7.146900118432457e-07,
      "loss": 0.74046278,
      "num_input_tokens_seen": 130609940,
      "step": 6075,
      "time_per_iteration": 2.7236416339874268
    },
    {
      "auxiliary_loss_clip": 0.01078551,
      "auxiliary_loss_mlp": 0.01024851,
      "balance_loss_clip": 1.03255081,
      "balance_loss_mlp": 1.01818752,
      "epoch": 0.7305958035231167,
      "flos": 23840483927040.0,
      "grad_norm": 1.6679542320729357,
      "language_loss": 0.85823202,
      "learning_rate": 7.140932957798753e-07,
      "loss": 0.87926602,
      "num_input_tokens_seen": 130628380,
      "step": 6076,
      "time_per_iteration": 2.923943281173706
    },
    {
      "auxiliary_loss_clip": 0.01145297,
      "auxiliary_loss_mlp": 0.01026866,
      "balance_loss_clip": 1.04400325,
      "balance_loss_mlp": 1.0196687,
      "epoch": 0.7307160464137558,
      "flos": 16726939597440.0,
      "grad_norm": 2.1147092281533304,
      "language_loss": 0.70737302,
      "learning_rate": 7.134967747875309e-07,
      "loss": 0.72909456,
      "num_input_tokens_seen": 130646590,
      "step": 6077,
      "time_per_iteration": 2.645317792892456
    },
    {
      "auxiliary_loss_clip": 0.01146929,
      "auxiliary_loss_mlp": 0.01024915,
      "balance_loss_clip": 1.04217148,
      "balance_loss_mlp": 1.01772034,
      "epoch": 0.7308362893043949,
      "flos": 21798280172160.0,
      "grad_norm": 2.2725630580416114,
      "language_loss": 0.81947792,
      "learning_rate": 7.129004489567014e-07,
      "loss": 0.8411963,
      "num_input_tokens_seen": 130664070,
      "step": 6078,
      "time_per_iteration": 2.632467746734619
    },
    {
      "auxiliary_loss_clip": 0.01129207,
      "auxiliary_loss_mlp": 0.01025145,
      "balance_loss_clip": 1.04286528,
      "balance_loss_mlp": 1.01793289,
      "epoch": 0.730956532195034,
      "flos": 10707377840640.0,
      "grad_norm": 2.2325612113101987,
      "language_loss": 0.77751648,
      "learning_rate": 7.123043183778512e-07,
      "loss": 0.79906005,
      "num_input_tokens_seen": 130681400,
      "step": 6079,
      "time_per_iteration": 2.6426584720611572
    },
    {
      "auxiliary_loss_clip": 0.01131183,
      "auxiliary_loss_mlp": 0.01031881,
      "balance_loss_clip": 1.04377627,
      "balance_loss_mlp": 1.02489567,
      "epoch": 0.731076775085673,
      "flos": 19791987039360.0,
      "grad_norm": 1.5467680036266793,
      "language_loss": 0.64803576,
      "learning_rate": 7.117083831414114e-07,
      "loss": 0.66966635,
      "num_input_tokens_seen": 130700675,
      "step": 6080,
      "time_per_iteration": 2.7225708961486816
    },
    {
      "auxiliary_loss_clip": 0.01166604,
      "auxiliary_loss_mlp": 0.01020294,
      "balance_loss_clip": 1.04679906,
      "balance_loss_mlp": 1.01345456,
      "epoch": 0.7311970179763122,
      "flos": 20447033414400.0,
      "grad_norm": 1.8519438125482652,
      "language_loss": 0.69611144,
      "learning_rate": 7.11112643337787e-07,
      "loss": 0.71798038,
      "num_input_tokens_seen": 130719720,
      "step": 6081,
      "time_per_iteration": 2.587704658508301
    },
    {
      "auxiliary_loss_clip": 0.01142237,
      "auxiliary_loss_mlp": 0.01022853,
      "balance_loss_clip": 1.04646039,
      "balance_loss_mlp": 1.01534843,
      "epoch": 0.7313172608669513,
      "flos": 18513818501760.0,
      "grad_norm": 2.516022164197006,
      "language_loss": 0.76848078,
      "learning_rate": 7.10517099057349e-07,
      "loss": 0.79013169,
      "num_input_tokens_seen": 130736670,
      "step": 6082,
      "time_per_iteration": 2.6389245986938477
    },
    {
      "auxiliary_loss_clip": 0.01143274,
      "auxiliary_loss_mlp": 0.01024411,
      "balance_loss_clip": 1.04400098,
      "balance_loss_mlp": 1.01701379,
      "epoch": 0.7314375037575903,
      "flos": 16180738410240.0,
      "grad_norm": 2.431267638794644,
      "language_loss": 0.61869383,
      "learning_rate": 7.099217503904411e-07,
      "loss": 0.64037061,
      "num_input_tokens_seen": 130754525,
      "step": 6083,
      "time_per_iteration": 2.689507246017456
    },
    {
      "auxiliary_loss_clip": 0.01142663,
      "auxiliary_loss_mlp": 0.01026319,
      "balance_loss_clip": 1.04364085,
      "balance_loss_mlp": 1.01925635,
      "epoch": 0.7315577466482295,
      "flos": 17967940536960.0,
      "grad_norm": 1.848987066115641,
      "language_loss": 0.89435959,
      "learning_rate": 7.093265974273788e-07,
      "loss": 0.91604936,
      "num_input_tokens_seen": 130772420,
      "step": 6084,
      "time_per_iteration": 2.6483230590820312
    },
    {
      "auxiliary_loss_clip": 0.01153098,
      "auxiliary_loss_mlp": 0.01023974,
      "balance_loss_clip": 1.04372191,
      "balance_loss_mlp": 1.01718807,
      "epoch": 0.7316779895388685,
      "flos": 18405440190720.0,
      "grad_norm": 2.324260853356819,
      "language_loss": 0.72290403,
      "learning_rate": 7.087316402584447e-07,
      "loss": 0.7446748,
      "num_input_tokens_seen": 130791245,
      "step": 6085,
      "time_per_iteration": 2.657064199447632
    },
    {
      "auxiliary_loss_clip": 0.01167611,
      "auxiliary_loss_mlp": 0.01024828,
      "balance_loss_clip": 1.04676402,
      "balance_loss_mlp": 1.01789606,
      "epoch": 0.7317982324295076,
      "flos": 17928294900480.0,
      "grad_norm": 1.8059641123729397,
      "language_loss": 0.86411631,
      "learning_rate": 7.081368789738953e-07,
      "loss": 0.88604081,
      "num_input_tokens_seen": 130808445,
      "step": 6086,
      "time_per_iteration": 2.6807308197021484
    },
    {
      "auxiliary_loss_clip": 0.01133245,
      "auxiliary_loss_mlp": 0.01023615,
      "balance_loss_clip": 1.03931236,
      "balance_loss_mlp": 1.01625669,
      "epoch": 0.7319184753201466,
      "flos": 27229840289280.0,
      "grad_norm": 1.9421357534881216,
      "language_loss": 0.77874804,
      "learning_rate": 7.075423136639537e-07,
      "loss": 0.80031669,
      "num_input_tokens_seen": 130827700,
      "step": 6087,
      "time_per_iteration": 2.7404978275299072
    },
    {
      "auxiliary_loss_clip": 0.01122055,
      "auxiliary_loss_mlp": 0.01028526,
      "balance_loss_clip": 1.04129291,
      "balance_loss_mlp": 1.02099752,
      "epoch": 0.7320387182107858,
      "flos": 37448544574080.0,
      "grad_norm": 1.7687022053844337,
      "language_loss": 0.74847019,
      "learning_rate": 7.069479444188149e-07,
      "loss": 0.76997596,
      "num_input_tokens_seen": 130848290,
      "step": 6088,
      "time_per_iteration": 2.8200392723083496
    },
    {
      "auxiliary_loss_clip": 0.01131981,
      "auxiliary_loss_mlp": 0.01026569,
      "balance_loss_clip": 1.04380238,
      "balance_loss_mlp": 1.01938391,
      "epoch": 0.7321589611014249,
      "flos": 17859023521920.0,
      "grad_norm": 2.514695109210698,
      "language_loss": 0.82167232,
      "learning_rate": 7.063537713286453e-07,
      "loss": 0.84325778,
      "num_input_tokens_seen": 130865970,
      "step": 6089,
      "time_per_iteration": 3.588059663772583
    },
    {
      "auxiliary_loss_clip": 0.01143785,
      "auxiliary_loss_mlp": 0.01027205,
      "balance_loss_clip": 1.04302943,
      "balance_loss_mlp": 1.01976013,
      "epoch": 0.7322792039920639,
      "flos": 26100593539200.0,
      "grad_norm": 2.0386376291392785,
      "language_loss": 0.80817878,
      "learning_rate": 7.057597944835803e-07,
      "loss": 0.82988864,
      "num_input_tokens_seen": 130885245,
      "step": 6090,
      "time_per_iteration": 2.7465696334838867
    },
    {
      "auxiliary_loss_clip": 0.01130876,
      "auxiliary_loss_mlp": 0.01024406,
      "balance_loss_clip": 1.04108882,
      "balance_loss_mlp": 1.01755738,
      "epoch": 0.7323994468827031,
      "flos": 25369093065600.0,
      "grad_norm": 1.6603237953045553,
      "language_loss": 0.74667007,
      "learning_rate": 7.051660139737253e-07,
      "loss": 0.76822287,
      "num_input_tokens_seen": 130903465,
      "step": 6091,
      "time_per_iteration": 2.721478223800659
    },
    {
      "auxiliary_loss_clip": 0.01152329,
      "auxiliary_loss_mlp": 0.00761875,
      "balance_loss_clip": 1.0470053,
      "balance_loss_mlp": 1.00037515,
      "epoch": 0.7325196897733421,
      "flos": 26907075653760.0,
      "grad_norm": 2.599001020473849,
      "language_loss": 0.76815605,
      "learning_rate": 7.045724298891565e-07,
      "loss": 0.78729808,
      "num_input_tokens_seen": 130922935,
      "step": 6092,
      "time_per_iteration": 3.6391196250915527
    },
    {
      "auxiliary_loss_clip": 0.01154427,
      "auxiliary_loss_mlp": 0.01023141,
      "balance_loss_clip": 1.04767847,
      "balance_loss_mlp": 1.016137,
      "epoch": 0.7326399326639812,
      "flos": 25775781828480.0,
      "grad_norm": 1.8900254938226724,
      "language_loss": 0.69346154,
      "learning_rate": 7.039790423199192e-07,
      "loss": 0.7152372,
      "num_input_tokens_seen": 130942575,
      "step": 6093,
      "time_per_iteration": 3.596498966217041
    },
    {
      "auxiliary_loss_clip": 0.01143828,
      "auxiliary_loss_mlp": 0.01019487,
      "balance_loss_clip": 1.04539418,
      "balance_loss_mlp": 1.0119884,
      "epoch": 0.7327601755546204,
      "flos": 21032269706880.0,
      "grad_norm": 1.956844456188359,
      "language_loss": 0.77277523,
      "learning_rate": 7.033858513560322e-07,
      "loss": 0.79440832,
      "num_input_tokens_seen": 130958870,
      "step": 6094,
      "time_per_iteration": 2.6508889198303223
    },
    {
      "auxiliary_loss_clip": 0.01156238,
      "auxiliary_loss_mlp": 0.01026914,
      "balance_loss_clip": 1.04832983,
      "balance_loss_mlp": 1.01970243,
      "epoch": 0.7328804184452594,
      "flos": 16289224462080.0,
      "grad_norm": 2.195745681618227,
      "language_loss": 0.75718939,
      "learning_rate": 7.027928570874794e-07,
      "loss": 0.77902091,
      "num_input_tokens_seen": 130977060,
      "step": 6095,
      "time_per_iteration": 3.5407028198242188
    },
    {
      "auxiliary_loss_clip": 0.0116588,
      "auxiliary_loss_mlp": 0.01027161,
      "balance_loss_clip": 1.04567385,
      "balance_loss_mlp": 1.02011538,
      "epoch": 0.7330006613358985,
      "flos": 17858233422720.0,
      "grad_norm": 2.1271134368896627,
      "language_loss": 0.85205698,
      "learning_rate": 7.022000596042194e-07,
      "loss": 0.87398738,
      "num_input_tokens_seen": 130994160,
      "step": 6096,
      "time_per_iteration": 2.5869038105010986
    },
    {
      "auxiliary_loss_clip": 0.01127139,
      "auxiliary_loss_mlp": 0.01023343,
      "balance_loss_clip": 1.04021788,
      "balance_loss_mlp": 1.01607978,
      "epoch": 0.7331209042265376,
      "flos": 22492074343680.0,
      "grad_norm": 2.112414518673481,
      "language_loss": 0.81813431,
      "learning_rate": 7.016074589961784e-07,
      "loss": 0.83963913,
      "num_input_tokens_seen": 131012725,
      "step": 6097,
      "time_per_iteration": 2.7254457473754883
    },
    {
      "auxiliary_loss_clip": 0.01136594,
      "auxiliary_loss_mlp": 0.01022544,
      "balance_loss_clip": 1.04444003,
      "balance_loss_mlp": 1.01540303,
      "epoch": 0.7332411471171767,
      "flos": 33072757937280.0,
      "grad_norm": 2.0078630190332856,
      "language_loss": 0.67061466,
      "learning_rate": 7.01015055353253e-07,
      "loss": 0.69220603,
      "num_input_tokens_seen": 131035150,
      "step": 6098,
      "time_per_iteration": 2.735308885574341
    },
    {
      "auxiliary_loss_clip": 0.01102566,
      "auxiliary_loss_mlp": 0.01026944,
      "balance_loss_clip": 1.04050553,
      "balance_loss_mlp": 1.01927865,
      "epoch": 0.7333613900078157,
      "flos": 22743017735040.0,
      "grad_norm": 1.6646637309088514,
      "language_loss": 0.7814225,
      "learning_rate": 7.004228487653123e-07,
      "loss": 0.80271757,
      "num_input_tokens_seen": 131055955,
      "step": 6099,
      "time_per_iteration": 2.786825656890869
    },
    {
      "auxiliary_loss_clip": 0.01121683,
      "auxiliary_loss_mlp": 0.01023959,
      "balance_loss_clip": 1.03855133,
      "balance_loss_mlp": 1.01706314,
      "epoch": 0.7334816328984549,
      "flos": 22346133384960.0,
      "grad_norm": 1.7017383862174718,
      "language_loss": 0.78292018,
      "learning_rate": 6.998308393221906e-07,
      "loss": 0.8043766,
      "num_input_tokens_seen": 131074360,
      "step": 6100,
      "time_per_iteration": 2.7142446041107178
    },
    {
      "auxiliary_loss_clip": 0.01131895,
      "auxiliary_loss_mlp": 0.01025532,
      "balance_loss_clip": 1.04542506,
      "balance_loss_mlp": 1.01916361,
      "epoch": 0.733601875789094,
      "flos": 20736149984640.0,
      "grad_norm": 2.63138820289893,
      "language_loss": 0.71046239,
      "learning_rate": 6.992390271136977e-07,
      "loss": 0.73203665,
      "num_input_tokens_seen": 131090070,
      "step": 6101,
      "time_per_iteration": 2.6897268295288086
    },
    {
      "auxiliary_loss_clip": 0.01146556,
      "auxiliary_loss_mlp": 0.01027683,
      "balance_loss_clip": 1.04303432,
      "balance_loss_mlp": 1.02063465,
      "epoch": 0.733722118679733,
      "flos": 22564362464640.0,
      "grad_norm": 1.9277950259977643,
      "language_loss": 0.85557026,
      "learning_rate": 6.986474122296094e-07,
      "loss": 0.87731266,
      "num_input_tokens_seen": 131109185,
      "step": 6102,
      "time_per_iteration": 2.7104127407073975
    },
    {
      "auxiliary_loss_clip": 0.01171092,
      "auxiliary_loss_mlp": 0.01025821,
      "balance_loss_clip": 1.04846251,
      "balance_loss_mlp": 1.01814389,
      "epoch": 0.7338423615703722,
      "flos": 20084192179200.0,
      "grad_norm": 3.697914800952412,
      "language_loss": 0.72823578,
      "learning_rate": 6.980559947596751e-07,
      "loss": 0.75020492,
      "num_input_tokens_seen": 131127725,
      "step": 6103,
      "time_per_iteration": 2.6069588661193848
    },
    {
      "auxiliary_loss_clip": 0.01110651,
      "auxiliary_loss_mlp": 0.01026824,
      "balance_loss_clip": 1.03949296,
      "balance_loss_mlp": 1.0194118,
      "epoch": 0.7339626044610112,
      "flos": 21687675217920.0,
      "grad_norm": 1.9903755556231217,
      "language_loss": 0.7576701,
      "learning_rate": 6.974647747936109e-07,
      "loss": 0.77904487,
      "num_input_tokens_seen": 131146110,
      "step": 6104,
      "time_per_iteration": 2.825244665145874
    },
    {
      "auxiliary_loss_clip": 0.01167891,
      "auxiliary_loss_mlp": 0.00761817,
      "balance_loss_clip": 1.04689729,
      "balance_loss_mlp": 1.00042892,
      "epoch": 0.7340828473516503,
      "flos": 15268248282240.0,
      "grad_norm": 2.0342524475475745,
      "language_loss": 0.82833529,
      "learning_rate": 6.968737524211039e-07,
      "loss": 0.84763241,
      "num_input_tokens_seen": 131162920,
      "step": 6105,
      "time_per_iteration": 2.5975215435028076
    },
    {
      "auxiliary_loss_clip": 0.01153968,
      "auxiliary_loss_mlp": 0.01023195,
      "balance_loss_clip": 1.04725409,
      "balance_loss_mlp": 1.0161078,
      "epoch": 0.7342030902422895,
      "flos": 22930112701440.0,
      "grad_norm": 2.234961974479447,
      "language_loss": 0.80044168,
      "learning_rate": 6.962829277318132e-07,
      "loss": 0.82221329,
      "num_input_tokens_seen": 131182515,
      "step": 6106,
      "time_per_iteration": 2.681307315826416
    },
    {
      "auxiliary_loss_clip": 0.01157299,
      "auxiliary_loss_mlp": 0.01024428,
      "balance_loss_clip": 1.04708707,
      "balance_loss_mlp": 1.01787448,
      "epoch": 0.7343233331329285,
      "flos": 25847890381440.0,
      "grad_norm": 7.203489426051874,
      "language_loss": 0.83189243,
      "learning_rate": 6.956923008153652e-07,
      "loss": 0.8537097,
      "num_input_tokens_seen": 131202280,
      "step": 6107,
      "time_per_iteration": 2.6482224464416504
    },
    {
      "auxiliary_loss_clip": 0.01151931,
      "auxiliary_loss_mlp": 0.01026013,
      "balance_loss_clip": 1.04225278,
      "balance_loss_mlp": 1.01909256,
      "epoch": 0.7344435760235676,
      "flos": 18478985287680.0,
      "grad_norm": 2.284717790901659,
      "language_loss": 0.84300148,
      "learning_rate": 6.951018717613593e-07,
      "loss": 0.8647809,
      "num_input_tokens_seen": 131221295,
      "step": 6108,
      "time_per_iteration": 2.683530807495117
    },
    {
      "auxiliary_loss_clip": 0.01153055,
      "auxiliary_loss_mlp": 0.01025558,
      "balance_loss_clip": 1.04590845,
      "balance_loss_mlp": 1.01861393,
      "epoch": 0.7345638189142067,
      "flos": 17640040256640.0,
      "grad_norm": 2.074661338240404,
      "language_loss": 0.78401726,
      "learning_rate": 6.945116406593614e-07,
      "loss": 0.80580342,
      "num_input_tokens_seen": 131240150,
      "step": 6109,
      "time_per_iteration": 2.661816358566284
    },
    {
      "auxiliary_loss_clip": 0.01114456,
      "auxiliary_loss_mlp": 0.01025119,
      "balance_loss_clip": 1.04267454,
      "balance_loss_mlp": 1.01791906,
      "epoch": 0.7346840618048458,
      "flos": 20260225756800.0,
      "grad_norm": 2.0592450051425,
      "language_loss": 0.74569356,
      "learning_rate": 6.939216075989089e-07,
      "loss": 0.76708937,
      "num_input_tokens_seen": 131258080,
      "step": 6110,
      "time_per_iteration": 2.7212815284729004
    },
    {
      "auxiliary_loss_clip": 0.0113694,
      "auxiliary_loss_mlp": 0.01026289,
      "balance_loss_clip": 1.04203832,
      "balance_loss_mlp": 1.0191927,
      "epoch": 0.7348043046954849,
      "flos": 29023183641600.0,
      "grad_norm": 2.0460165824631478,
      "language_loss": 0.65902704,
      "learning_rate": 6.933317726695109e-07,
      "loss": 0.68065929,
      "num_input_tokens_seen": 131279310,
      "step": 6111,
      "time_per_iteration": 2.6604928970336914
    },
    {
      "auxiliary_loss_clip": 0.01124483,
      "auxiliary_loss_mlp": 0.01021088,
      "balance_loss_clip": 1.04504037,
      "balance_loss_mlp": 1.01418018,
      "epoch": 0.734924547586124,
      "flos": 17931203902080.0,
      "grad_norm": 2.6281996606430384,
      "language_loss": 0.80100441,
      "learning_rate": 6.92742135960644e-07,
      "loss": 0.82246006,
      "num_input_tokens_seen": 131297010,
      "step": 6112,
      "time_per_iteration": 2.7247776985168457
    },
    {
      "auxiliary_loss_clip": 0.0105774,
      "auxiliary_loss_mlp": 0.01000864,
      "balance_loss_clip": 1.0117979,
      "balance_loss_mlp": 0.9996835,
      "epoch": 0.7350447904767631,
      "flos": 63588319850880.0,
      "grad_norm": 0.8248769957439724,
      "language_loss": 0.55651057,
      "learning_rate": 6.921526975617556e-07,
      "loss": 0.57709658,
      "num_input_tokens_seen": 131356470,
      "step": 6113,
      "time_per_iteration": 3.2584221363067627
    },
    {
      "auxiliary_loss_clip": 0.0114409,
      "auxiliary_loss_mlp": 0.01024351,
      "balance_loss_clip": 1.04375446,
      "balance_loss_mlp": 1.0168556,
      "epoch": 0.7351650333674021,
      "flos": 21580015178880.0,
      "grad_norm": 1.9153751081803623,
      "language_loss": 0.75655913,
      "learning_rate": 6.915634575622631e-07,
      "loss": 0.77824354,
      "num_input_tokens_seen": 131374985,
      "step": 6114,
      "time_per_iteration": 2.685204267501831
    },
    {
      "auxiliary_loss_clip": 0.01164807,
      "auxiliary_loss_mlp": 0.01022969,
      "balance_loss_clip": 1.04503083,
      "balance_loss_mlp": 1.01580477,
      "epoch": 0.7352852762580413,
      "flos": 18186349184640.0,
      "grad_norm": 1.9002517603334987,
      "language_loss": 0.7124114,
      "learning_rate": 6.909744160515532e-07,
      "loss": 0.73428917,
      "num_input_tokens_seen": 131393125,
      "step": 6115,
      "time_per_iteration": 3.6154119968414307
    },
    {
      "auxiliary_loss_clip": 0.01138339,
      "auxiliary_loss_mlp": 0.01023337,
      "balance_loss_clip": 1.04455948,
      "balance_loss_mlp": 1.01588619,
      "epoch": 0.7354055191486804,
      "flos": 38910073063680.0,
      "grad_norm": 2.0563367034157762,
      "language_loss": 0.69645005,
      "learning_rate": 6.903855731189849e-07,
      "loss": 0.71806681,
      "num_input_tokens_seen": 131415760,
      "step": 6116,
      "time_per_iteration": 2.8522636890411377
    },
    {
      "auxiliary_loss_clip": 0.01145538,
      "auxiliary_loss_mlp": 0.01030377,
      "balance_loss_clip": 1.04394782,
      "balance_loss_mlp": 1.02267587,
      "epoch": 0.7355257620393194,
      "flos": 16289978647680.0,
      "grad_norm": 2.0658872808129614,
      "language_loss": 0.8164078,
      "learning_rate": 6.897969288538825e-07,
      "loss": 0.83816689,
      "num_input_tokens_seen": 131433705,
      "step": 6117,
      "time_per_iteration": 2.6889140605926514
    },
    {
      "auxiliary_loss_clip": 0.01134037,
      "auxiliary_loss_mlp": 0.0102378,
      "balance_loss_clip": 1.04170787,
      "balance_loss_mlp": 1.01679707,
      "epoch": 0.7356460049299585,
      "flos": 18114240631680.0,
      "grad_norm": 1.8389390514381192,
      "language_loss": 0.81356007,
      "learning_rate": 6.892084833455452e-07,
      "loss": 0.83513826,
      "num_input_tokens_seen": 131453275,
      "step": 6118,
      "time_per_iteration": 3.685389757156372
    },
    {
      "auxiliary_loss_clip": 0.01153269,
      "auxiliary_loss_mlp": 0.01020717,
      "balance_loss_clip": 1.04668689,
      "balance_loss_mlp": 1.01398134,
      "epoch": 0.7357662478205976,
      "flos": 21325193118720.0,
      "grad_norm": 1.4415584071879186,
      "language_loss": 0.83714473,
      "learning_rate": 6.886202366832384e-07,
      "loss": 0.85888469,
      "num_input_tokens_seen": 131474960,
      "step": 6119,
      "time_per_iteration": 3.6610970497131348
    },
    {
      "auxiliary_loss_clip": 0.0111033,
      "auxiliary_loss_mlp": 0.01024022,
      "balance_loss_clip": 1.04309869,
      "balance_loss_mlp": 1.01679778,
      "epoch": 0.7358864907112367,
      "flos": 14246841139200.0,
      "grad_norm": 2.238615281896033,
      "language_loss": 0.7363233,
      "learning_rate": 6.880321889561987e-07,
      "loss": 0.75766677,
      "num_input_tokens_seen": 131492935,
      "step": 6120,
      "time_per_iteration": 2.7224111557006836
    },
    {
      "auxiliary_loss_clip": 0.01119619,
      "auxiliary_loss_mlp": 0.0102503,
      "balance_loss_clip": 1.04149139,
      "balance_loss_mlp": 1.01675725,
      "epoch": 0.7360067336018757,
      "flos": 22309684058880.0,
      "grad_norm": 2.866509689302074,
      "language_loss": 0.64800429,
      "learning_rate": 6.874443402536338e-07,
      "loss": 0.66945088,
      "num_input_tokens_seen": 131512025,
      "step": 6121,
      "time_per_iteration": 3.6904549598693848
    },
    {
      "auxiliary_loss_clip": 0.0114401,
      "auxiliary_loss_mlp": 0.01023259,
      "balance_loss_clip": 1.04593706,
      "balance_loss_mlp": 1.01614857,
      "epoch": 0.7361269764925149,
      "flos": 25554607833600.0,
      "grad_norm": 2.1278581255609277,
      "language_loss": 0.80672872,
      "learning_rate": 6.868566906647177e-07,
      "loss": 0.82840145,
      "num_input_tokens_seen": 131532975,
      "step": 6122,
      "time_per_iteration": 2.666492462158203
    },
    {
      "auxiliary_loss_clip": 0.01156282,
      "auxiliary_loss_mlp": 0.01027164,
      "balance_loss_clip": 1.04644883,
      "balance_loss_mlp": 1.019508,
      "epoch": 0.736247219383154,
      "flos": 20376505059840.0,
      "grad_norm": 1.9326593252309239,
      "language_loss": 0.83392537,
      "learning_rate": 6.862692402785984e-07,
      "loss": 0.8557598,
      "num_input_tokens_seen": 131553225,
      "step": 6123,
      "time_per_iteration": 2.744079351425171
    },
    {
      "auxiliary_loss_clip": 0.01037955,
      "auxiliary_loss_mlp": 0.01000677,
      "balance_loss_clip": 1.01759815,
      "balance_loss_mlp": 0.99937785,
      "epoch": 0.736367462273793,
      "flos": 70339525735680.0,
      "grad_norm": 0.6814638255791329,
      "language_loss": 0.4961538,
      "learning_rate": 6.856819891843899e-07,
      "loss": 0.51654017,
      "num_input_tokens_seen": 131617930,
      "step": 6124,
      "time_per_iteration": 3.357239246368408
    },
    {
      "auxiliary_loss_clip": 0.01099121,
      "auxiliary_loss_mlp": 0.01023124,
      "balance_loss_clip": 1.04019141,
      "balance_loss_mlp": 1.01605499,
      "epoch": 0.7364877051644322,
      "flos": 22412711243520.0,
      "grad_norm": 1.9801678944034977,
      "language_loss": 0.72386044,
      "learning_rate": 6.8509493747118e-07,
      "loss": 0.74508286,
      "num_input_tokens_seen": 131636740,
      "step": 6125,
      "time_per_iteration": 2.7425270080566406
    },
    {
      "auxiliary_loss_clip": 0.01168817,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.04805648,
      "balance_loss_mlp": 1.02288449,
      "epoch": 0.7366079480550712,
      "flos": 12130266274560.0,
      "grad_norm": 3.502325688495394,
      "language_loss": 0.88581395,
      "learning_rate": 6.845080852280221e-07,
      "loss": 0.90779781,
      "num_input_tokens_seen": 131653810,
      "step": 6126,
      "time_per_iteration": 2.5839459896087646
    },
    {
      "auxiliary_loss_clip": 0.01126044,
      "auxiliary_loss_mlp": 0.01022169,
      "balance_loss_clip": 1.04225349,
      "balance_loss_mlp": 1.015347,
      "epoch": 0.7367281909457103,
      "flos": 15049336844160.0,
      "grad_norm": 1.8123386553167553,
      "language_loss": 0.74197537,
      "learning_rate": 6.839214325439409e-07,
      "loss": 0.76345754,
      "num_input_tokens_seen": 131671505,
      "step": 6127,
      "time_per_iteration": 2.666963815689087
    },
    {
      "auxiliary_loss_clip": 0.01133677,
      "auxiliary_loss_mlp": 0.010312,
      "balance_loss_clip": 1.04400754,
      "balance_loss_mlp": 1.02447414,
      "epoch": 0.7368484338363495,
      "flos": 23510752053120.0,
      "grad_norm": 1.7097160009710766,
      "language_loss": 0.71880198,
      "learning_rate": 6.833349795079327e-07,
      "loss": 0.74045074,
      "num_input_tokens_seen": 131690615,
      "step": 6128,
      "time_per_iteration": 2.70161771774292
    },
    {
      "auxiliary_loss_clip": 0.01127505,
      "auxiliary_loss_mlp": 0.01023456,
      "balance_loss_clip": 1.04527569,
      "balance_loss_mlp": 1.01648521,
      "epoch": 0.7369686767269885,
      "flos": 27417833095680.0,
      "grad_norm": 1.808217919827775,
      "language_loss": 0.68385935,
      "learning_rate": 6.827487262089613e-07,
      "loss": 0.70536894,
      "num_input_tokens_seen": 131711120,
      "step": 6129,
      "time_per_iteration": 2.8333590030670166
    },
    {
      "auxiliary_loss_clip": 0.01040251,
      "auxiliary_loss_mlp": 0.01000826,
      "balance_loss_clip": 1.01017439,
      "balance_loss_mlp": 0.99964631,
      "epoch": 0.7370889196176276,
      "flos": 70293343824000.0,
      "grad_norm": 0.892938912722892,
      "language_loss": 0.5684284,
      "learning_rate": 6.821626727359606e-07,
      "loss": 0.58883917,
      "num_input_tokens_seen": 131776680,
      "step": 6130,
      "time_per_iteration": 3.3624086380004883
    },
    {
      "auxiliary_loss_clip": 0.0113823,
      "auxiliary_loss_mlp": 0.01028494,
      "balance_loss_clip": 1.04475737,
      "balance_loss_mlp": 1.02020252,
      "epoch": 0.7372091625082667,
      "flos": 18040839189120.0,
      "grad_norm": 3.1919237366180155,
      "language_loss": 0.77428907,
      "learning_rate": 6.815768191778348e-07,
      "loss": 0.79595637,
      "num_input_tokens_seen": 131794760,
      "step": 6131,
      "time_per_iteration": 2.718435525894165
    },
    {
      "auxiliary_loss_clip": 0.01149373,
      "auxiliary_loss_mlp": 0.01023869,
      "balance_loss_clip": 1.04390895,
      "balance_loss_mlp": 1.01605487,
      "epoch": 0.7373294053989058,
      "flos": 33726331854720.0,
      "grad_norm": 1.759335844518035,
      "language_loss": 0.72766936,
      "learning_rate": 6.809911656234569e-07,
      "loss": 0.74940181,
      "num_input_tokens_seen": 131816735,
      "step": 6132,
      "time_per_iteration": 2.781639814376831
    },
    {
      "auxiliary_loss_clip": 0.01126753,
      "auxiliary_loss_mlp": 0.01026103,
      "balance_loss_clip": 1.04064047,
      "balance_loss_mlp": 1.01896262,
      "epoch": 0.7374496482895448,
      "flos": 21506326427520.0,
      "grad_norm": 2.0289294203118997,
      "language_loss": 0.78189903,
      "learning_rate": 6.804057121616707e-07,
      "loss": 0.80342758,
      "num_input_tokens_seen": 131834940,
      "step": 6133,
      "time_per_iteration": 2.733464002609253
    },
    {
      "auxiliary_loss_clip": 0.01156362,
      "auxiliary_loss_mlp": 0.01023743,
      "balance_loss_clip": 1.04602313,
      "balance_loss_mlp": 1.01636338,
      "epoch": 0.737569891180184,
      "flos": 24936908624640.0,
      "grad_norm": 1.8225271731410826,
      "language_loss": 0.71994406,
      "learning_rate": 6.798204588812888e-07,
      "loss": 0.74174511,
      "num_input_tokens_seen": 131854355,
      "step": 6134,
      "time_per_iteration": 2.732590436935425
    },
    {
      "auxiliary_loss_clip": 0.01088261,
      "auxiliary_loss_mlp": 0.00762605,
      "balance_loss_clip": 1.03734326,
      "balance_loss_mlp": 1.00032258,
      "epoch": 0.7376901340708231,
      "flos": 20664544222080.0,
      "grad_norm": 1.9399441165931683,
      "language_loss": 0.7533108,
      "learning_rate": 6.792354058710937e-07,
      "loss": 0.77181947,
      "num_input_tokens_seen": 131871825,
      "step": 6135,
      "time_per_iteration": 2.833524465560913
    },
    {
      "auxiliary_loss_clip": 0.01162145,
      "auxiliary_loss_mlp": 0.010203,
      "balance_loss_clip": 1.04581976,
      "balance_loss_mlp": 1.01377332,
      "epoch": 0.7378103769614621,
      "flos": 23805794367360.0,
      "grad_norm": 2.1575796254350936,
      "language_loss": 0.65367758,
      "learning_rate": 6.786505532198374e-07,
      "loss": 0.67550206,
      "num_input_tokens_seen": 131890770,
      "step": 6136,
      "time_per_iteration": 2.6107802391052246
    },
    {
      "auxiliary_loss_clip": 0.01170643,
      "auxiliary_loss_mlp": 0.01024665,
      "balance_loss_clip": 1.04824233,
      "balance_loss_mlp": 1.0179565,
      "epoch": 0.7379306198521013,
      "flos": 22237216369920.0,
      "grad_norm": 1.6630128771375878,
      "language_loss": 0.85195726,
      "learning_rate": 6.780659010162411e-07,
      "loss": 0.87391031,
      "num_input_tokens_seen": 131909720,
      "step": 6137,
      "time_per_iteration": 2.6605801582336426
    },
    {
      "auxiliary_loss_clip": 0.01128664,
      "auxiliary_loss_mlp": 0.01021449,
      "balance_loss_clip": 1.04304051,
      "balance_loss_mlp": 1.01457059,
      "epoch": 0.7380508627427403,
      "flos": 14903108576640.0,
      "grad_norm": 2.221249823598007,
      "language_loss": 0.83147383,
      "learning_rate": 6.774814493489975e-07,
      "loss": 0.85297489,
      "num_input_tokens_seen": 131927395,
      "step": 6138,
      "time_per_iteration": 2.6516075134277344
    },
    {
      "auxiliary_loss_clip": 0.01147902,
      "auxiliary_loss_mlp": 0.01021874,
      "balance_loss_clip": 1.04400373,
      "balance_loss_mlp": 1.0152756,
      "epoch": 0.7381711056333794,
      "flos": 21685843624320.0,
      "grad_norm": 2.6914548967087124,
      "language_loss": 0.66469336,
      "learning_rate": 6.768971983067655e-07,
      "loss": 0.68639106,
      "num_input_tokens_seen": 131947725,
      "step": 6139,
      "time_per_iteration": 2.652316093444824
    },
    {
      "auxiliary_loss_clip": 0.01066724,
      "auxiliary_loss_mlp": 0.0100128,
      "balance_loss_clip": 1.01143932,
      "balance_loss_mlp": 1.00017691,
      "epoch": 0.7382913485240186,
      "flos": 52404263596800.0,
      "grad_norm": 1.0358542440511391,
      "language_loss": 0.67792279,
      "learning_rate": 6.763131479781772e-07,
      "loss": 0.69860286,
      "num_input_tokens_seen": 131997485,
      "step": 6140,
      "time_per_iteration": 3.0506324768066406
    },
    {
      "auxiliary_loss_clip": 0.01134148,
      "auxiliary_loss_mlp": 0.01022972,
      "balance_loss_clip": 1.04591441,
      "balance_loss_mlp": 1.01643085,
      "epoch": 0.7384115914146576,
      "flos": 21798818876160.0,
      "grad_norm": 1.8192883155409,
      "language_loss": 0.7598536,
      "learning_rate": 6.757292984518316e-07,
      "loss": 0.78142476,
      "num_input_tokens_seen": 132016885,
      "step": 6141,
      "time_per_iteration": 3.6502699851989746
    },
    {
      "auxiliary_loss_clip": 0.01057816,
      "auxiliary_loss_mlp": 0.01002055,
      "balance_loss_clip": 1.0117054,
      "balance_loss_mlp": 1.00093412,
      "epoch": 0.7385318343052967,
      "flos": 61494331662720.0,
      "grad_norm": 0.742588597527093,
      "language_loss": 0.56435442,
      "learning_rate": 6.751456498162981e-07,
      "loss": 0.58495313,
      "num_input_tokens_seen": 132075920,
      "step": 6142,
      "time_per_iteration": 3.1735730171203613
    },
    {
      "auxiliary_loss_clip": 0.0115061,
      "auxiliary_loss_mlp": 0.01020828,
      "balance_loss_clip": 1.04159772,
      "balance_loss_mlp": 1.01447964,
      "epoch": 0.7386520771959358,
      "flos": 17013757697280.0,
      "grad_norm": 1.8059934994640525,
      "language_loss": 0.84910655,
      "learning_rate": 6.745622021601174e-07,
      "loss": 0.870821,
      "num_input_tokens_seen": 132092945,
      "step": 6143,
      "time_per_iteration": 2.579279661178589
    },
    {
      "auxiliary_loss_clip": 0.01127825,
      "auxiliary_loss_mlp": 0.01025174,
      "balance_loss_clip": 1.04210639,
      "balance_loss_mlp": 1.01752687,
      "epoch": 0.7387723200865749,
      "flos": 18770759464320.0,
      "grad_norm": 3.1796999162567205,
      "language_loss": 0.69921428,
      "learning_rate": 6.739789555717954e-07,
      "loss": 0.72074425,
      "num_input_tokens_seen": 132109920,
      "step": 6144,
      "time_per_iteration": 3.619631767272949
    },
    {
      "auxiliary_loss_clip": 0.01164342,
      "auxiliary_loss_mlp": 0.0102766,
      "balance_loss_clip": 1.04446328,
      "balance_loss_mlp": 1.02112484,
      "epoch": 0.738892562977214,
      "flos": 22525542840960.0,
      "grad_norm": 1.9400843925302267,
      "language_loss": 0.77214098,
      "learning_rate": 6.733959101398124e-07,
      "loss": 0.79406095,
      "num_input_tokens_seen": 132128050,
      "step": 6145,
      "time_per_iteration": 3.5423765182495117
    },
    {
      "auxiliary_loss_clip": 0.01136346,
      "auxiliary_loss_mlp": 0.01028842,
      "balance_loss_clip": 1.04183698,
      "balance_loss_mlp": 1.02232146,
      "epoch": 0.7390128058678531,
      "flos": 21501478091520.0,
      "grad_norm": 1.6293756708988685,
      "language_loss": 0.81573701,
      "learning_rate": 6.728130659526143e-07,
      "loss": 0.83738887,
      "num_input_tokens_seen": 132145860,
      "step": 6146,
      "time_per_iteration": 2.675356388092041
    },
    {
      "auxiliary_loss_clip": 0.01143133,
      "auxiliary_loss_mlp": 0.01031307,
      "balance_loss_clip": 1.04584646,
      "balance_loss_mlp": 1.02400541,
      "epoch": 0.7391330487584922,
      "flos": 25776176878080.0,
      "grad_norm": 3.7715962278292854,
      "language_loss": 0.70765883,
      "learning_rate": 6.7223042309862e-07,
      "loss": 0.72940326,
      "num_input_tokens_seen": 132166060,
      "step": 6147,
      "time_per_iteration": 3.622819423675537
    },
    {
      "auxiliary_loss_clip": 0.01149088,
      "auxiliary_loss_mlp": 0.01027724,
      "balance_loss_clip": 1.04287314,
      "balance_loss_mlp": 1.02035069,
      "epoch": 0.7392532916491312,
      "flos": 28366736636160.0,
      "grad_norm": 2.2515622299296325,
      "language_loss": 0.73953688,
      "learning_rate": 6.716479816662144e-07,
      "loss": 0.76130497,
      "num_input_tokens_seen": 132187790,
      "step": 6148,
      "time_per_iteration": 2.7269699573516846
    },
    {
      "auxiliary_loss_clip": 0.01142956,
      "auxiliary_loss_mlp": 0.01026883,
      "balance_loss_clip": 1.04342246,
      "balance_loss_mlp": 1.01971292,
      "epoch": 0.7393735345397703,
      "flos": 23585877348480.0,
      "grad_norm": 1.9986583679071979,
      "language_loss": 0.73205453,
      "learning_rate": 6.710657417437531e-07,
      "loss": 0.75375295,
      "num_input_tokens_seen": 132207495,
      "step": 6149,
      "time_per_iteration": 2.673571825027466
    },
    {
      "auxiliary_loss_clip": 0.01137523,
      "auxiliary_loss_mlp": 0.01024475,
      "balance_loss_clip": 1.04323149,
      "balance_loss_mlp": 1.01763797,
      "epoch": 0.7394937774304094,
      "flos": 19974772373760.0,
      "grad_norm": 2.077001166597141,
      "language_loss": 0.80304366,
      "learning_rate": 6.704837034195628e-07,
      "loss": 0.82466364,
      "num_input_tokens_seen": 132225960,
      "step": 6150,
      "time_per_iteration": 2.639875888824463
    },
    {
      "auxiliary_loss_clip": 0.0114937,
      "auxiliary_loss_mlp": 0.0102832,
      "balance_loss_clip": 1.04698658,
      "balance_loss_mlp": 1.02125061,
      "epoch": 0.7396140203210485,
      "flos": 23478037741440.0,
      "grad_norm": 1.6936409704951199,
      "language_loss": 0.85031468,
      "learning_rate": 6.699018667819376e-07,
      "loss": 0.87209153,
      "num_input_tokens_seen": 132245360,
      "step": 6151,
      "time_per_iteration": 2.757138252258301
    },
    {
      "auxiliary_loss_clip": 0.0115036,
      "auxiliary_loss_mlp": 0.01029054,
      "balance_loss_clip": 1.04483044,
      "balance_loss_mlp": 1.02143013,
      "epoch": 0.7397342632116876,
      "flos": 25555433846400.0,
      "grad_norm": 1.542186676860542,
      "language_loss": 0.72961366,
      "learning_rate": 6.693202319191415e-07,
      "loss": 0.7514078,
      "num_input_tokens_seen": 132267095,
      "step": 6152,
      "time_per_iteration": 2.702171802520752
    },
    {
      "auxiliary_loss_clip": 0.01167266,
      "auxiliary_loss_mlp": 0.01031521,
      "balance_loss_clip": 1.0494566,
      "balance_loss_mlp": 1.02404332,
      "epoch": 0.7398545061023267,
      "flos": 24755021130240.0,
      "grad_norm": 13.260195463512833,
      "language_loss": 0.74896753,
      "learning_rate": 6.687387989194084e-07,
      "loss": 0.77095544,
      "num_input_tokens_seen": 132286610,
      "step": 6153,
      "time_per_iteration": 2.613478660583496
    },
    {
      "auxiliary_loss_clip": 0.01131357,
      "auxiliary_loss_mlp": 0.01024296,
      "balance_loss_clip": 1.04274118,
      "balance_loss_mlp": 1.01739645,
      "epoch": 0.7399747489929658,
      "flos": 16508602776960.0,
      "grad_norm": 1.8932424676161752,
      "language_loss": 0.79156268,
      "learning_rate": 6.681575678709404e-07,
      "loss": 0.81311917,
      "num_input_tokens_seen": 132305300,
      "step": 6154,
      "time_per_iteration": 2.6608335971832275
    },
    {
      "auxiliary_loss_clip": 0.01153166,
      "auxiliary_loss_mlp": 0.01028663,
      "balance_loss_clip": 1.04603541,
      "balance_loss_mlp": 1.021752,
      "epoch": 0.7400949918836048,
      "flos": 24097065753600.0,
      "grad_norm": 2.011080587499535,
      "language_loss": 0.70579457,
      "learning_rate": 6.67576538861911e-07,
      "loss": 0.72761285,
      "num_input_tokens_seen": 132323875,
      "step": 6155,
      "time_per_iteration": 2.58485746383667
    },
    {
      "auxiliary_loss_clip": 0.01137159,
      "auxiliary_loss_mlp": 0.01022853,
      "balance_loss_clip": 1.04439569,
      "balance_loss_mlp": 1.01618052,
      "epoch": 0.740215234774244,
      "flos": 21802517976960.0,
      "grad_norm": 1.5327827975595518,
      "language_loss": 0.82253277,
      "learning_rate": 6.669957119804612e-07,
      "loss": 0.8441329,
      "num_input_tokens_seen": 132345510,
      "step": 6156,
      "time_per_iteration": 2.7540125846862793
    },
    {
      "auxiliary_loss_clip": 0.01147084,
      "auxiliary_loss_mlp": 0.01023176,
      "balance_loss_clip": 1.04396665,
      "balance_loss_mlp": 1.01556492,
      "epoch": 0.7403354776648831,
      "flos": 18733196816640.0,
      "grad_norm": 3.31545718488185,
      "language_loss": 0.73210192,
      "learning_rate": 6.66415087314702e-07,
      "loss": 0.75380456,
      "num_input_tokens_seen": 132360465,
      "step": 6157,
      "time_per_iteration": 2.6089510917663574
    },
    {
      "auxiliary_loss_clip": 0.01139749,
      "auxiliary_loss_mlp": 0.01021445,
      "balance_loss_clip": 1.04238176,
      "balance_loss_mlp": 1.01436055,
      "epoch": 0.7404557205555221,
      "flos": 16909581277440.0,
      "grad_norm": 2.6886413052750595,
      "language_loss": 0.73544782,
      "learning_rate": 6.65834664952714e-07,
      "loss": 0.75705969,
      "num_input_tokens_seen": 132377915,
      "step": 6158,
      "time_per_iteration": 2.67838978767395
    },
    {
      "auxiliary_loss_clip": 0.01126677,
      "auxiliary_loss_mlp": 0.01021295,
      "balance_loss_clip": 1.0417192,
      "balance_loss_mlp": 1.01444316,
      "epoch": 0.7405759634461613,
      "flos": 21214408596480.0,
      "grad_norm": 1.755785839360571,
      "language_loss": 0.75983554,
      "learning_rate": 6.652544449825457e-07,
      "loss": 0.78131521,
      "num_input_tokens_seen": 132398170,
      "step": 6159,
      "time_per_iteration": 2.699551820755005
    },
    {
      "auxiliary_loss_clip": 0.01147358,
      "auxiliary_loss_mlp": 0.01026041,
      "balance_loss_clip": 1.04426515,
      "balance_loss_mlp": 1.0186677,
      "epoch": 0.7406962063368003,
      "flos": 20480106862080.0,
      "grad_norm": 1.6452843978623748,
      "language_loss": 0.76552039,
      "learning_rate": 6.646744274922182e-07,
      "loss": 0.78725439,
      "num_input_tokens_seen": 132416615,
      "step": 6160,
      "time_per_iteration": 2.7571797370910645
    },
    {
      "auxiliary_loss_clip": 0.01141219,
      "auxiliary_loss_mlp": 0.01023363,
      "balance_loss_clip": 1.04379106,
      "balance_loss_mlp": 1.01647294,
      "epoch": 0.7408164492274394,
      "flos": 19791915212160.0,
      "grad_norm": 4.016063246530093,
      "language_loss": 0.75948834,
      "learning_rate": 6.640946125697171e-07,
      "loss": 0.78113413,
      "num_input_tokens_seen": 132434145,
      "step": 6161,
      "time_per_iteration": 2.672748327255249
    },
    {
      "auxiliary_loss_clip": 0.01154503,
      "auxiliary_loss_mlp": 0.0102632,
      "balance_loss_clip": 1.04459548,
      "balance_loss_mlp": 1.01885772,
      "epoch": 0.7409366921180786,
      "flos": 29204855654400.0,
      "grad_norm": 2.464715969395424,
      "language_loss": 0.76026523,
      "learning_rate": 6.635150003030017e-07,
      "loss": 0.7820735,
      "num_input_tokens_seen": 132452670,
      "step": 6162,
      "time_per_iteration": 2.781038999557495
    },
    {
      "auxiliary_loss_clip": 0.01110683,
      "auxiliary_loss_mlp": 0.01023184,
      "balance_loss_clip": 1.03815889,
      "balance_loss_mlp": 1.01624632,
      "epoch": 0.7410569350087176,
      "flos": 22930004960640.0,
      "grad_norm": 2.1607997882612326,
      "language_loss": 0.86309177,
      "learning_rate": 6.629355907799981e-07,
      "loss": 0.88443041,
      "num_input_tokens_seen": 132472475,
      "step": 6163,
      "time_per_iteration": 2.791494846343994
    },
    {
      "auxiliary_loss_clip": 0.01154148,
      "auxiliary_loss_mlp": 0.0102452,
      "balance_loss_clip": 1.04369307,
      "balance_loss_mlp": 1.01690292,
      "epoch": 0.7411771778993567,
      "flos": 30440397726720.0,
      "grad_norm": 1.5983854892380709,
      "language_loss": 0.68873477,
      "learning_rate": 6.623563840886015e-07,
      "loss": 0.71052146,
      "num_input_tokens_seen": 132493400,
      "step": 6164,
      "time_per_iteration": 2.7225863933563232
    },
    {
      "auxiliary_loss_clip": 0.01148115,
      "auxiliary_loss_mlp": 0.01027753,
      "balance_loss_clip": 1.04225874,
      "balance_loss_mlp": 1.02035594,
      "epoch": 0.7412974207899958,
      "flos": 20522050968960.0,
      "grad_norm": 1.590874097813442,
      "language_loss": 0.69388294,
      "learning_rate": 6.617773803166795e-07,
      "loss": 0.71564162,
      "num_input_tokens_seen": 132511725,
      "step": 6165,
      "time_per_iteration": 2.6535584926605225
    },
    {
      "auxiliary_loss_clip": 0.01144024,
      "auxiliary_loss_mlp": 0.00762333,
      "balance_loss_clip": 1.04511309,
      "balance_loss_mlp": 1.00046992,
      "epoch": 0.7414176636806349,
      "flos": 22090700793600.0,
      "grad_norm": 2.260000285038872,
      "language_loss": 0.81651258,
      "learning_rate": 6.611985795520634e-07,
      "loss": 0.83557612,
      "num_input_tokens_seen": 132530270,
      "step": 6166,
      "time_per_iteration": 2.705854654312134
    },
    {
      "auxiliary_loss_clip": 0.01135122,
      "auxiliary_loss_mlp": 0.0103128,
      "balance_loss_clip": 1.0463618,
      "balance_loss_mlp": 1.02372825,
      "epoch": 0.7415379065712739,
      "flos": 25155245445120.0,
      "grad_norm": 1.9417569411543147,
      "language_loss": 0.77616417,
      "learning_rate": 6.606199818825588e-07,
      "loss": 0.7978282,
      "num_input_tokens_seen": 132550725,
      "step": 6167,
      "time_per_iteration": 3.665947675704956
    },
    {
      "auxiliary_loss_clip": 0.01143141,
      "auxiliary_loss_mlp": 0.01023684,
      "balance_loss_clip": 1.04160309,
      "balance_loss_mlp": 1.01637018,
      "epoch": 0.7416581494619131,
      "flos": 16871731320960.0,
      "grad_norm": 1.9621960352584757,
      "language_loss": 0.81717956,
      "learning_rate": 6.600415873959377e-07,
      "loss": 0.83884776,
      "num_input_tokens_seen": 132568600,
      "step": 6168,
      "time_per_iteration": 2.641655206680298
    },
    {
      "auxiliary_loss_clip": 0.01094446,
      "auxiliary_loss_mlp": 0.00761831,
      "balance_loss_clip": 1.03798568,
      "balance_loss_mlp": 1.00029337,
      "epoch": 0.7417783923525522,
      "flos": 28438881102720.0,
      "grad_norm": 1.917392602003218,
      "language_loss": 0.64600259,
      "learning_rate": 6.594633961799437e-07,
      "loss": 0.66456532,
      "num_input_tokens_seen": 132587640,
      "step": 6169,
      "time_per_iteration": 2.8498239517211914
    },
    {
      "auxiliary_loss_clip": 0.01132784,
      "auxiliary_loss_mlp": 0.01029098,
      "balance_loss_clip": 1.04280353,
      "balance_loss_mlp": 1.02211237,
      "epoch": 0.7418986352431912,
      "flos": 20084299920000.0,
      "grad_norm": 7.335748716475869,
      "language_loss": 0.81735098,
      "learning_rate": 6.588854083222857e-07,
      "loss": 0.83896983,
      "num_input_tokens_seen": 132607075,
      "step": 6170,
      "time_per_iteration": 3.6786270141601562
    },
    {
      "auxiliary_loss_clip": 0.01146575,
      "auxiliary_loss_mlp": 0.01023755,
      "balance_loss_clip": 1.04670024,
      "balance_loss_mlp": 1.01657581,
      "epoch": 0.7420188781338304,
      "flos": 18259571059200.0,
      "grad_norm": 2.1381310175807644,
      "language_loss": 0.80877745,
      "learning_rate": 6.583076239106444e-07,
      "loss": 0.83048069,
      "num_input_tokens_seen": 132625580,
      "step": 6171,
      "time_per_iteration": 3.5799245834350586
    },
    {
      "auxiliary_loss_clip": 0.01145718,
      "auxiliary_loss_mlp": 0.01025826,
      "balance_loss_clip": 1.04415154,
      "balance_loss_mlp": 1.01801765,
      "epoch": 0.7421391210244694,
      "flos": 13771994319360.0,
      "grad_norm": 2.6444057254324447,
      "language_loss": 0.75089389,
      "learning_rate": 6.577300430326707e-07,
      "loss": 0.77260941,
      "num_input_tokens_seen": 132640525,
      "step": 6172,
      "time_per_iteration": 2.663411855697632
    },
    {
      "auxiliary_loss_clip": 0.01122971,
      "auxiliary_loss_mlp": 0.01023425,
      "balance_loss_clip": 1.04321146,
      "balance_loss_mlp": 1.01642704,
      "epoch": 0.7422593639151085,
      "flos": 15961683317760.0,
      "grad_norm": 2.5550883715118684,
      "language_loss": 0.7262525,
      "learning_rate": 6.571526657759821e-07,
      "loss": 0.74771643,
      "num_input_tokens_seen": 132656265,
      "step": 6173,
      "time_per_iteration": 3.7747671604156494
    },
    {
      "auxiliary_loss_clip": 0.01145114,
      "auxiliary_loss_mlp": 0.01021839,
      "balance_loss_clip": 1.04170871,
      "balance_loss_mlp": 1.01474071,
      "epoch": 0.7423796068057477,
      "flos": 30114400867200.0,
      "grad_norm": 1.5757989332268019,
      "language_loss": 0.70717388,
      "learning_rate": 6.565754922281663e-07,
      "loss": 0.72884345,
      "num_input_tokens_seen": 132678510,
      "step": 6174,
      "time_per_iteration": 2.7009828090667725
    },
    {
      "auxiliary_loss_clip": 0.01137283,
      "auxiliary_loss_mlp": 0.01024701,
      "balance_loss_clip": 1.04186022,
      "balance_loss_mlp": 1.0175184,
      "epoch": 0.7424998496963867,
      "flos": 20521907314560.0,
      "grad_norm": 1.9657415361527921,
      "language_loss": 0.7870422,
      "learning_rate": 6.559985224767801e-07,
      "loss": 0.808662,
      "num_input_tokens_seen": 132696385,
      "step": 6175,
      "time_per_iteration": 2.680285930633545
    },
    {
      "auxiliary_loss_clip": 0.01133103,
      "auxiliary_loss_mlp": 0.01024691,
      "balance_loss_clip": 1.04471266,
      "balance_loss_mlp": 1.01727569,
      "epoch": 0.7426200925870258,
      "flos": 21871573873920.0,
      "grad_norm": 2.4692111678006805,
      "language_loss": 0.75406039,
      "learning_rate": 6.55421756609349e-07,
      "loss": 0.77563834,
      "num_input_tokens_seen": 132714640,
      "step": 6176,
      "time_per_iteration": 2.643416404724121
    },
    {
      "auxiliary_loss_clip": 0.01150099,
      "auxiliary_loss_mlp": 0.01026566,
      "balance_loss_clip": 1.04653215,
      "balance_loss_mlp": 1.01871896,
      "epoch": 0.7427403354776649,
      "flos": 26432049265920.0,
      "grad_norm": 1.816109598237587,
      "language_loss": 0.79422861,
      "learning_rate": 6.54845194713369e-07,
      "loss": 0.81599522,
      "num_input_tokens_seen": 132735590,
      "step": 6177,
      "time_per_iteration": 2.6848556995391846
    },
    {
      "auxiliary_loss_clip": 0.01151869,
      "auxiliary_loss_mlp": 0.01026927,
      "balance_loss_clip": 1.04709256,
      "balance_loss_mlp": 1.01951814,
      "epoch": 0.742860578368304,
      "flos": 19898390102400.0,
      "grad_norm": 2.00371013866887,
      "language_loss": 0.80104649,
      "learning_rate": 6.542688368763034e-07,
      "loss": 0.82283449,
      "num_input_tokens_seen": 132753995,
      "step": 6178,
      "time_per_iteration": 2.590877056121826
    },
    {
      "auxiliary_loss_clip": 0.01153056,
      "auxiliary_loss_mlp": 0.01024574,
      "balance_loss_clip": 1.04698443,
      "balance_loss_mlp": 1.01750505,
      "epoch": 0.742980821258943,
      "flos": 24827201510400.0,
      "grad_norm": 1.7240918772624372,
      "language_loss": 0.76909059,
      "learning_rate": 6.536926831855854e-07,
      "loss": 0.79086685,
      "num_input_tokens_seen": 132773160,
      "step": 6179,
      "time_per_iteration": 2.672665596008301
    },
    {
      "auxiliary_loss_clip": 0.01134528,
      "auxiliary_loss_mlp": 0.0102586,
      "balance_loss_clip": 1.04347181,
      "balance_loss_mlp": 1.01896727,
      "epoch": 0.7431010641495821,
      "flos": 25228646887680.0,
      "grad_norm": 2.1188472283900994,
      "language_loss": 0.73579532,
      "learning_rate": 6.531167337286165e-07,
      "loss": 0.7573992,
      "num_input_tokens_seen": 132793180,
      "step": 6180,
      "time_per_iteration": 2.6938347816467285
    },
    {
      "auxiliary_loss_clip": 0.01136597,
      "auxiliary_loss_mlp": 0.01024819,
      "balance_loss_clip": 1.04511428,
      "balance_loss_mlp": 1.01795566,
      "epoch": 0.7432213070402213,
      "flos": 21762369550080.0,
      "grad_norm": 1.4347856741176404,
      "language_loss": 0.80086946,
      "learning_rate": 6.52540988592768e-07,
      "loss": 0.82248366,
      "num_input_tokens_seen": 132814200,
      "step": 6181,
      "time_per_iteration": 2.6921794414520264
    },
    {
      "auxiliary_loss_clip": 0.01142173,
      "auxiliary_loss_mlp": 0.0102712,
      "balance_loss_clip": 1.04395092,
      "balance_loss_mlp": 1.02014589,
      "epoch": 0.7433415499308603,
      "flos": 14793832425600.0,
      "grad_norm": 2.0101811982103635,
      "language_loss": 0.8376357,
      "learning_rate": 6.519654478653814e-07,
      "loss": 0.85932863,
      "num_input_tokens_seen": 132832565,
      "step": 6182,
      "time_per_iteration": 2.718783140182495
    },
    {
      "auxiliary_loss_clip": 0.01050848,
      "auxiliary_loss_mlp": 0.01000422,
      "balance_loss_clip": 1.01354313,
      "balance_loss_mlp": 0.99921781,
      "epoch": 0.7434617928214994,
      "flos": 67155577297920.0,
      "grad_norm": 0.7453184303340871,
      "language_loss": 0.56029809,
      "learning_rate": 6.51390111633763e-07,
      "loss": 0.58081079,
      "num_input_tokens_seen": 132897840,
      "step": 6183,
      "time_per_iteration": 3.365290880203247
    },
    {
      "auxiliary_loss_clip": 0.0109761,
      "auxiliary_loss_mlp": 0.01023665,
      "balance_loss_clip": 1.0396148,
      "balance_loss_mlp": 1.01669407,
      "epoch": 0.7435820357121385,
      "flos": 27377576928000.0,
      "grad_norm": 1.6831489474510961,
      "language_loss": 0.76521581,
      "learning_rate": 6.508149799851932e-07,
      "loss": 0.78642857,
      "num_input_tokens_seen": 132919505,
      "step": 6184,
      "time_per_iteration": 2.9354984760284424
    },
    {
      "auxiliary_loss_clip": 0.01133051,
      "auxiliary_loss_mlp": 0.01022251,
      "balance_loss_clip": 1.04284203,
      "balance_loss_mlp": 1.01578045,
      "epoch": 0.7437022786027776,
      "flos": 23987645948160.0,
      "grad_norm": 1.9986587032979253,
      "language_loss": 0.60991544,
      "learning_rate": 6.502400530069183e-07,
      "loss": 0.63146842,
      "num_input_tokens_seen": 132939390,
      "step": 6185,
      "time_per_iteration": 2.727757215499878
    },
    {
      "auxiliary_loss_clip": 0.01124675,
      "auxiliary_loss_mlp": 0.01030557,
      "balance_loss_clip": 1.04331207,
      "balance_loss_mlp": 1.02276659,
      "epoch": 0.7438225214934167,
      "flos": 21866761451520.0,
      "grad_norm": 2.02037388477216,
      "language_loss": 0.68976486,
      "learning_rate": 6.496653307861535e-07,
      "loss": 0.71131718,
      "num_input_tokens_seen": 132960060,
      "step": 6186,
      "time_per_iteration": 2.74678111076355
    },
    {
      "auxiliary_loss_clip": 0.01158623,
      "auxiliary_loss_mlp": 0.01029739,
      "balance_loss_clip": 1.04666233,
      "balance_loss_mlp": 1.02244091,
      "epoch": 0.7439427643840558,
      "flos": 20230097224320.0,
      "grad_norm": 1.9108927214583882,
      "language_loss": 0.66179669,
      "learning_rate": 6.490908134100857e-07,
      "loss": 0.6836803,
      "num_input_tokens_seen": 132978525,
      "step": 6187,
      "time_per_iteration": 2.7154691219329834
    },
    {
      "auxiliary_loss_clip": 0.0115762,
      "auxiliary_loss_mlp": 0.01028259,
      "balance_loss_clip": 1.04561043,
      "balance_loss_mlp": 1.02096963,
      "epoch": 0.7440630072746949,
      "flos": 20849915335680.0,
      "grad_norm": 2.263680346218758,
      "language_loss": 0.69878304,
      "learning_rate": 6.48516500965866e-07,
      "loss": 0.72064185,
      "num_input_tokens_seen": 132998460,
      "step": 6188,
      "time_per_iteration": 2.7056281566619873
    },
    {
      "auxiliary_loss_clip": 0.01156016,
      "auxiliary_loss_mlp": 0.01022572,
      "balance_loss_clip": 1.04372478,
      "balance_loss_mlp": 1.01570833,
      "epoch": 0.7441832501653339,
      "flos": 26503762769280.0,
      "grad_norm": 1.6387093689579877,
      "language_loss": 0.81713057,
      "learning_rate": 6.479423935406192e-07,
      "loss": 0.83891642,
      "num_input_tokens_seen": 133018445,
      "step": 6189,
      "time_per_iteration": 2.710979461669922
    },
    {
      "auxiliary_loss_clip": 0.01039946,
      "auxiliary_loss_mlp": 0.0100231,
      "balance_loss_clip": 1.01147199,
      "balance_loss_mlp": 1.00123084,
      "epoch": 0.7443034930559731,
      "flos": 68602848088320.0,
      "grad_norm": 0.8131682532173082,
      "language_loss": 0.62020528,
      "learning_rate": 6.473684912214357e-07,
      "loss": 0.64062786,
      "num_input_tokens_seen": 133082005,
      "step": 6190,
      "time_per_iteration": 3.3992176055908203
    },
    {
      "auxiliary_loss_clip": 0.01153844,
      "auxiliary_loss_mlp": 0.01020349,
      "balance_loss_clip": 1.04664409,
      "balance_loss_mlp": 1.01240098,
      "epoch": 0.7444237359466122,
      "flos": 18654982951680.0,
      "grad_norm": 2.0435947838739406,
      "language_loss": 0.69623053,
      "learning_rate": 6.467947940953778e-07,
      "loss": 0.71797246,
      "num_input_tokens_seen": 133100530,
      "step": 6191,
      "time_per_iteration": 2.652797222137451
    },
    {
      "auxiliary_loss_clip": 0.01137913,
      "auxiliary_loss_mlp": 0.0102193,
      "balance_loss_clip": 1.04418826,
      "balance_loss_mlp": 1.01517045,
      "epoch": 0.7445439788372512,
      "flos": 22817604326400.0,
      "grad_norm": 1.8150350836994886,
      "language_loss": 0.72858584,
      "learning_rate": 6.462213022494732e-07,
      "loss": 0.75018424,
      "num_input_tokens_seen": 133119775,
      "step": 6192,
      "time_per_iteration": 2.6620213985443115
    },
    {
      "auxiliary_loss_clip": 0.01058806,
      "auxiliary_loss_mlp": 0.01001157,
      "balance_loss_clip": 1.01320601,
      "balance_loss_mlp": 1.00005388,
      "epoch": 0.7446642217278904,
      "flos": 67045690615680.0,
      "grad_norm": 0.7703784984523696,
      "language_loss": 0.609743,
      "learning_rate": 6.456480157707201e-07,
      "loss": 0.6303426,
      "num_input_tokens_seen": 133184550,
      "step": 6193,
      "time_per_iteration": 4.086431264877319
    },
    {
      "auxiliary_loss_clip": 0.01119351,
      "auxiliary_loss_mlp": 0.01030476,
      "balance_loss_clip": 1.04086328,
      "balance_loss_mlp": 1.02318311,
      "epoch": 0.7447844646185294,
      "flos": 17417465631360.0,
      "grad_norm": 2.242495338294364,
      "language_loss": 0.84979081,
      "learning_rate": 6.450749347460866e-07,
      "loss": 0.87128907,
      "num_input_tokens_seen": 133201525,
      "step": 6194,
      "time_per_iteration": 2.724835157394409
    },
    {
      "auxiliary_loss_clip": 0.01166589,
      "auxiliary_loss_mlp": 0.01026241,
      "balance_loss_clip": 1.0457859,
      "balance_loss_mlp": 1.01919556,
      "epoch": 0.7449047075091685,
      "flos": 26615876094720.0,
      "grad_norm": 1.8996971260714226,
      "language_loss": 0.79105735,
      "learning_rate": 6.445020592625083e-07,
      "loss": 0.81298566,
      "num_input_tokens_seen": 133222175,
      "step": 6195,
      "time_per_iteration": 2.7371091842651367
    },
    {
      "auxiliary_loss_clip": 0.01164583,
      "auxiliary_loss_mlp": 0.01024413,
      "balance_loss_clip": 1.04396796,
      "balance_loss_mlp": 1.01793718,
      "epoch": 0.7450249503998077,
      "flos": 14170458867840.0,
      "grad_norm": 1.9934078505012374,
      "language_loss": 0.80482644,
      "learning_rate": 6.4392938940689e-07,
      "loss": 0.82671642,
      "num_input_tokens_seen": 133237590,
      "step": 6196,
      "time_per_iteration": 3.5921947956085205
    },
    {
      "auxiliary_loss_clip": 0.01108994,
      "auxiliary_loss_mlp": 0.00762413,
      "balance_loss_clip": 1.04156744,
      "balance_loss_mlp": 1.00039983,
      "epoch": 0.7451451932904467,
      "flos": 19606687752960.0,
      "grad_norm": 2.119713770102674,
      "language_loss": 0.71533179,
      "learning_rate": 6.433569252661049e-07,
      "loss": 0.73404586,
      "num_input_tokens_seen": 133255590,
      "step": 6197,
      "time_per_iteration": 3.598384141921997
    },
    {
      "auxiliary_loss_clip": 0.01116992,
      "auxiliary_loss_mlp": 0.01025382,
      "balance_loss_clip": 1.04030859,
      "balance_loss_mlp": 1.01888561,
      "epoch": 0.7452654361810858,
      "flos": 12495405980160.0,
      "grad_norm": 2.6214943572619287,
      "language_loss": 0.71287155,
      "learning_rate": 6.427846669269952e-07,
      "loss": 0.73429537,
      "num_input_tokens_seen": 133273210,
      "step": 6198,
      "time_per_iteration": 2.7548880577087402
    },
    {
      "auxiliary_loss_clip": 0.01172093,
      "auxiliary_loss_mlp": 0.01023313,
      "balance_loss_clip": 1.05139315,
      "balance_loss_mlp": 1.01636612,
      "epoch": 0.7453856790717249,
      "flos": 22127329687680.0,
      "grad_norm": 2.066059859363878,
      "language_loss": 0.82059526,
      "learning_rate": 6.422126144763729e-07,
      "loss": 0.84254932,
      "num_input_tokens_seen": 133292600,
      "step": 6199,
      "time_per_iteration": 3.5965919494628906
    },
    {
      "auxiliary_loss_clip": 0.0112144,
      "auxiliary_loss_mlp": 0.00762448,
      "balance_loss_clip": 1.03797698,
      "balance_loss_mlp": 1.00034308,
      "epoch": 0.745505921962364,
      "flos": 20010682995840.0,
      "grad_norm": 2.177534380578065,
      "language_loss": 0.76929462,
      "learning_rate": 6.416407680010174e-07,
      "loss": 0.7881335,
      "num_input_tokens_seen": 133306960,
      "step": 6200,
      "time_per_iteration": 2.6976382732391357
    },
    {
      "auxiliary_loss_clip": 0.01129615,
      "auxiliary_loss_mlp": 0.01024936,
      "balance_loss_clip": 1.04622293,
      "balance_loss_mlp": 1.01782823,
      "epoch": 0.745626164853003,
      "flos": 24677884673280.0,
      "grad_norm": 1.9918685500171576,
      "language_loss": 0.81095117,
      "learning_rate": 6.410691275876774e-07,
      "loss": 0.83249676,
      "num_input_tokens_seen": 133326380,
      "step": 6201,
      "time_per_iteration": 2.7952194213867188
    },
    {
      "auxiliary_loss_clip": 0.01148004,
      "auxiliary_loss_mlp": 0.01029834,
      "balance_loss_clip": 1.04552293,
      "balance_loss_mlp": 1.02254128,
      "epoch": 0.7457464077436422,
      "flos": 14538830797440.0,
      "grad_norm": 2.320940626682077,
      "language_loss": 0.76632571,
      "learning_rate": 6.404976933230704e-07,
      "loss": 0.78810412,
      "num_input_tokens_seen": 133342900,
      "step": 6202,
      "time_per_iteration": 2.6755716800689697
    },
    {
      "auxiliary_loss_clip": 0.01145553,
      "auxiliary_loss_mlp": 0.01024914,
      "balance_loss_clip": 1.04513848,
      "balance_loss_mlp": 1.01724267,
      "epoch": 0.7458666506342813,
      "flos": 34021194600960.0,
      "grad_norm": 2.0306120178633646,
      "language_loss": 0.72755754,
      "learning_rate": 6.399264652938813e-07,
      "loss": 0.74926221,
      "num_input_tokens_seen": 133363805,
      "step": 6203,
      "time_per_iteration": 2.762225866317749
    },
    {
      "auxiliary_loss_clip": 0.01136895,
      "auxiliary_loss_mlp": 0.0102437,
      "balance_loss_clip": 1.04284859,
      "balance_loss_mlp": 1.01728022,
      "epoch": 0.7459868935249203,
      "flos": 24279025075200.0,
      "grad_norm": 1.8858521106181385,
      "language_loss": 0.74555045,
      "learning_rate": 6.393554435867679e-07,
      "loss": 0.76716316,
      "num_input_tokens_seen": 133384655,
      "step": 6204,
      "time_per_iteration": 2.725489377975464
    },
    {
      "auxiliary_loss_clip": 0.01120467,
      "auxiliary_loss_mlp": 0.01031119,
      "balance_loss_clip": 1.04069018,
      "balance_loss_mlp": 1.02393639,
      "epoch": 0.7461071364155595,
      "flos": 21908777385600.0,
      "grad_norm": 4.475044887978384,
      "language_loss": 0.83342665,
      "learning_rate": 6.387846282883502e-07,
      "loss": 0.8549425,
      "num_input_tokens_seen": 133401185,
      "step": 6205,
      "time_per_iteration": 2.6904854774475098
    },
    {
      "auxiliary_loss_clip": 0.0116907,
      "auxiliary_loss_mlp": 0.01024745,
      "balance_loss_clip": 1.04903984,
      "balance_loss_mlp": 1.01740813,
      "epoch": 0.7462273793061985,
      "flos": 22889712879360.0,
      "grad_norm": 2.207906022797722,
      "language_loss": 0.76870573,
      "learning_rate": 6.38214019485223e-07,
      "loss": 0.79064387,
      "num_input_tokens_seen": 133420010,
      "step": 6206,
      "time_per_iteration": 2.6240029335021973
    },
    {
      "auxiliary_loss_clip": 0.01092009,
      "auxiliary_loss_mlp": 0.01022763,
      "balance_loss_clip": 1.03897583,
      "balance_loss_mlp": 1.01531529,
      "epoch": 0.7463476221968376,
      "flos": 19968451580160.0,
      "grad_norm": 2.1413635296819353,
      "language_loss": 0.72002679,
      "learning_rate": 6.376436172639461e-07,
      "loss": 0.74117458,
      "num_input_tokens_seen": 133437855,
      "step": 6207,
      "time_per_iteration": 2.762192726135254
    },
    {
      "auxiliary_loss_clip": 0.01086217,
      "auxiliary_loss_mlp": 0.0102523,
      "balance_loss_clip": 1.03874004,
      "balance_loss_mlp": 1.0176245,
      "epoch": 0.7464678650874768,
      "flos": 16836610798080.0,
      "grad_norm": 7.172271799418539,
      "language_loss": 0.65237814,
      "learning_rate": 6.370734217110487e-07,
      "loss": 0.67349261,
      "num_input_tokens_seen": 133456600,
      "step": 6208,
      "time_per_iteration": 2.785538911819458
    },
    {
      "auxiliary_loss_clip": 0.01146005,
      "auxiliary_loss_mlp": 0.01023859,
      "balance_loss_clip": 1.04878271,
      "balance_loss_mlp": 1.01612854,
      "epoch": 0.7465881079781158,
      "flos": 48100869843840.0,
      "grad_norm": 1.424048303769729,
      "language_loss": 0.64227974,
      "learning_rate": 6.36503432913031e-07,
      "loss": 0.66397834,
      "num_input_tokens_seen": 133479745,
      "step": 6209,
      "time_per_iteration": 2.8787760734558105
    },
    {
      "auxiliary_loss_clip": 0.01152254,
      "auxiliary_loss_mlp": 0.01021465,
      "balance_loss_clip": 1.04579377,
      "balance_loss_mlp": 1.01402009,
      "epoch": 0.7467083508687549,
      "flos": 19677359761920.0,
      "grad_norm": 2.0549736748124747,
      "language_loss": 0.69023716,
      "learning_rate": 6.359336509563569e-07,
      "loss": 0.71197432,
      "num_input_tokens_seen": 133495765,
      "step": 6210,
      "time_per_iteration": 2.601499557495117
    },
    {
      "auxiliary_loss_clip": 0.01115892,
      "auxiliary_loss_mlp": 0.01023254,
      "balance_loss_clip": 1.04204023,
      "balance_loss_mlp": 1.01638198,
      "epoch": 0.7468285937593939,
      "flos": 17895436934400.0,
      "grad_norm": 1.9197369274249187,
      "language_loss": 0.80489951,
      "learning_rate": 6.353640759274641e-07,
      "loss": 0.82629097,
      "num_input_tokens_seen": 133514655,
      "step": 6211,
      "time_per_iteration": 2.685817003250122
    },
    {
      "auxiliary_loss_clip": 0.01152078,
      "auxiliary_loss_mlp": 0.01022784,
      "balance_loss_clip": 1.04339683,
      "balance_loss_mlp": 1.01545,
      "epoch": 0.7469488366500331,
      "flos": 23141446369920.0,
      "grad_norm": 3.022521627437731,
      "language_loss": 0.74735475,
      "learning_rate": 6.347947079127556e-07,
      "loss": 0.76910329,
      "num_input_tokens_seen": 133532555,
      "step": 6212,
      "time_per_iteration": 2.6430695056915283
    },
    {
      "auxiliary_loss_clip": 0.01132323,
      "auxiliary_loss_mlp": 0.0102735,
      "balance_loss_clip": 1.04197931,
      "balance_loss_mlp": 1.02001524,
      "epoch": 0.7470690795406721,
      "flos": 16690849407360.0,
      "grad_norm": 2.2250280184706046,
      "language_loss": 0.76897228,
      "learning_rate": 6.342255469986053e-07,
      "loss": 0.79056901,
      "num_input_tokens_seen": 133551300,
      "step": 6213,
      "time_per_iteration": 2.6677963733673096
    },
    {
      "auxiliary_loss_clip": 0.0116835,
      "auxiliary_loss_mlp": 0.01024922,
      "balance_loss_clip": 1.04816616,
      "balance_loss_mlp": 1.0183351,
      "epoch": 0.7471893224313112,
      "flos": 25192700352000.0,
      "grad_norm": 1.9595972353033755,
      "language_loss": 0.76286519,
      "learning_rate": 6.336565932713533e-07,
      "loss": 0.78479791,
      "num_input_tokens_seen": 133570725,
      "step": 6214,
      "time_per_iteration": 2.7073559761047363
    },
    {
      "auxiliary_loss_clip": 0.01136412,
      "auxiliary_loss_mlp": 0.01024519,
      "balance_loss_clip": 1.04531765,
      "balance_loss_mlp": 1.01697028,
      "epoch": 0.7473095653219504,
      "flos": 22526225199360.0,
      "grad_norm": 1.9266797020248434,
      "language_loss": 0.77770662,
      "learning_rate": 6.330878468173088e-07,
      "loss": 0.79931593,
      "num_input_tokens_seen": 133590790,
      "step": 6215,
      "time_per_iteration": 2.7077155113220215
    },
    {
      "auxiliary_loss_clip": 0.0114552,
      "auxiliary_loss_mlp": 0.01026175,
      "balance_loss_clip": 1.0430181,
      "balance_loss_mlp": 1.01931417,
      "epoch": 0.7474298082125894,
      "flos": 18113989236480.0,
      "grad_norm": 1.7342427734497174,
      "language_loss": 0.73191398,
      "learning_rate": 6.32519307722752e-07,
      "loss": 0.75363088,
      "num_input_tokens_seen": 133608685,
      "step": 6216,
      "time_per_iteration": 2.5879409313201904
    },
    {
      "auxiliary_loss_clip": 0.01036042,
      "auxiliary_loss_mlp": 0.01002538,
      "balance_loss_clip": 1.01693916,
      "balance_loss_mlp": 1.00128603,
      "epoch": 0.7475500511032285,
      "flos": 62086535193600.0,
      "grad_norm": 0.8114796853298277,
      "language_loss": 0.54947698,
      "learning_rate": 6.31950976073929e-07,
      "loss": 0.56986284,
      "num_input_tokens_seen": 133662775,
      "step": 6217,
      "time_per_iteration": 3.254901647567749
    },
    {
      "auxiliary_loss_clip": 0.0110598,
      "auxiliary_loss_mlp": 0.01022972,
      "balance_loss_clip": 1.04088235,
      "balance_loss_mlp": 1.01577783,
      "epoch": 0.7476702939938676,
      "flos": 17785586165760.0,
      "grad_norm": 2.313469277267881,
      "language_loss": 0.80730742,
      "learning_rate": 6.31382851957055e-07,
      "loss": 0.82859683,
      "num_input_tokens_seen": 133679595,
      "step": 6218,
      "time_per_iteration": 2.74605131149292
    },
    {
      "auxiliary_loss_clip": 0.01123067,
      "auxiliary_loss_mlp": 0.00761962,
      "balance_loss_clip": 1.04384816,
      "balance_loss_mlp": 1.0003432,
      "epoch": 0.7477905368845067,
      "flos": 27927944092800.0,
      "grad_norm": 2.029662682670831,
      "language_loss": 0.7143659,
      "learning_rate": 6.308149354583143e-07,
      "loss": 0.73321617,
      "num_input_tokens_seen": 133699000,
      "step": 6219,
      "time_per_iteration": 3.669180393218994
    },
    {
      "auxiliary_loss_clip": 0.01159456,
      "auxiliary_loss_mlp": 0.01024154,
      "balance_loss_clip": 1.0477128,
      "balance_loss_mlp": 1.01677191,
      "epoch": 0.7479107797751458,
      "flos": 26870374932480.0,
      "grad_norm": 6.392390538490855,
      "language_loss": 0.81786942,
      "learning_rate": 6.302472266638586e-07,
      "loss": 0.83970553,
      "num_input_tokens_seen": 133719540,
      "step": 6220,
      "time_per_iteration": 2.685568332672119
    },
    {
      "auxiliary_loss_clip": 0.01172781,
      "auxiliary_loss_mlp": 0.01027566,
      "balance_loss_clip": 1.04796565,
      "balance_loss_mlp": 1.01853609,
      "epoch": 0.7480310226657849,
      "flos": 33943375785600.0,
      "grad_norm": 2.8540530534478967,
      "language_loss": 0.69759542,
      "learning_rate": 6.296797256598101e-07,
      "loss": 0.71959889,
      "num_input_tokens_seen": 133741020,
      "step": 6221,
      "time_per_iteration": 2.700759172439575
    },
    {
      "auxiliary_loss_clip": 0.0111621,
      "auxiliary_loss_mlp": 0.01023856,
      "balance_loss_clip": 1.0404377,
      "balance_loss_mlp": 1.01720679,
      "epoch": 0.748151265556424,
      "flos": 24826555065600.0,
      "grad_norm": 1.7188702061654149,
      "language_loss": 0.81237924,
      "learning_rate": 6.291124325322576e-07,
      "loss": 0.83377987,
      "num_input_tokens_seen": 133761145,
      "step": 6222,
      "time_per_iteration": 3.7109224796295166
    },
    {
      "auxiliary_loss_clip": 0.01145707,
      "auxiliary_loss_mlp": 0.01025544,
      "balance_loss_clip": 1.04561114,
      "balance_loss_mlp": 1.01832545,
      "epoch": 0.748271508447063,
      "flos": 38399351535360.0,
      "grad_norm": 1.6779396215167273,
      "language_loss": 0.62579292,
      "learning_rate": 6.285453473672595e-07,
      "loss": 0.6475054,
      "num_input_tokens_seen": 133783715,
      "step": 6223,
      "time_per_iteration": 3.6455330848693848
    },
    {
      "auxiliary_loss_clip": 0.01163661,
      "auxiliary_loss_mlp": 0.01018872,
      "balance_loss_clip": 1.04399359,
      "balance_loss_mlp": 1.01211619,
      "epoch": 0.7483917513377022,
      "flos": 21541842000000.0,
      "grad_norm": 1.8781218910827473,
      "language_loss": 0.75506663,
      "learning_rate": 6.279784702508415e-07,
      "loss": 0.77689195,
      "num_input_tokens_seen": 133804465,
      "step": 6224,
      "time_per_iteration": 2.61055588722229
    },
    {
      "auxiliary_loss_clip": 0.01041667,
      "auxiliary_loss_mlp": 0.01001958,
      "balance_loss_clip": 1.01223099,
      "balance_loss_mlp": 1.00086176,
      "epoch": 0.7485119942283412,
      "flos": 62314532772480.0,
      "grad_norm": 0.7871278899150429,
      "language_loss": 0.58513474,
      "learning_rate": 6.274118012689979e-07,
      "loss": 0.60557103,
      "num_input_tokens_seen": 133866365,
      "step": 6225,
      "time_per_iteration": 4.301759958267212
    },
    {
      "auxiliary_loss_clip": 0.01132988,
      "auxiliary_loss_mlp": 0.0102583,
      "balance_loss_clip": 1.04318023,
      "balance_loss_mlp": 1.01863313,
      "epoch": 0.7486322371189803,
      "flos": 29937613104000.0,
      "grad_norm": 1.5201661080899442,
      "language_loss": 0.68416023,
      "learning_rate": 6.268453405076943e-07,
      "loss": 0.70574844,
      "num_input_tokens_seen": 133888760,
      "step": 6226,
      "time_per_iteration": 2.747171401977539
    },
    {
      "auxiliary_loss_clip": 0.01141902,
      "auxiliary_loss_mlp": 0.01023239,
      "balance_loss_clip": 1.04579926,
      "balance_loss_mlp": 1.01594329,
      "epoch": 0.7487524800096195,
      "flos": 18949414734720.0,
      "grad_norm": 1.9065342277314996,
      "language_loss": 0.82434368,
      "learning_rate": 6.262790880528592e-07,
      "loss": 0.84599507,
      "num_input_tokens_seen": 133906380,
      "step": 6227,
      "time_per_iteration": 2.6401023864746094
    },
    {
      "auxiliary_loss_clip": 0.01135495,
      "auxiliary_loss_mlp": 0.01023107,
      "balance_loss_clip": 1.04067087,
      "balance_loss_mlp": 1.01534665,
      "epoch": 0.7488727229002585,
      "flos": 18697393935360.0,
      "grad_norm": 2.769678878544884,
      "language_loss": 0.8004936,
      "learning_rate": 6.257130439903951e-07,
      "loss": 0.82207966,
      "num_input_tokens_seen": 133922875,
      "step": 6228,
      "time_per_iteration": 2.797952890396118
    },
    {
      "auxiliary_loss_clip": 0.01171787,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.05088305,
      "balance_loss_mlp": 1.02200902,
      "epoch": 0.7489929657908976,
      "flos": 23623368168960.0,
      "grad_norm": 2.3649690285552265,
      "language_loss": 0.81214434,
      "learning_rate": 6.251472084061695e-07,
      "loss": 0.83415353,
      "num_input_tokens_seen": 133941795,
      "step": 6229,
      "time_per_iteration": 2.654466152191162
    },
    {
      "auxiliary_loss_clip": 0.01152695,
      "auxiliary_loss_mlp": 0.01029301,
      "balance_loss_clip": 1.04611039,
      "balance_loss_mlp": 1.0226047,
      "epoch": 0.7491132086815367,
      "flos": 20551533056640.0,
      "grad_norm": 2.1104288353780696,
      "language_loss": 0.89003074,
      "learning_rate": 6.245815813860191e-07,
      "loss": 0.91185069,
      "num_input_tokens_seen": 133957305,
      "step": 6230,
      "time_per_iteration": 2.674736976623535
    },
    {
      "auxiliary_loss_clip": 0.01168995,
      "auxiliary_loss_mlp": 0.01025352,
      "balance_loss_clip": 1.04644489,
      "balance_loss_mlp": 1.01758599,
      "epoch": 0.7492334515721758,
      "flos": 23003011353600.0,
      "grad_norm": 1.966029225582525,
      "language_loss": 0.70264196,
      "learning_rate": 6.240161630157495e-07,
      "loss": 0.72458547,
      "num_input_tokens_seen": 133976660,
      "step": 6231,
      "time_per_iteration": 2.6121299266815186
    },
    {
      "auxiliary_loss_clip": 0.01173144,
      "auxiliary_loss_mlp": 0.01024055,
      "balance_loss_clip": 1.04872179,
      "balance_loss_mlp": 1.01674449,
      "epoch": 0.7493536944628149,
      "flos": 16398823835520.0,
      "grad_norm": 2.8033757958578116,
      "language_loss": 0.70824087,
      "learning_rate": 6.23450953381133e-07,
      "loss": 0.73021287,
      "num_input_tokens_seen": 133994750,
      "step": 6232,
      "time_per_iteration": 2.5583012104034424
    },
    {
      "auxiliary_loss_clip": 0.01131681,
      "auxiliary_loss_mlp": 0.0102197,
      "balance_loss_clip": 1.04079473,
      "balance_loss_mlp": 1.01529408,
      "epoch": 0.749473937353454,
      "flos": 15338561155200.0,
      "grad_norm": 2.071902799841961,
      "language_loss": 0.68056297,
      "learning_rate": 6.228859525679131e-07,
      "loss": 0.70209944,
      "num_input_tokens_seen": 134009165,
      "step": 6233,
      "time_per_iteration": 2.71573805809021
    },
    {
      "auxiliary_loss_clip": 0.01154081,
      "auxiliary_loss_mlp": 0.01022456,
      "balance_loss_clip": 1.0457387,
      "balance_loss_mlp": 1.01558995,
      "epoch": 0.7495941802440931,
      "flos": 18951138587520.0,
      "grad_norm": 2.0610921477578454,
      "language_loss": 0.80419755,
      "learning_rate": 6.223211606617986e-07,
      "loss": 0.8259629,
      "num_input_tokens_seen": 134027585,
      "step": 6234,
      "time_per_iteration": 2.616936683654785
    },
    {
      "auxiliary_loss_clip": 0.01148718,
      "auxiliary_loss_mlp": 0.01026941,
      "balance_loss_clip": 1.04645324,
      "balance_loss_mlp": 1.02069139,
      "epoch": 0.7497144231347321,
      "flos": 22492469393280.0,
      "grad_norm": 1.8414731788455145,
      "language_loss": 0.84365821,
      "learning_rate": 6.217565777484701e-07,
      "loss": 0.8654148,
      "num_input_tokens_seen": 134046680,
      "step": 6235,
      "time_per_iteration": 2.6252260208129883
    },
    {
      "auxiliary_loss_clip": 0.01140598,
      "auxiliary_loss_mlp": 0.00761927,
      "balance_loss_clip": 1.04623854,
      "balance_loss_mlp": 1.00031388,
      "epoch": 0.7498346660253713,
      "flos": 24243509502720.0,
      "grad_norm": 2.4115485461250405,
      "language_loss": 0.80196118,
      "learning_rate": 6.211922039135722e-07,
      "loss": 0.82098639,
      "num_input_tokens_seen": 134066825,
      "step": 6236,
      "time_per_iteration": 2.7281253337860107
    },
    {
      "auxiliary_loss_clip": 0.01167969,
      "auxiliary_loss_mlp": 0.01027233,
      "balance_loss_clip": 1.04765785,
      "balance_loss_mlp": 1.02029192,
      "epoch": 0.7499549089160104,
      "flos": 24387080163840.0,
      "grad_norm": 1.898353851000896,
      "language_loss": 0.81011218,
      "learning_rate": 6.206280392427201e-07,
      "loss": 0.83206415,
      "num_input_tokens_seen": 134086410,
      "step": 6237,
      "time_per_iteration": 2.658667802810669
    },
    {
      "auxiliary_loss_clip": 0.01146407,
      "auxiliary_loss_mlp": 0.01024703,
      "balance_loss_clip": 1.0429287,
      "balance_loss_mlp": 1.01772928,
      "epoch": 0.7500751518066494,
      "flos": 34057320704640.0,
      "grad_norm": 1.7419043120328517,
      "language_loss": 0.73659861,
      "learning_rate": 6.200640838214983e-07,
      "loss": 0.75830972,
      "num_input_tokens_seen": 134109185,
      "step": 6238,
      "time_per_iteration": 2.7898871898651123
    },
    {
      "auxiliary_loss_clip": 0.0116755,
      "auxiliary_loss_mlp": 0.0102161,
      "balance_loss_clip": 1.04710031,
      "balance_loss_mlp": 1.01506889,
      "epoch": 0.7501953946972886,
      "flos": 18843586289280.0,
      "grad_norm": 2.100078406878541,
      "language_loss": 0.67188585,
      "learning_rate": 6.195003377354578e-07,
      "loss": 0.6937775,
      "num_input_tokens_seen": 134128455,
      "step": 6239,
      "time_per_iteration": 2.606393337249756
    },
    {
      "auxiliary_loss_clip": 0.01148993,
      "auxiliary_loss_mlp": 0.01026965,
      "balance_loss_clip": 1.04317188,
      "balance_loss_mlp": 1.01985395,
      "epoch": 0.7503156375879276,
      "flos": 20257675891200.0,
      "grad_norm": 2.740858987216959,
      "language_loss": 0.73855174,
      "learning_rate": 6.189368010701183e-07,
      "loss": 0.76031125,
      "num_input_tokens_seen": 134145515,
      "step": 6240,
      "time_per_iteration": 2.740865468978882
    },
    {
      "auxiliary_loss_clip": 0.01158915,
      "auxiliary_loss_mlp": 0.01023662,
      "balance_loss_clip": 1.04536927,
      "balance_loss_mlp": 1.01658702,
      "epoch": 0.7504358804785667,
      "flos": 13480040574720.0,
      "grad_norm": 1.8732397830881025,
      "language_loss": 0.76746249,
      "learning_rate": 6.183734739109683e-07,
      "loss": 0.78928822,
      "num_input_tokens_seen": 134163335,
      "step": 6241,
      "time_per_iteration": 2.6277778148651123
    },
    {
      "auxiliary_loss_clip": 0.01162419,
      "auxiliary_loss_mlp": 0.01027703,
      "balance_loss_clip": 1.04798865,
      "balance_loss_mlp": 1.02012706,
      "epoch": 0.7505561233692057,
      "flos": 29461042431360.0,
      "grad_norm": 2.387386378919897,
      "language_loss": 0.68524528,
      "learning_rate": 6.178103563434629e-07,
      "loss": 0.70714647,
      "num_input_tokens_seen": 134182335,
      "step": 6242,
      "time_per_iteration": 2.6819751262664795
    },
    {
      "auxiliary_loss_clip": 0.0116893,
      "auxiliary_loss_mlp": 0.01025698,
      "balance_loss_clip": 1.04800785,
      "balance_loss_mlp": 1.01837599,
      "epoch": 0.7506763662598449,
      "flos": 20302457172480.0,
      "grad_norm": 1.5757705144198286,
      "language_loss": 0.83963561,
      "learning_rate": 6.172474484530283e-07,
      "loss": 0.86158192,
      "num_input_tokens_seen": 134201070,
      "step": 6243,
      "time_per_iteration": 2.655743360519409
    },
    {
      "auxiliary_loss_clip": 0.01127742,
      "auxiliary_loss_mlp": 0.01024089,
      "balance_loss_clip": 1.0398035,
      "balance_loss_mlp": 1.01691318,
      "epoch": 0.750796609150484,
      "flos": 37230961939200.0,
      "grad_norm": 1.8384526624224773,
      "language_loss": 0.76193881,
      "learning_rate": 6.166847503250563e-07,
      "loss": 0.78345716,
      "num_input_tokens_seen": 134223310,
      "step": 6244,
      "time_per_iteration": 2.829166889190674
    },
    {
      "auxiliary_loss_clip": 0.01138492,
      "auxiliary_loss_mlp": 0.01022396,
      "balance_loss_clip": 1.04188788,
      "balance_loss_mlp": 1.01467109,
      "epoch": 0.750916852041123,
      "flos": 19609417186560.0,
      "grad_norm": 4.3962857688074815,
      "language_loss": 0.79270506,
      "learning_rate": 6.161222620449078e-07,
      "loss": 0.81431389,
      "num_input_tokens_seen": 134242085,
      "step": 6245,
      "time_per_iteration": 3.6289732456207275
    },
    {
      "auxiliary_loss_clip": 0.01127559,
      "auxiliary_loss_mlp": 0.01023462,
      "balance_loss_clip": 1.0438149,
      "balance_loss_mlp": 1.01585054,
      "epoch": 0.7510370949317622,
      "flos": 25112690807040.0,
      "grad_norm": 2.173995947960605,
      "language_loss": 0.80073798,
      "learning_rate": 6.155599836979117e-07,
      "loss": 0.82224822,
      "num_input_tokens_seen": 134260770,
      "step": 6246,
      "time_per_iteration": 2.7568163871765137
    },
    {
      "auxiliary_loss_clip": 0.01111068,
      "auxiliary_loss_mlp": 0.01027821,
      "balance_loss_clip": 1.0392102,
      "balance_loss_mlp": 1.02051318,
      "epoch": 0.7511573378224012,
      "flos": 19062282245760.0,
      "grad_norm": 1.9568674146395753,
      "language_loss": 0.81919175,
      "learning_rate": 6.149979153693649e-07,
      "loss": 0.84058058,
      "num_input_tokens_seen": 134278025,
      "step": 6247,
      "time_per_iteration": 2.7254300117492676
    },
    {
      "auxiliary_loss_clip": 0.01151893,
      "auxiliary_loss_mlp": 0.01030623,
      "balance_loss_clip": 1.0448662,
      "balance_loss_mlp": 1.0231545,
      "epoch": 0.7512775807130403,
      "flos": 19937676602880.0,
      "grad_norm": 1.902443383708817,
      "language_loss": 0.76787221,
      "learning_rate": 6.144360571445343e-07,
      "loss": 0.78969735,
      "num_input_tokens_seen": 134297170,
      "step": 6248,
      "time_per_iteration": 3.6428720951080322
    },
    {
      "auxiliary_loss_clip": 0.0115165,
      "auxiliary_loss_mlp": 0.01027596,
      "balance_loss_clip": 1.04630089,
      "balance_loss_mlp": 1.02063775,
      "epoch": 0.7513978236036795,
      "flos": 20739920912640.0,
      "grad_norm": 1.6579267789732461,
      "language_loss": 0.80161631,
      "learning_rate": 6.138744091086509e-07,
      "loss": 0.82340872,
      "num_input_tokens_seen": 134316755,
      "step": 6249,
      "time_per_iteration": 3.6026813983917236
    },
    {
      "auxiliary_loss_clip": 0.01130555,
      "auxiliary_loss_mlp": 0.01025326,
      "balance_loss_clip": 1.04616833,
      "balance_loss_mlp": 1.01805711,
      "epoch": 0.7515180664943185,
      "flos": 27563163523200.0,
      "grad_norm": 2.2298137738863137,
      "language_loss": 0.73009145,
      "learning_rate": 6.133129713469183e-07,
      "loss": 0.75165021,
      "num_input_tokens_seen": 134335960,
      "step": 6250,
      "time_per_iteration": 2.8123538494110107
    },
    {
      "auxiliary_loss_clip": 0.0113161,
      "auxiliary_loss_mlp": 0.01021473,
      "balance_loss_clip": 1.03925323,
      "balance_loss_mlp": 1.01411736,
      "epoch": 0.7516383093849576,
      "flos": 33803181002880.0,
      "grad_norm": 1.9319404483968476,
      "language_loss": 0.64297009,
      "learning_rate": 6.127517439445053e-07,
      "loss": 0.66450095,
      "num_input_tokens_seen": 134356805,
      "step": 6251,
      "time_per_iteration": 3.787219762802124
    },
    {
      "auxiliary_loss_clip": 0.01106763,
      "auxiliary_loss_mlp": 0.01025151,
      "balance_loss_clip": 1.04335487,
      "balance_loss_mlp": 1.01821876,
      "epoch": 0.7517585522755967,
      "flos": 29746172592000.0,
      "grad_norm": 2.714070573129891,
      "language_loss": 0.81470859,
      "learning_rate": 6.121907269865498e-07,
      "loss": 0.83602774,
      "num_input_tokens_seen": 134376295,
      "step": 6252,
      "time_per_iteration": 2.7680513858795166
    },
    {
      "auxiliary_loss_clip": 0.01031149,
      "auxiliary_loss_mlp": 0.01002061,
      "balance_loss_clip": 1.01180577,
      "balance_loss_mlp": 1.0010004,
      "epoch": 0.7518787951662358,
      "flos": 69807974319360.0,
      "grad_norm": 0.9216200366596135,
      "language_loss": 0.67246389,
      "learning_rate": 6.116299205581577e-07,
      "loss": 0.69279599,
      "num_input_tokens_seen": 134431125,
      "step": 6253,
      "time_per_iteration": 3.250988721847534
    },
    {
      "auxiliary_loss_clip": 0.0117543,
      "auxiliary_loss_mlp": 0.01023179,
      "balance_loss_clip": 1.05035937,
      "balance_loss_mlp": 1.01551056,
      "epoch": 0.7519990380568748,
      "flos": 34203225749760.0,
      "grad_norm": 1.8874540851359076,
      "language_loss": 0.68144083,
      "learning_rate": 6.110693247444018e-07,
      "loss": 0.70342696,
      "num_input_tokens_seen": 134452960,
      "step": 6254,
      "time_per_iteration": 2.6999900341033936
    },
    {
      "auxiliary_loss_clip": 0.01116706,
      "auxiliary_loss_mlp": 0.01026811,
      "balance_loss_clip": 1.04227161,
      "balance_loss_mlp": 1.01934004,
      "epoch": 0.752119280947514,
      "flos": 21725704742400.0,
      "grad_norm": 1.7365892652781167,
      "language_loss": 0.82121229,
      "learning_rate": 6.105089396303258e-07,
      "loss": 0.84264749,
      "num_input_tokens_seen": 134471350,
      "step": 6255,
      "time_per_iteration": 2.7478363513946533
    },
    {
      "auxiliary_loss_clip": 0.01140972,
      "auxiliary_loss_mlp": 0.01025814,
      "balance_loss_clip": 1.04336381,
      "balance_loss_mlp": 1.01799703,
      "epoch": 0.7522395238381531,
      "flos": 32742774668160.0,
      "grad_norm": 1.7950138831081461,
      "language_loss": 0.75520027,
      "learning_rate": 6.099487653009383e-07,
      "loss": 0.7768681,
      "num_input_tokens_seen": 134490695,
      "step": 6256,
      "time_per_iteration": 2.756467580795288
    },
    {
      "auxiliary_loss_clip": 0.01150528,
      "auxiliary_loss_mlp": 0.01027792,
      "balance_loss_clip": 1.04397273,
      "balance_loss_mlp": 1.02157807,
      "epoch": 0.7523597667287921,
      "flos": 23476026579840.0,
      "grad_norm": 1.9650594538959594,
      "language_loss": 0.82834804,
      "learning_rate": 6.093888018412192e-07,
      "loss": 0.85013115,
      "num_input_tokens_seen": 134506885,
      "step": 6257,
      "time_per_iteration": 2.6924142837524414
    },
    {
      "auxiliary_loss_clip": 0.01059174,
      "auxiliary_loss_mlp": 0.01000818,
      "balance_loss_clip": 1.01287508,
      "balance_loss_mlp": 0.99971503,
      "epoch": 0.7524800096194313,
      "flos": 67346730501120.0,
      "grad_norm": 0.7212498088554035,
      "language_loss": 0.54658824,
      "learning_rate": 6.088290493361125e-07,
      "loss": 0.56718814,
      "num_input_tokens_seen": 134571770,
      "step": 6258,
      "time_per_iteration": 3.4147026538848877
    },
    {
      "auxiliary_loss_clip": 0.01102411,
      "auxiliary_loss_mlp": 0.01027819,
      "balance_loss_clip": 1.038854,
      "balance_loss_mlp": 1.02052057,
      "epoch": 0.7526002525100703,
      "flos": 13006055681280.0,
      "grad_norm": 1.92976448761389,
      "language_loss": 0.7166273,
      "learning_rate": 6.082695078705322e-07,
      "loss": 0.73792958,
      "num_input_tokens_seen": 134589250,
      "step": 6259,
      "time_per_iteration": 2.7262258529663086
    },
    {
      "auxiliary_loss_clip": 0.0114762,
      "auxiliary_loss_mlp": 0.01026099,
      "balance_loss_clip": 1.04550135,
      "balance_loss_mlp": 1.01839244,
      "epoch": 0.7527204954007094,
      "flos": 21397229844480.0,
      "grad_norm": 2.3586720533326906,
      "language_loss": 0.68780529,
      "learning_rate": 6.077101775293618e-07,
      "loss": 0.70954251,
      "num_input_tokens_seen": 134608075,
      "step": 6260,
      "time_per_iteration": 2.6716935634613037
    },
    {
      "auxiliary_loss_clip": 0.01152837,
      "auxiliary_loss_mlp": 0.01027175,
      "balance_loss_clip": 1.04348385,
      "balance_loss_mlp": 1.01972485,
      "epoch": 0.7528407382913486,
      "flos": 18947188091520.0,
      "grad_norm": 3.2127736831010156,
      "language_loss": 0.82659423,
      "learning_rate": 6.071510583974504e-07,
      "loss": 0.84839433,
      "num_input_tokens_seen": 134623260,
      "step": 6261,
      "time_per_iteration": 2.6255314350128174
    },
    {
      "auxiliary_loss_clip": 0.01169446,
      "auxiliary_loss_mlp": 0.01025248,
      "balance_loss_clip": 1.04838395,
      "balance_loss_mlp": 1.01792598,
      "epoch": 0.7529609811819876,
      "flos": 15231798956160.0,
      "grad_norm": 2.4914882366134874,
      "language_loss": 0.72004449,
      "learning_rate": 6.065921505596161e-07,
      "loss": 0.7419914,
      "num_input_tokens_seen": 134641540,
      "step": 6262,
      "time_per_iteration": 2.6495933532714844
    },
    {
      "auxiliary_loss_clip": 0.01125055,
      "auxiliary_loss_mlp": 0.01023078,
      "balance_loss_clip": 1.04328561,
      "balance_loss_mlp": 1.0162828,
      "epoch": 0.7530812240726267,
      "flos": 19354487385600.0,
      "grad_norm": 1.619160954196703,
      "language_loss": 0.7707786,
      "learning_rate": 6.060334541006445e-07,
      "loss": 0.79225993,
      "num_input_tokens_seen": 134660035,
      "step": 6263,
      "time_per_iteration": 2.7348968982696533
    },
    {
      "auxiliary_loss_clip": 0.01125662,
      "auxiliary_loss_mlp": 0.01023989,
      "balance_loss_clip": 1.03913116,
      "balance_loss_mlp": 1.01731038,
      "epoch": 0.7532014669632658,
      "flos": 27748247328000.0,
      "grad_norm": 1.68107783548805,
      "language_loss": 0.69220817,
      "learning_rate": 6.05474969105289e-07,
      "loss": 0.71370465,
      "num_input_tokens_seen": 134683025,
      "step": 6264,
      "time_per_iteration": 2.850309133529663
    },
    {
      "auxiliary_loss_clip": 0.01155215,
      "auxiliary_loss_mlp": 0.01035135,
      "balance_loss_clip": 1.04662347,
      "balance_loss_mlp": 1.02681148,
      "epoch": 0.7533217098539049,
      "flos": 14137421333760.0,
      "grad_norm": 2.1339999725996104,
      "language_loss": 0.73980951,
      "learning_rate": 6.049166956582725e-07,
      "loss": 0.76171303,
      "num_input_tokens_seen": 134701290,
      "step": 6265,
      "time_per_iteration": 2.6398746967315674
    },
    {
      "auxiliary_loss_clip": 0.01146748,
      "auxiliary_loss_mlp": 0.01020411,
      "balance_loss_clip": 1.04198337,
      "balance_loss_mlp": 1.01370001,
      "epoch": 0.753441952744544,
      "flos": 26429068437120.0,
      "grad_norm": 2.2045831395176227,
      "language_loss": 0.87417853,
      "learning_rate": 6.043586338442841e-07,
      "loss": 0.89585012,
      "num_input_tokens_seen": 134720345,
      "step": 6266,
      "time_per_iteration": 2.644554853439331
    },
    {
      "auxiliary_loss_clip": 0.01165176,
      "auxiliary_loss_mlp": 0.01023187,
      "balance_loss_clip": 1.04734397,
      "balance_loss_mlp": 1.016783,
      "epoch": 0.7535621956351831,
      "flos": 23878621192320.0,
      "grad_norm": 2.0890091406381948,
      "language_loss": 0.73176897,
      "learning_rate": 6.038007837479815e-07,
      "loss": 0.75365257,
      "num_input_tokens_seen": 134741450,
      "step": 6267,
      "time_per_iteration": 2.680832862854004
    },
    {
      "auxiliary_loss_clip": 0.01150759,
      "auxiliary_loss_mlp": 0.01020997,
      "balance_loss_clip": 1.0453527,
      "balance_loss_mlp": 1.01413631,
      "epoch": 0.7536824385258222,
      "flos": 21795873960960.0,
      "grad_norm": 1.9098141274986997,
      "language_loss": 0.64090174,
      "learning_rate": 6.032431454539897e-07,
      "loss": 0.66261935,
      "num_input_tokens_seen": 134760295,
      "step": 6268,
      "time_per_iteration": 2.623098850250244
    },
    {
      "auxiliary_loss_clip": 0.01125534,
      "auxiliary_loss_mlp": 0.01026599,
      "balance_loss_clip": 1.04108059,
      "balance_loss_mlp": 1.01968229,
      "epoch": 0.7538026814164612,
      "flos": 28911644933760.0,
      "grad_norm": 2.1044252606337093,
      "language_loss": 0.81484723,
      "learning_rate": 6.026857190469014e-07,
      "loss": 0.83636862,
      "num_input_tokens_seen": 134782050,
      "step": 6269,
      "time_per_iteration": 2.7838385105133057
    },
    {
      "auxiliary_loss_clip": 0.01139588,
      "auxiliary_loss_mlp": 0.01022751,
      "balance_loss_clip": 1.04302239,
      "balance_loss_mlp": 1.01606953,
      "epoch": 0.7539229243071004,
      "flos": 21104701482240.0,
      "grad_norm": 1.845781272394144,
      "language_loss": 0.73953533,
      "learning_rate": 6.0212850461128e-07,
      "loss": 0.76115876,
      "num_input_tokens_seen": 134801170,
      "step": 6270,
      "time_per_iteration": 2.6962227821350098
    },
    {
      "auxiliary_loss_clip": 0.0114211,
      "auxiliary_loss_mlp": 0.01024368,
      "balance_loss_clip": 1.04209018,
      "balance_loss_mlp": 1.0170846,
      "epoch": 0.7540431671977395,
      "flos": 15158469340800.0,
      "grad_norm": 1.991840436484,
      "language_loss": 0.74586666,
      "learning_rate": 6.015715022316516e-07,
      "loss": 0.76753151,
      "num_input_tokens_seen": 134819150,
      "step": 6271,
      "time_per_iteration": 3.6456661224365234
    },
    {
      "auxiliary_loss_clip": 0.01113716,
      "auxiliary_loss_mlp": 0.01025175,
      "balance_loss_clip": 1.04074001,
      "balance_loss_mlp": 1.01766753,
      "epoch": 0.7541634100883785,
      "flos": 18770579896320.0,
      "grad_norm": 2.4982370853941607,
      "language_loss": 0.77915907,
      "learning_rate": 6.010147119925154e-07,
      "loss": 0.80054802,
      "num_input_tokens_seen": 134836905,
      "step": 6272,
      "time_per_iteration": 2.77127742767334
    },
    {
      "auxiliary_loss_clip": 0.01116688,
      "auxiliary_loss_mlp": 0.01029193,
      "balance_loss_clip": 1.04189539,
      "balance_loss_mlp": 1.0219779,
      "epoch": 0.7542836529790176,
      "flos": 20594770053120.0,
      "grad_norm": 2.0037858236753183,
      "language_loss": 0.66322893,
      "learning_rate": 6.004581339783348e-07,
      "loss": 0.68468773,
      "num_input_tokens_seen": 134855225,
      "step": 6273,
      "time_per_iteration": 2.779801607131958
    },
    {
      "auxiliary_loss_clip": 0.0115582,
      "auxiliary_loss_mlp": 0.01026099,
      "balance_loss_clip": 1.04462612,
      "balance_loss_mlp": 1.01792741,
      "epoch": 0.7544038958696567,
      "flos": 19095104298240.0,
      "grad_norm": 3.2822716948708957,
      "language_loss": 0.68447018,
      "learning_rate": 5.999017682735425e-07,
      "loss": 0.70628935,
      "num_input_tokens_seen": 134871615,
      "step": 6274,
      "time_per_iteration": 3.6571638584136963
    },
    {
      "auxiliary_loss_clip": 0.01104622,
      "auxiliary_loss_mlp": 0.01026139,
      "balance_loss_clip": 1.03943324,
      "balance_loss_mlp": 1.01911449,
      "epoch": 0.7545241387602958,
      "flos": 31723306859520.0,
      "grad_norm": 2.0742244832419043,
      "language_loss": 0.66205305,
      "learning_rate": 5.993456149625387e-07,
      "loss": 0.6833607,
      "num_input_tokens_seen": 134892765,
      "step": 6275,
      "time_per_iteration": 3.8331286907196045
    },
    {
      "auxiliary_loss_clip": 0.01115563,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.04071331,
      "balance_loss_mlp": 1.01777995,
      "epoch": 0.7546443816509348,
      "flos": 20296495514880.0,
      "grad_norm": 1.8478160775785186,
      "language_loss": 0.82517052,
      "learning_rate": 5.987896741296909e-07,
      "loss": 0.846578,
      "num_input_tokens_seen": 134910505,
      "step": 6276,
      "time_per_iteration": 2.6575846672058105
    },
    {
      "auxiliary_loss_clip": 0.01140702,
      "auxiliary_loss_mlp": 0.01027951,
      "balance_loss_clip": 1.04748929,
      "balance_loss_mlp": 1.02131987,
      "epoch": 0.754764624541574,
      "flos": 23696159080320.0,
      "grad_norm": 2.292307033888658,
      "language_loss": 0.78502434,
      "learning_rate": 5.982339458593361e-07,
      "loss": 0.80671084,
      "num_input_tokens_seen": 134930445,
      "step": 6277,
      "time_per_iteration": 3.625814914703369
    },
    {
      "auxiliary_loss_clip": 0.01149517,
      "auxiliary_loss_mlp": 0.00761355,
      "balance_loss_clip": 1.04548478,
      "balance_loss_mlp": 1.00030673,
      "epoch": 0.7548848674322131,
      "flos": 25337204766720.0,
      "grad_norm": 1.6476814673120361,
      "language_loss": 0.84002179,
      "learning_rate": 5.976784302357767e-07,
      "loss": 0.8591305,
      "num_input_tokens_seen": 134951010,
      "step": 6278,
      "time_per_iteration": 2.6776795387268066
    },
    {
      "auxiliary_loss_clip": 0.01157508,
      "auxiliary_loss_mlp": 0.0102141,
      "balance_loss_clip": 1.04604244,
      "balance_loss_mlp": 1.01454067,
      "epoch": 0.7550051103228521,
      "flos": 19573147428480.0,
      "grad_norm": 2.4891009882633632,
      "language_loss": 0.73635876,
      "learning_rate": 5.971231273432855e-07,
      "loss": 0.75814795,
      "num_input_tokens_seen": 134970495,
      "step": 6279,
      "time_per_iteration": 2.5919971466064453
    },
    {
      "auxiliary_loss_clip": 0.01058302,
      "auxiliary_loss_mlp": 0.01000411,
      "balance_loss_clip": 1.01246738,
      "balance_loss_mlp": 0.99931473,
      "epoch": 0.7551253532134913,
      "flos": 64150068648960.0,
      "grad_norm": 0.8209992599150006,
      "language_loss": 0.54554403,
      "learning_rate": 5.965680372661e-07,
      "loss": 0.56613111,
      "num_input_tokens_seen": 135028060,
      "step": 6280,
      "time_per_iteration": 3.155048131942749
    },
    {
      "auxiliary_loss_clip": 0.01140195,
      "auxiliary_loss_mlp": 0.01022911,
      "balance_loss_clip": 1.0452441,
      "balance_loss_mlp": 1.01650357,
      "epoch": 0.7552455961041303,
      "flos": 26067986968320.0,
      "grad_norm": 1.8259104257775596,
      "language_loss": 0.56418502,
      "learning_rate": 5.960131600884266e-07,
      "loss": 0.58581614,
      "num_input_tokens_seen": 135047330,
      "step": 6281,
      "time_per_iteration": 2.71712327003479
    },
    {
      "auxiliary_loss_clip": 0.01127012,
      "auxiliary_loss_mlp": 0.01019824,
      "balance_loss_clip": 1.04306006,
      "balance_loss_mlp": 1.0128355,
      "epoch": 0.7553658389947694,
      "flos": 24498223822080.0,
      "grad_norm": 1.7241970089663998,
      "language_loss": 0.75961745,
      "learning_rate": 5.954584958944413e-07,
      "loss": 0.78108585,
      "num_input_tokens_seen": 135065995,
      "step": 6282,
      "time_per_iteration": 2.787997245788574
    },
    {
      "auxiliary_loss_clip": 0.01125872,
      "auxiliary_loss_mlp": 0.00761967,
      "balance_loss_clip": 1.04066503,
      "balance_loss_mlp": 1.0003531,
      "epoch": 0.7554860818854086,
      "flos": 21799465320960.0,
      "grad_norm": 2.410344293148737,
      "language_loss": 0.81771076,
      "learning_rate": 5.949040447682854e-07,
      "loss": 0.83658916,
      "num_input_tokens_seen": 135085820,
      "step": 6283,
      "time_per_iteration": 2.814040184020996
    },
    {
      "auxiliary_loss_clip": 0.01144912,
      "auxiliary_loss_mlp": 0.01024653,
      "balance_loss_clip": 1.04377341,
      "balance_loss_mlp": 1.0174017,
      "epoch": 0.7556063247760476,
      "flos": 16362123114240.0,
      "grad_norm": 2.1649497122356043,
      "language_loss": 0.677809,
      "learning_rate": 5.943498067940686e-07,
      "loss": 0.69950467,
      "num_input_tokens_seen": 135102845,
      "step": 6284,
      "time_per_iteration": 2.807706832885742
    },
    {
      "auxiliary_loss_clip": 0.01133512,
      "auxiliary_loss_mlp": 0.01023528,
      "balance_loss_clip": 1.0471431,
      "balance_loss_mlp": 1.01707244,
      "epoch": 0.7557265676666867,
      "flos": 27235155502080.0,
      "grad_norm": 1.8223366815993103,
      "language_loss": 0.81475723,
      "learning_rate": 5.937957820558686e-07,
      "loss": 0.83632755,
      "num_input_tokens_seen": 135122190,
      "step": 6285,
      "time_per_iteration": 2.815964937210083
    },
    {
      "auxiliary_loss_clip": 0.01049897,
      "auxiliary_loss_mlp": 0.01001324,
      "balance_loss_clip": 1.01238608,
      "balance_loss_mlp": 1.00016212,
      "epoch": 0.7558468105573258,
      "flos": 62189131415040.0,
      "grad_norm": 0.8460875014979042,
      "language_loss": 0.65260708,
      "learning_rate": 5.932419706377296e-07,
      "loss": 0.67311931,
      "num_input_tokens_seen": 135180495,
      "step": 6286,
      "time_per_iteration": 3.3413124084472656
    },
    {
      "auxiliary_loss_clip": 0.01121812,
      "auxiliary_loss_mlp": 0.01021534,
      "balance_loss_clip": 1.04454362,
      "balance_loss_mlp": 1.01358259,
      "epoch": 0.7559670534479649,
      "flos": 33249078823680.0,
      "grad_norm": 1.9123176685865677,
      "language_loss": 0.74025619,
      "learning_rate": 5.92688372623666e-07,
      "loss": 0.76168966,
      "num_input_tokens_seen": 135199200,
      "step": 6287,
      "time_per_iteration": 2.8124232292175293
    },
    {
      "auxiliary_loss_clip": 0.01151264,
      "auxiliary_loss_mlp": 0.01021414,
      "balance_loss_clip": 1.04313421,
      "balance_loss_mlp": 1.01372528,
      "epoch": 0.7560872963386039,
      "flos": 14064379027200.0,
      "grad_norm": 2.013960057381503,
      "language_loss": 0.74263859,
      "learning_rate": 5.921349880976574e-07,
      "loss": 0.76436538,
      "num_input_tokens_seen": 135217035,
      "step": 6288,
      "time_per_iteration": 2.920696496963501
    },
    {
      "auxiliary_loss_clip": 0.01141504,
      "auxiliary_loss_mlp": 0.00762606,
      "balance_loss_clip": 1.04253244,
      "balance_loss_mlp": 1.00036418,
      "epoch": 0.7562075392292431,
      "flos": 20412307941120.0,
      "grad_norm": 1.9475340084046477,
      "language_loss": 0.81498098,
      "learning_rate": 5.915818171436515e-07,
      "loss": 0.8340221,
      "num_input_tokens_seen": 135236370,
      "step": 6289,
      "time_per_iteration": 2.689699649810791
    },
    {
      "auxiliary_loss_clip": 0.01138041,
      "auxiliary_loss_mlp": 0.01025891,
      "balance_loss_clip": 1.04131842,
      "balance_loss_mlp": 1.01816893,
      "epoch": 0.7563277821198822,
      "flos": 20376792368640.0,
      "grad_norm": 1.9624529616483757,
      "language_loss": 0.74726409,
      "learning_rate": 5.910288598455642e-07,
      "loss": 0.76890337,
      "num_input_tokens_seen": 135255720,
      "step": 6290,
      "time_per_iteration": 2.732272148132324
    },
    {
      "auxiliary_loss_clip": 0.01159495,
      "auxiliary_loss_mlp": 0.01026252,
      "balance_loss_clip": 1.04491186,
      "balance_loss_mlp": 1.01832485,
      "epoch": 0.7564480250105212,
      "flos": 18588261438720.0,
      "grad_norm": 2.1740456671534782,
      "language_loss": 0.74038965,
      "learning_rate": 5.90476116287278e-07,
      "loss": 0.76224715,
      "num_input_tokens_seen": 135273320,
      "step": 6291,
      "time_per_iteration": 2.669858932495117
    },
    {
      "auxiliary_loss_clip": 0.01141575,
      "auxiliary_loss_mlp": 0.01026765,
      "balance_loss_clip": 1.04640484,
      "balance_loss_mlp": 1.01952279,
      "epoch": 0.7565682679011604,
      "flos": 21215521918080.0,
      "grad_norm": 1.98625073258502,
      "language_loss": 0.67494714,
      "learning_rate": 5.899235865526456e-07,
      "loss": 0.69663048,
      "num_input_tokens_seen": 135292615,
      "step": 6292,
      "time_per_iteration": 2.654615879058838
    },
    {
      "auxiliary_loss_clip": 0.01117018,
      "auxiliary_loss_mlp": 0.01023383,
      "balance_loss_clip": 1.0410049,
      "balance_loss_mlp": 1.01685333,
      "epoch": 0.7566885107917994,
      "flos": 20449008662400.0,
      "grad_norm": 1.673209367793022,
      "language_loss": 0.82049394,
      "learning_rate": 5.893712707254825e-07,
      "loss": 0.84189796,
      "num_input_tokens_seen": 135310075,
      "step": 6293,
      "time_per_iteration": 2.7207071781158447
    },
    {
      "auxiliary_loss_clip": 0.01105763,
      "auxiliary_loss_mlp": 0.01022079,
      "balance_loss_clip": 1.0379169,
      "balance_loss_mlp": 1.01462269,
      "epoch": 0.7568087536824385,
      "flos": 19025832919680.0,
      "grad_norm": 2.6471451883831163,
      "language_loss": 0.66429102,
      "learning_rate": 5.888191688895769e-07,
      "loss": 0.68556941,
      "num_input_tokens_seen": 135327335,
      "step": 6294,
      "time_per_iteration": 2.8102660179138184
    },
    {
      "auxiliary_loss_clip": 0.01166598,
      "auxiliary_loss_mlp": 0.01023965,
      "balance_loss_clip": 1.04454422,
      "balance_loss_mlp": 1.01659775,
      "epoch": 0.7569289965730777,
      "flos": 15225442248960.0,
      "grad_norm": 2.4121090557416864,
      "language_loss": 0.62389433,
      "learning_rate": 5.882672811286813e-07,
      "loss": 0.64579993,
      "num_input_tokens_seen": 135343615,
      "step": 6295,
      "time_per_iteration": 2.718003988265991
    },
    {
      "auxiliary_loss_clip": 0.0116788,
      "auxiliary_loss_mlp": 0.01028573,
      "balance_loss_clip": 1.04474258,
      "balance_loss_mlp": 1.0209527,
      "epoch": 0.7570492394637167,
      "flos": 20769367086720.0,
      "grad_norm": 2.084705568977647,
      "language_loss": 0.6964274,
      "learning_rate": 5.877156075265166e-07,
      "loss": 0.71839195,
      "num_input_tokens_seen": 135359880,
      "step": 6296,
      "time_per_iteration": 2.7622368335723877
    },
    {
      "auxiliary_loss_clip": 0.01136152,
      "auxiliary_loss_mlp": 0.01028691,
      "balance_loss_clip": 1.04038179,
      "balance_loss_mlp": 1.02154446,
      "epoch": 0.7571694823543558,
      "flos": 15664091137920.0,
      "grad_norm": 3.2434354007642128,
      "language_loss": 0.70006472,
      "learning_rate": 5.871641481667715e-07,
      "loss": 0.72171319,
      "num_input_tokens_seen": 135374325,
      "step": 6297,
      "time_per_iteration": 3.6909573078155518
    },
    {
      "auxiliary_loss_clip": 0.0111357,
      "auxiliary_loss_mlp": 0.01029748,
      "balance_loss_clip": 1.04157495,
      "balance_loss_mlp": 1.02236915,
      "epoch": 0.7572897252449949,
      "flos": 25409241492480.0,
      "grad_norm": 1.9724741967299242,
      "language_loss": 0.84393597,
      "learning_rate": 5.866129031331011e-07,
      "loss": 0.8653692,
      "num_input_tokens_seen": 135393980,
      "step": 6298,
      "time_per_iteration": 2.8981802463531494
    },
    {
      "auxiliary_loss_clip": 0.01139921,
      "auxiliary_loss_mlp": 0.01028847,
      "balance_loss_clip": 1.0420382,
      "balance_loss_mlp": 1.0217452,
      "epoch": 0.757409968135634,
      "flos": 24279348297600.0,
      "grad_norm": 1.958729723651007,
      "language_loss": 0.83320141,
      "learning_rate": 5.8606187250913e-07,
      "loss": 0.85488909,
      "num_input_tokens_seen": 135412030,
      "step": 6299,
      "time_per_iteration": 2.7568917274475098
    },
    {
      "auxiliary_loss_clip": 0.01152813,
      "auxiliary_loss_mlp": 0.00761785,
      "balance_loss_clip": 1.04767704,
      "balance_loss_mlp": 1.00040877,
      "epoch": 0.757530211026273,
      "flos": 24133766474880.0,
      "grad_norm": 1.9047325448432209,
      "language_loss": 0.84423959,
      "learning_rate": 5.855110563784482e-07,
      "loss": 0.86338556,
      "num_input_tokens_seen": 135430565,
      "step": 6300,
      "time_per_iteration": 3.72444748878479
    },
    {
      "auxiliary_loss_clip": 0.01146225,
      "auxiliary_loss_mlp": 0.00762271,
      "balance_loss_clip": 1.04313529,
      "balance_loss_mlp": 1.0003593,
      "epoch": 0.7576504539169122,
      "flos": 23951807153280.0,
      "grad_norm": 1.6676511778433556,
      "language_loss": 0.64084601,
      "learning_rate": 5.849604548246156e-07,
      "loss": 0.65993094,
      "num_input_tokens_seen": 135451675,
      "step": 6301,
      "time_per_iteration": 3.6140666007995605
    },
    {
      "auxiliary_loss_clip": 0.01146713,
      "auxiliary_loss_mlp": 0.00762015,
      "balance_loss_clip": 1.04754448,
      "balance_loss_mlp": 1.00039482,
      "epoch": 0.7577706968075513,
      "flos": 21251360712960.0,
      "grad_norm": 4.049176499081369,
      "language_loss": 0.80364913,
      "learning_rate": 5.844100679311565e-07,
      "loss": 0.82273638,
      "num_input_tokens_seen": 135470635,
      "step": 6302,
      "time_per_iteration": 2.712498188018799
    },
    {
      "auxiliary_loss_clip": 0.01139141,
      "auxiliary_loss_mlp": 0.01025571,
      "balance_loss_clip": 1.0440588,
      "balance_loss_mlp": 1.01811481,
      "epoch": 0.7578909396981903,
      "flos": 18296595002880.0,
      "grad_norm": 2.613512436020556,
      "language_loss": 0.76440573,
      "learning_rate": 5.838598957815637e-07,
      "loss": 0.78605282,
      "num_input_tokens_seen": 135487865,
      "step": 6303,
      "time_per_iteration": 3.628082752227783
    },
    {
      "auxiliary_loss_clip": 0.01133443,
      "auxiliary_loss_mlp": 0.01029205,
      "balance_loss_clip": 1.04291451,
      "balance_loss_mlp": 1.02134883,
      "epoch": 0.7580111825888295,
      "flos": 25373869574400.0,
      "grad_norm": 2.109339797373425,
      "language_loss": 0.853652,
      "learning_rate": 5.833099384592996e-07,
      "loss": 0.87527847,
      "num_input_tokens_seen": 135508440,
      "step": 6304,
      "time_per_iteration": 2.781425952911377
    },
    {
      "auxiliary_loss_clip": 0.01136721,
      "auxiliary_loss_mlp": 0.01021591,
      "balance_loss_clip": 1.04525411,
      "balance_loss_mlp": 1.01417625,
      "epoch": 0.7581314254794685,
      "flos": 23768662682880.0,
      "grad_norm": 2.103217233452294,
      "language_loss": 0.71307147,
      "learning_rate": 5.827601960477913e-07,
      "loss": 0.73465455,
      "num_input_tokens_seen": 135526365,
      "step": 6305,
      "time_per_iteration": 2.834982395172119
    },
    {
      "auxiliary_loss_clip": 0.01148564,
      "auxiliary_loss_mlp": 0.01026133,
      "balance_loss_clip": 1.04179239,
      "balance_loss_mlp": 1.01910853,
      "epoch": 0.7582516683701076,
      "flos": 22054610603520.0,
      "grad_norm": 1.9277816443594353,
      "language_loss": 0.70513153,
      "learning_rate": 5.822106686304344e-07,
      "loss": 0.72687846,
      "num_input_tokens_seen": 135545655,
      "step": 6306,
      "time_per_iteration": 2.737581253051758
    },
    {
      "auxiliary_loss_clip": 0.01129601,
      "auxiliary_loss_mlp": 0.01026962,
      "balance_loss_clip": 1.04175305,
      "balance_loss_mlp": 1.01944017,
      "epoch": 0.7583719112607467,
      "flos": 31649725848960.0,
      "grad_norm": 1.9914277316950908,
      "language_loss": 0.57841611,
      "learning_rate": 5.816613562905919e-07,
      "loss": 0.59998178,
      "num_input_tokens_seen": 135566840,
      "step": 6307,
      "time_per_iteration": 2.7760770320892334
    },
    {
      "auxiliary_loss_clip": 0.01118079,
      "auxiliary_loss_mlp": 0.01021959,
      "balance_loss_clip": 1.04292512,
      "balance_loss_mlp": 1.0152688,
      "epoch": 0.7584921541513858,
      "flos": 33068376478080.0,
      "grad_norm": 1.8379004856668884,
      "language_loss": 0.70063239,
      "learning_rate": 5.811122591115933e-07,
      "loss": 0.72203279,
      "num_input_tokens_seen": 135587825,
      "step": 6308,
      "time_per_iteration": 2.811708927154541
    },
    {
      "auxiliary_loss_clip": 0.01122638,
      "auxiliary_loss_mlp": 0.01028988,
      "balance_loss_clip": 1.04425991,
      "balance_loss_mlp": 1.02158487,
      "epoch": 0.7586123970420249,
      "flos": 23326350606720.0,
      "grad_norm": 2.6134725234031464,
      "language_loss": 0.71178102,
      "learning_rate": 5.805633771767376e-07,
      "loss": 0.73329729,
      "num_input_tokens_seen": 135605220,
      "step": 6309,
      "time_per_iteration": 2.726496934890747
    },
    {
      "auxiliary_loss_clip": 0.01138096,
      "auxiliary_loss_mlp": 0.01028745,
      "balance_loss_clip": 1.04478073,
      "balance_loss_mlp": 1.02207184,
      "epoch": 0.7587326399326639,
      "flos": 18334229477760.0,
      "grad_norm": 1.7321532030601894,
      "language_loss": 0.77890545,
      "learning_rate": 5.800147105692888e-07,
      "loss": 0.80057389,
      "num_input_tokens_seen": 135624795,
      "step": 6310,
      "time_per_iteration": 2.730590343475342
    },
    {
      "auxiliary_loss_clip": 0.01152011,
      "auxiliary_loss_mlp": 0.01021953,
      "balance_loss_clip": 1.04198146,
      "balance_loss_mlp": 1.01490748,
      "epoch": 0.7588528828233031,
      "flos": 17275080119040.0,
      "grad_norm": 1.7672889420574105,
      "language_loss": 0.78823024,
      "learning_rate": 5.794662593724795e-07,
      "loss": 0.8099699,
      "num_input_tokens_seen": 135643800,
      "step": 6311,
      "time_per_iteration": 2.7096822261810303
    },
    {
      "auxiliary_loss_clip": 0.01171258,
      "auxiliary_loss_mlp": 0.01025534,
      "balance_loss_clip": 1.0502758,
      "balance_loss_mlp": 1.01806855,
      "epoch": 0.7589731257139422,
      "flos": 17713621267200.0,
      "grad_norm": 1.8929846373044263,
      "language_loss": 0.75111616,
      "learning_rate": 5.789180236695091e-07,
      "loss": 0.77308416,
      "num_input_tokens_seen": 135660655,
      "step": 6312,
      "time_per_iteration": 2.646646738052368
    },
    {
      "auxiliary_loss_clip": 0.01147637,
      "auxiliary_loss_mlp": 0.0102353,
      "balance_loss_clip": 1.04461527,
      "balance_loss_mlp": 1.01679218,
      "epoch": 0.7590933686045812,
      "flos": 15961072786560.0,
      "grad_norm": 2.3258460225069677,
      "language_loss": 0.85047269,
      "learning_rate": 5.78370003543544e-07,
      "loss": 0.8721844,
      "num_input_tokens_seen": 135679410,
      "step": 6313,
      "time_per_iteration": 2.649960517883301
    },
    {
      "auxiliary_loss_clip": 0.01153324,
      "auxiliary_loss_mlp": 0.00762347,
      "balance_loss_clip": 1.04476666,
      "balance_loss_mlp": 1.00034904,
      "epoch": 0.7592136114952204,
      "flos": 21068072588160.0,
      "grad_norm": 1.9855329163534234,
      "language_loss": 0.83486533,
      "learning_rate": 5.778221990777203e-07,
      "loss": 0.85402203,
      "num_input_tokens_seen": 135697150,
      "step": 6314,
      "time_per_iteration": 2.594696521759033
    },
    {
      "auxiliary_loss_clip": 0.01141951,
      "auxiliary_loss_mlp": 0.01025396,
      "balance_loss_clip": 1.04573607,
      "balance_loss_mlp": 1.01848507,
      "epoch": 0.7593338543858594,
      "flos": 25297666871040.0,
      "grad_norm": 2.63222025993635,
      "language_loss": 0.82578337,
      "learning_rate": 5.772746103551372e-07,
      "loss": 0.84745681,
      "num_input_tokens_seen": 135712545,
      "step": 6315,
      "time_per_iteration": 2.6482956409454346
    },
    {
      "auxiliary_loss_clip": 0.01135959,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.04506969,
      "balance_loss_mlp": 1.01991963,
      "epoch": 0.7594540972764985,
      "flos": 31832367528960.0,
      "grad_norm": 1.934637948013311,
      "language_loss": 0.71399981,
      "learning_rate": 5.767272374588648e-07,
      "loss": 0.7356267,
      "num_input_tokens_seen": 135733950,
      "step": 6316,
      "time_per_iteration": 2.7368102073669434
    },
    {
      "auxiliary_loss_clip": 0.01151554,
      "auxiliary_loss_mlp": 0.01028886,
      "balance_loss_clip": 1.04541874,
      "balance_loss_mlp": 1.02230525,
      "epoch": 0.7595743401671377,
      "flos": 37597250880000.0,
      "grad_norm": 1.701811029863731,
      "language_loss": 0.77918267,
      "learning_rate": 5.76180080471939e-07,
      "loss": 0.80098701,
      "num_input_tokens_seen": 135757120,
      "step": 6317,
      "time_per_iteration": 2.830780029296875
    },
    {
      "auxiliary_loss_clip": 0.01170569,
      "auxiliary_loss_mlp": 0.010253,
      "balance_loss_clip": 1.04730892,
      "balance_loss_mlp": 1.0175221,
      "epoch": 0.7596945830577767,
      "flos": 18287724343680.0,
      "grad_norm": 2.1494147126794445,
      "language_loss": 0.72327065,
      "learning_rate": 5.756331394773631e-07,
      "loss": 0.74522936,
      "num_input_tokens_seen": 135773335,
      "step": 6318,
      "time_per_iteration": 2.5440547466278076
    },
    {
      "auxiliary_loss_clip": 0.010949,
      "auxiliary_loss_mlp": 0.00762553,
      "balance_loss_clip": 1.03854334,
      "balance_loss_mlp": 1.00035882,
      "epoch": 0.7598148259484158,
      "flos": 22233122219520.0,
      "grad_norm": 2.0509616240386,
      "language_loss": 0.76128769,
      "learning_rate": 5.750864145581071e-07,
      "loss": 0.77986217,
      "num_input_tokens_seen": 135792555,
      "step": 6319,
      "time_per_iteration": 2.7838962078094482
    },
    {
      "auxiliary_loss_clip": 0.01167056,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.04672647,
      "balance_loss_mlp": 1.01815319,
      "epoch": 0.7599350688390549,
      "flos": 27161718145920.0,
      "grad_norm": 1.8486041513525464,
      "language_loss": 0.86402762,
      "learning_rate": 5.745399057971085e-07,
      "loss": 0.88594592,
      "num_input_tokens_seen": 135813690,
      "step": 6320,
      "time_per_iteration": 2.6997833251953125
    },
    {
      "auxiliary_loss_clip": 0.01158223,
      "auxiliary_loss_mlp": 0.01025118,
      "balance_loss_clip": 1.04598546,
      "balance_loss_mlp": 1.01755428,
      "epoch": 0.760055311729694,
      "flos": 15560704817280.0,
      "grad_norm": 2.3002389889355617,
      "language_loss": 0.75059491,
      "learning_rate": 5.739936132772738e-07,
      "loss": 0.77242839,
      "num_input_tokens_seen": 135832255,
      "step": 6321,
      "time_per_iteration": 2.6230013370513916
    },
    {
      "auxiliary_loss_clip": 0.01165725,
      "auxiliary_loss_mlp": 0.01025354,
      "balance_loss_clip": 1.04598498,
      "balance_loss_mlp": 1.01857364,
      "epoch": 0.760175554620333,
      "flos": 25155496840320.0,
      "grad_norm": 2.207653447871622,
      "language_loss": 0.74417233,
      "learning_rate": 5.734475370814733e-07,
      "loss": 0.76608312,
      "num_input_tokens_seen": 135851935,
      "step": 6322,
      "time_per_iteration": 3.543412446975708
    },
    {
      "auxiliary_loss_clip": 0.01156963,
      "auxiliary_loss_mlp": 0.01023815,
      "balance_loss_clip": 1.04455805,
      "balance_loss_mlp": 1.01653087,
      "epoch": 0.7602957975109722,
      "flos": 24353791234560.0,
      "grad_norm": 1.577349970172988,
      "language_loss": 0.78493333,
      "learning_rate": 5.729016772925483e-07,
      "loss": 0.80674112,
      "num_input_tokens_seen": 135873510,
      "step": 6323,
      "time_per_iteration": 2.6469149589538574
    },
    {
      "auxiliary_loss_clip": 0.01105845,
      "auxiliary_loss_mlp": 0.01021996,
      "balance_loss_clip": 1.04204798,
      "balance_loss_mlp": 1.01518941,
      "epoch": 0.7604160404016113,
      "flos": 25192664438400.0,
      "grad_norm": 1.8601796321705393,
      "language_loss": 0.7075597,
      "learning_rate": 5.723560339933038e-07,
      "loss": 0.72883815,
      "num_input_tokens_seen": 135893845,
      "step": 6324,
      "time_per_iteration": 2.8287336826324463
    },
    {
      "auxiliary_loss_clip": 0.01150857,
      "auxiliary_loss_mlp": 0.00761968,
      "balance_loss_clip": 1.04407251,
      "balance_loss_mlp": 1.00038838,
      "epoch": 0.7605362832922503,
      "flos": 29861841363840.0,
      "grad_norm": 1.9548997033109565,
      "language_loss": 0.65133667,
      "learning_rate": 5.71810607266513e-07,
      "loss": 0.67046487,
      "num_input_tokens_seen": 135912430,
      "step": 6325,
      "time_per_iteration": 2.7100303173065186
    },
    {
      "auxiliary_loss_clip": 0.01151256,
      "auxiliary_loss_mlp": 0.01033697,
      "balance_loss_clip": 1.04244256,
      "balance_loss_mlp": 1.02669084,
      "epoch": 0.7606565261828895,
      "flos": 13917935278080.0,
      "grad_norm": 1.8599563653281472,
      "language_loss": 0.60572016,
      "learning_rate": 5.712653971949184e-07,
      "loss": 0.62756968,
      "num_input_tokens_seen": 135930550,
      "step": 6326,
      "time_per_iteration": 3.9500253200531006
    },
    {
      "auxiliary_loss_clip": 0.01146184,
      "auxiliary_loss_mlp": 0.01024847,
      "balance_loss_clip": 1.04258132,
      "balance_loss_mlp": 1.01780438,
      "epoch": 0.7607767690735285,
      "flos": 18551273408640.0,
      "grad_norm": 2.5646036818967715,
      "language_loss": 0.75374019,
      "learning_rate": 5.707204038612268e-07,
      "loss": 0.77545047,
      "num_input_tokens_seen": 135947980,
      "step": 6327,
      "time_per_iteration": 3.6326229572296143
    },
    {
      "auxiliary_loss_clip": 0.01147881,
      "auxiliary_loss_mlp": 0.01030697,
      "balance_loss_clip": 1.0504446,
      "balance_loss_mlp": 1.02288282,
      "epoch": 0.7608970119641676,
      "flos": 20922993555840.0,
      "grad_norm": 2.288692000046473,
      "language_loss": 0.74147248,
      "learning_rate": 5.701756273481138e-07,
      "loss": 0.76325828,
      "num_input_tokens_seen": 135965400,
      "step": 6328,
      "time_per_iteration": 2.7432796955108643
    },
    {
      "auxiliary_loss_clip": 0.01143011,
      "auxiliary_loss_mlp": 0.01026098,
      "balance_loss_clip": 1.04347086,
      "balance_loss_mlp": 1.01945186,
      "epoch": 0.7610172548548068,
      "flos": 23807302738560.0,
      "grad_norm": 1.506493781097288,
      "language_loss": 0.74011517,
      "learning_rate": 5.696310677382212e-07,
      "loss": 0.76180625,
      "num_input_tokens_seen": 135986795,
      "step": 6329,
      "time_per_iteration": 2.7720091342926025
    },
    {
      "auxiliary_loss_clip": 0.01039133,
      "auxiliary_loss_mlp": 0.01000719,
      "balance_loss_clip": 1.0167532,
      "balance_loss_mlp": 0.99961627,
      "epoch": 0.7611374977454458,
      "flos": 66496580426880.0,
      "grad_norm": 0.8785653125016024,
      "language_loss": 0.6181308,
      "learning_rate": 5.690867251141576e-07,
      "loss": 0.6385293,
      "num_input_tokens_seen": 136053450,
      "step": 6330,
      "time_per_iteration": 4.412827014923096
    },
    {
      "auxiliary_loss_clip": 0.01159875,
      "auxiliary_loss_mlp": 0.01021172,
      "balance_loss_clip": 1.04485536,
      "balance_loss_mlp": 1.01412678,
      "epoch": 0.7612577406360849,
      "flos": 15633136592640.0,
      "grad_norm": 2.5654260077837896,
      "language_loss": 0.91710877,
      "learning_rate": 5.685425995585013e-07,
      "loss": 0.93891919,
      "num_input_tokens_seen": 136071375,
      "step": 6331,
      "time_per_iteration": 2.6193904876708984
    },
    {
      "auxiliary_loss_clip": 0.01050158,
      "auxiliary_loss_mlp": 0.01001462,
      "balance_loss_clip": 1.01288557,
      "balance_loss_mlp": 1.0003531,
      "epoch": 0.761377983526724,
      "flos": 60526253237760.0,
      "grad_norm": 0.7488647414752886,
      "language_loss": 0.59012389,
      "learning_rate": 5.679986911537935e-07,
      "loss": 0.61064005,
      "num_input_tokens_seen": 136138905,
      "step": 6332,
      "time_per_iteration": 3.3278043270111084
    },
    {
      "auxiliary_loss_clip": 0.01098419,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 1.04054308,
      "balance_loss_mlp": 1.01779056,
      "epoch": 0.7614982264173631,
      "flos": 35772522019200.0,
      "grad_norm": 1.8952726120378358,
      "language_loss": 0.66765219,
      "learning_rate": 5.674549999825462e-07,
      "loss": 0.68888259,
      "num_input_tokens_seen": 136161720,
      "step": 6333,
      "time_per_iteration": 2.86212420463562
    },
    {
      "auxiliary_loss_clip": 0.01058914,
      "auxiliary_loss_mlp": 0.0100197,
      "balance_loss_clip": 1.01274467,
      "balance_loss_mlp": 1.00092077,
      "epoch": 0.7616184693080021,
      "flos": 67925502345600.0,
      "grad_norm": 0.9139416715771943,
      "language_loss": 0.71372342,
      "learning_rate": 5.669115261272363e-07,
      "loss": 0.7343322,
      "num_input_tokens_seen": 136222040,
      "step": 6334,
      "time_per_iteration": 3.1977081298828125
    },
    {
      "auxiliary_loss_clip": 0.01156395,
      "auxiliary_loss_mlp": 0.01024446,
      "balance_loss_clip": 1.04763567,
      "balance_loss_mlp": 1.01739812,
      "epoch": 0.7617387121986413,
      "flos": 20521979141760.0,
      "grad_norm": 3.538803022576379,
      "language_loss": 0.72793818,
      "learning_rate": 5.663682696703081e-07,
      "loss": 0.74974668,
      "num_input_tokens_seen": 136240305,
      "step": 6335,
      "time_per_iteration": 2.6603240966796875
    },
    {
      "auxiliary_loss_clip": 0.01164982,
      "auxiliary_loss_mlp": 0.0102478,
      "balance_loss_clip": 1.04656601,
      "balance_loss_mlp": 1.01854563,
      "epoch": 0.7618589550892804,
      "flos": 18624495283200.0,
      "grad_norm": 1.8975780685722188,
      "language_loss": 0.81633151,
      "learning_rate": 5.658252306941746e-07,
      "loss": 0.83822912,
      "num_input_tokens_seen": 136259625,
      "step": 6336,
      "time_per_iteration": 2.5953853130340576
    },
    {
      "auxiliary_loss_clip": 0.01112332,
      "auxiliary_loss_mlp": 0.0103196,
      "balance_loss_clip": 1.04148638,
      "balance_loss_mlp": 1.0238899,
      "epoch": 0.7619791979799194,
      "flos": 17453735389440.0,
      "grad_norm": 2.0464550096887226,
      "language_loss": 0.75247151,
      "learning_rate": 5.65282409281212e-07,
      "loss": 0.77391446,
      "num_input_tokens_seen": 136277090,
      "step": 6337,
      "time_per_iteration": 2.761497974395752
    },
    {
      "auxiliary_loss_clip": 0.0113674,
      "auxiliary_loss_mlp": 0.01030616,
      "balance_loss_clip": 1.04385972,
      "balance_loss_mlp": 1.02317739,
      "epoch": 0.7620994408705585,
      "flos": 14137421333760.0,
      "grad_norm": 2.1791527272542366,
      "language_loss": 0.70025623,
      "learning_rate": 5.64739805513768e-07,
      "loss": 0.72192979,
      "num_input_tokens_seen": 136294635,
      "step": 6338,
      "time_per_iteration": 2.8429782390594482
    },
    {
      "auxiliary_loss_clip": 0.01052953,
      "auxiliary_loss_mlp": 0.0075377,
      "balance_loss_clip": 1.01101446,
      "balance_loss_mlp": 1.00028217,
      "epoch": 0.7622196837611976,
      "flos": 70708792527360.0,
      "grad_norm": 0.7873552627367454,
      "language_loss": 0.55733567,
      "learning_rate": 5.641974194741541e-07,
      "loss": 0.57540292,
      "num_input_tokens_seen": 136350320,
      "step": 6339,
      "time_per_iteration": 3.2008166313171387
    },
    {
      "auxiliary_loss_clip": 0.01043834,
      "auxiliary_loss_mlp": 0.01002054,
      "balance_loss_clip": 1.02042353,
      "balance_loss_mlp": 1.00082636,
      "epoch": 0.7623399266518367,
      "flos": 60684150447360.0,
      "grad_norm": 0.7793429766241025,
      "language_loss": 0.63742387,
      "learning_rate": 5.636552512446502e-07,
      "loss": 0.65788275,
      "num_input_tokens_seen": 136411375,
      "step": 6340,
      "time_per_iteration": 3.185504674911499
    },
    {
      "auxiliary_loss_clip": 0.01148551,
      "auxiliary_loss_mlp": 0.01024083,
      "balance_loss_clip": 1.04396617,
      "balance_loss_mlp": 1.0171361,
      "epoch": 0.7624601695424758,
      "flos": 26468893641600.0,
      "grad_norm": 1.74669787990895,
      "language_loss": 0.77981937,
      "learning_rate": 5.631133009075027e-07,
      "loss": 0.80154568,
      "num_input_tokens_seen": 136430560,
      "step": 6341,
      "time_per_iteration": 2.67034649848938
    },
    {
      "auxiliary_loss_clip": 0.0115496,
      "auxiliary_loss_mlp": 0.00761511,
      "balance_loss_clip": 1.04593194,
      "balance_loss_mlp": 1.00027597,
      "epoch": 0.7625804124331149,
      "flos": 19135755515520.0,
      "grad_norm": 2.50728885068984,
      "language_loss": 0.68546939,
      "learning_rate": 5.625715685449242e-07,
      "loss": 0.70463401,
      "num_input_tokens_seen": 136448665,
      "step": 6342,
      "time_per_iteration": 2.6798665523529053
    },
    {
      "auxiliary_loss_clip": 0.01121216,
      "auxiliary_loss_mlp": 0.01025769,
      "balance_loss_clip": 1.04516602,
      "balance_loss_mlp": 1.0189501,
      "epoch": 0.762700655323754,
      "flos": 26213101914240.0,
      "grad_norm": 1.5596743939029112,
      "language_loss": 0.71842062,
      "learning_rate": 5.620300542390966e-07,
      "loss": 0.73989046,
      "num_input_tokens_seen": 136469710,
      "step": 6343,
      "time_per_iteration": 2.7036850452423096
    },
    {
      "auxiliary_loss_clip": 0.01133492,
      "auxiliary_loss_mlp": 0.01022673,
      "balance_loss_clip": 1.0401938,
      "balance_loss_mlp": 1.01611972,
      "epoch": 0.762820898214393,
      "flos": 22382582711040.0,
      "grad_norm": 1.902244675232021,
      "language_loss": 0.8502233,
      "learning_rate": 5.614887580721659e-07,
      "loss": 0.87178493,
      "num_input_tokens_seen": 136489855,
      "step": 6344,
      "time_per_iteration": 2.682849168777466
    },
    {
      "auxiliary_loss_clip": 0.01119389,
      "auxiliary_loss_mlp": 0.01024952,
      "balance_loss_clip": 1.04425931,
      "balance_loss_mlp": 1.0180912,
      "epoch": 0.7629411411050322,
      "flos": 15700504550400.0,
      "grad_norm": 2.0102255678684124,
      "language_loss": 0.73986721,
      "learning_rate": 5.609476801262481e-07,
      "loss": 0.76131058,
      "num_input_tokens_seen": 136504715,
      "step": 6345,
      "time_per_iteration": 2.6353251934051514
    },
    {
      "auxiliary_loss_clip": 0.01122146,
      "auxiliary_loss_mlp": 0.01027645,
      "balance_loss_clip": 1.04372835,
      "balance_loss_mlp": 1.02015829,
      "epoch": 0.7630613839956712,
      "flos": 13770342293760.0,
      "grad_norm": 2.311413130055111,
      "language_loss": 0.64524722,
      "learning_rate": 5.604068204834223e-07,
      "loss": 0.66674513,
      "num_input_tokens_seen": 136521610,
      "step": 6346,
      "time_per_iteration": 2.73693585395813
    },
    {
      "auxiliary_loss_clip": 0.01105011,
      "auxiliary_loss_mlp": 0.00762409,
      "balance_loss_clip": 1.03952456,
      "balance_loss_mlp": 1.00034845,
      "epoch": 0.7631816268863103,
      "flos": 14569569861120.0,
      "grad_norm": 1.9812480603412777,
      "language_loss": 0.76561207,
      "learning_rate": 5.598661792257367e-07,
      "loss": 0.78428626,
      "num_input_tokens_seen": 136538655,
      "step": 6347,
      "time_per_iteration": 2.685518741607666
    },
    {
      "auxiliary_loss_clip": 0.01150189,
      "auxiliary_loss_mlp": 0.01024747,
      "balance_loss_clip": 1.04259109,
      "balance_loss_mlp": 1.01796365,
      "epoch": 0.7633018697769495,
      "flos": 19062210418560.0,
      "grad_norm": 2.5827807757956402,
      "language_loss": 0.76025748,
      "learning_rate": 5.593257564352071e-07,
      "loss": 0.78200686,
      "num_input_tokens_seen": 136557095,
      "step": 6348,
      "time_per_iteration": 3.5536322593688965
    },
    {
      "auxiliary_loss_clip": 0.01148201,
      "auxiliary_loss_mlp": 0.01022563,
      "balance_loss_clip": 1.04392862,
      "balance_loss_mlp": 1.01606655,
      "epoch": 0.7634221126675885,
      "flos": 22052958577920.0,
      "grad_norm": 1.5142091001149238,
      "language_loss": 0.75567925,
      "learning_rate": 5.58785552193815e-07,
      "loss": 0.7773869,
      "num_input_tokens_seen": 136577340,
      "step": 6349,
      "time_per_iteration": 2.6998536586761475
    },
    {
      "auxiliary_loss_clip": 0.01166919,
      "auxiliary_loss_mlp": 0.01021793,
      "balance_loss_clip": 1.04712105,
      "balance_loss_mlp": 1.01462841,
      "epoch": 0.7635423555582276,
      "flos": 29382720825600.0,
      "grad_norm": 1.82661615421463,
      "language_loss": 0.75546229,
      "learning_rate": 5.582455665835086e-07,
      "loss": 0.77734947,
      "num_input_tokens_seen": 136597635,
      "step": 6350,
      "time_per_iteration": 2.6439085006713867
    },
    {
      "auxiliary_loss_clip": 0.01150047,
      "auxiliary_loss_mlp": 0.01027394,
      "balance_loss_clip": 1.04400098,
      "balance_loss_mlp": 1.01936531,
      "epoch": 0.7636625984488667,
      "flos": 17784903807360.0,
      "grad_norm": 2.6605995533690088,
      "language_loss": 0.7228651,
      "learning_rate": 5.577057996862036e-07,
      "loss": 0.74463952,
      "num_input_tokens_seen": 136615260,
      "step": 6351,
      "time_per_iteration": 2.6813716888427734
    },
    {
      "auxiliary_loss_clip": 0.011642,
      "auxiliary_loss_mlp": 0.01022926,
      "balance_loss_clip": 1.04624605,
      "balance_loss_mlp": 1.01634896,
      "epoch": 0.7637828413395058,
      "flos": 23734583654400.0,
      "grad_norm": 1.6352495113543222,
      "language_loss": 0.76404732,
      "learning_rate": 5.571662515837814e-07,
      "loss": 0.78591859,
      "num_input_tokens_seen": 136637220,
      "step": 6352,
      "time_per_iteration": 3.5616705417633057
    },
    {
      "auxiliary_loss_clip": 0.01139656,
      "auxiliary_loss_mlp": 0.0102717,
      "balance_loss_clip": 1.04367375,
      "balance_loss_mlp": 1.02039027,
      "epoch": 0.7639030842301449,
      "flos": 36283279461120.0,
      "grad_norm": 2.0516868437954585,
      "language_loss": 0.83547992,
      "learning_rate": 5.566269223580926e-07,
      "loss": 0.85714817,
      "num_input_tokens_seen": 136658930,
      "step": 6353,
      "time_per_iteration": 3.6651198863983154
    },
    {
      "auxiliary_loss_clip": 0.0115741,
      "auxiliary_loss_mlp": 0.01027687,
      "balance_loss_clip": 1.04811454,
      "balance_loss_mlp": 1.02102351,
      "epoch": 0.764023327120784,
      "flos": 28878104609280.0,
      "grad_norm": 1.7261056785343323,
      "language_loss": 0.75433356,
      "learning_rate": 5.560878120909511e-07,
      "loss": 0.77618456,
      "num_input_tokens_seen": 136681530,
      "step": 6354,
      "time_per_iteration": 2.7422101497650146
    },
    {
      "auxiliary_loss_clip": 0.01058752,
      "auxiliary_loss_mlp": 0.0100234,
      "balance_loss_clip": 1.01327538,
      "balance_loss_mlp": 1.00124955,
      "epoch": 0.7641435700114231,
      "flos": 64789711067520.0,
      "grad_norm": 0.8571558431857784,
      "language_loss": 0.5851692,
      "learning_rate": 5.55548920864141e-07,
      "loss": 0.60578012,
      "num_input_tokens_seen": 136742185,
      "step": 6355,
      "time_per_iteration": 4.180971145629883
    },
    {
      "auxiliary_loss_clip": 0.0115465,
      "auxiliary_loss_mlp": 0.01024788,
      "balance_loss_clip": 1.04810858,
      "balance_loss_mlp": 1.01815677,
      "epoch": 0.7642638129020621,
      "flos": 16835784785280.0,
      "grad_norm": 1.588978036743412,
      "language_loss": 0.77831405,
      "learning_rate": 5.550102487594113e-07,
      "loss": 0.80010843,
      "num_input_tokens_seen": 136760855,
      "step": 6356,
      "time_per_iteration": 2.6153922080993652
    },
    {
      "auxiliary_loss_clip": 0.01111972,
      "auxiliary_loss_mlp": 0.00762198,
      "balance_loss_clip": 1.03862047,
      "balance_loss_mlp": 1.00045729,
      "epoch": 0.7643840557927013,
      "flos": 30408940391040.0,
      "grad_norm": 1.6256586618319135,
      "language_loss": 0.71459651,
      "learning_rate": 5.54471795858477e-07,
      "loss": 0.73333824,
      "num_input_tokens_seen": 136780925,
      "step": 6357,
      "time_per_iteration": 2.7946784496307373
    },
    {
      "auxiliary_loss_clip": 0.01124082,
      "auxiliary_loss_mlp": 0.01020972,
      "balance_loss_clip": 1.03943038,
      "balance_loss_mlp": 1.01452017,
      "epoch": 0.7645042986833404,
      "flos": 16983234115200.0,
      "grad_norm": 1.8972945962205445,
      "language_loss": 0.82961899,
      "learning_rate": 5.539335622430235e-07,
      "loss": 0.85106957,
      "num_input_tokens_seen": 136799545,
      "step": 6358,
      "time_per_iteration": 2.648392677307129
    },
    {
      "auxiliary_loss_clip": 0.01145322,
      "auxiliary_loss_mlp": 0.01022217,
      "balance_loss_clip": 1.04190731,
      "balance_loss_mlp": 1.01554406,
      "epoch": 0.7646245415739794,
      "flos": 17311493531520.0,
      "grad_norm": 1.9391787754919432,
      "language_loss": 0.74411118,
      "learning_rate": 5.533955479946975e-07,
      "loss": 0.76578653,
      "num_input_tokens_seen": 136818325,
      "step": 6359,
      "time_per_iteration": 2.6658008098602295
    },
    {
      "auxiliary_loss_clip": 0.01036751,
      "auxiliary_loss_mlp": 0.00753627,
      "balance_loss_clip": 1.02209842,
      "balance_loss_mlp": 1.00019169,
      "epoch": 0.7647447844646186,
      "flos": 70402332666240.0,
      "grad_norm": 0.8588667615629321,
      "language_loss": 0.6581229,
      "learning_rate": 5.528577531951173e-07,
      "loss": 0.6760267,
      "num_input_tokens_seen": 136878730,
      "step": 6360,
      "time_per_iteration": 3.2270116806030273
    },
    {
      "auxiliary_loss_clip": 0.0113993,
      "auxiliary_loss_mlp": 0.01022884,
      "balance_loss_clip": 1.04232228,
      "balance_loss_mlp": 1.01636052,
      "epoch": 0.7648650273552576,
      "flos": 17675914965120.0,
      "grad_norm": 1.9049725639900656,
      "language_loss": 0.73698354,
      "learning_rate": 5.523201779258653e-07,
      "loss": 0.75861168,
      "num_input_tokens_seen": 136897705,
      "step": 6361,
      "time_per_iteration": 2.606013536453247
    },
    {
      "auxiliary_loss_clip": 0.01165237,
      "auxiliary_loss_mlp": 0.01025246,
      "balance_loss_clip": 1.04492331,
      "balance_loss_mlp": 1.0177958,
      "epoch": 0.7649852702458967,
      "flos": 22162019247360.0,
      "grad_norm": 1.703063274518249,
      "language_loss": 0.84262884,
      "learning_rate": 5.517828222684912e-07,
      "loss": 0.86453366,
      "num_input_tokens_seen": 136918360,
      "step": 6362,
      "time_per_iteration": 2.6017954349517822
    },
    {
      "auxiliary_loss_clip": 0.01043643,
      "auxiliary_loss_mlp": 0.01000853,
      "balance_loss_clip": 1.01116824,
      "balance_loss_mlp": 0.99974996,
      "epoch": 0.7651055131365359,
      "flos": 69848338227840.0,
      "grad_norm": 0.7729796986747877,
      "language_loss": 0.59037185,
      "learning_rate": 5.512456863045117e-07,
      "loss": 0.61081672,
      "num_input_tokens_seen": 136979050,
      "step": 6363,
      "time_per_iteration": 3.2389886379241943
    },
    {
      "auxiliary_loss_clip": 0.01168057,
      "auxiliary_loss_mlp": 0.01024908,
      "balance_loss_clip": 1.0462383,
      "balance_loss_mlp": 1.01824093,
      "epoch": 0.7652257560271749,
      "flos": 19464014931840.0,
      "grad_norm": 1.7538903395187544,
      "language_loss": 0.74465001,
      "learning_rate": 5.507087701154089e-07,
      "loss": 0.76657969,
      "num_input_tokens_seen": 136998970,
      "step": 6364,
      "time_per_iteration": 2.6605122089385986
    },
    {
      "auxiliary_loss_clip": 0.01114568,
      "auxiliary_loss_mlp": 0.01027344,
      "balance_loss_clip": 1.04169679,
      "balance_loss_mlp": 1.02022398,
      "epoch": 0.765345998917814,
      "flos": 15961108700160.0,
      "grad_norm": 1.9669878462908983,
      "language_loss": 0.75689614,
      "learning_rate": 5.50172073782634e-07,
      "loss": 0.77831519,
      "num_input_tokens_seen": 137016950,
      "step": 6365,
      "time_per_iteration": 2.7000157833099365
    },
    {
      "auxiliary_loss_clip": 0.01124328,
      "auxiliary_loss_mlp": 0.01021195,
      "balance_loss_clip": 1.04524255,
      "balance_loss_mlp": 1.0147934,
      "epoch": 0.7654662418084531,
      "flos": 23659853408640.0,
      "grad_norm": 1.715404366456008,
      "language_loss": 0.87721127,
      "learning_rate": 5.496355973876023e-07,
      "loss": 0.8986665,
      "num_input_tokens_seen": 137036205,
      "step": 6366,
      "time_per_iteration": 2.8411295413970947
    },
    {
      "auxiliary_loss_clip": 0.01123729,
      "auxiliary_loss_mlp": 0.0076232,
      "balance_loss_clip": 1.04118633,
      "balance_loss_mlp": 1.00037766,
      "epoch": 0.7655864846990922,
      "flos": 41463608878080.0,
      "grad_norm": 1.7124703536233545,
      "language_loss": 0.70787239,
      "learning_rate": 5.490993410116984e-07,
      "loss": 0.72673285,
      "num_input_tokens_seen": 137059195,
      "step": 6367,
      "time_per_iteration": 2.9521539211273193
    },
    {
      "auxiliary_loss_clip": 0.01117395,
      "auxiliary_loss_mlp": 0.01027402,
      "balance_loss_clip": 1.04118943,
      "balance_loss_mlp": 1.01968265,
      "epoch": 0.7657067275897312,
      "flos": 43142684088960.0,
      "grad_norm": 1.8016545720064288,
      "language_loss": 0.69636142,
      "learning_rate": 5.485633047362704e-07,
      "loss": 0.71780938,
      "num_input_tokens_seen": 137081200,
      "step": 6368,
      "time_per_iteration": 2.962754726409912
    },
    {
      "auxiliary_loss_clip": 0.01172731,
      "auxiliary_loss_mlp": 0.01030218,
      "balance_loss_clip": 1.04939687,
      "balance_loss_mlp": 1.02227592,
      "epoch": 0.7658269704803703,
      "flos": 17311780840320.0,
      "grad_norm": 2.1352557196145625,
      "language_loss": 0.78828049,
      "learning_rate": 5.480274886426341e-07,
      "loss": 0.81030989,
      "num_input_tokens_seen": 137097840,
      "step": 6369,
      "time_per_iteration": 2.552774429321289
    },
    {
      "auxiliary_loss_clip": 0.01146926,
      "auxiliary_loss_mlp": 0.01025921,
      "balance_loss_clip": 1.04545712,
      "balance_loss_mlp": 1.01937652,
      "epoch": 0.7659472133710095,
      "flos": 12568160977920.0,
      "grad_norm": 2.620189900900933,
      "language_loss": 0.77579921,
      "learning_rate": 5.474918928120744e-07,
      "loss": 0.79752767,
      "num_input_tokens_seen": 137114335,
      "step": 6370,
      "time_per_iteration": 2.650662660598755
    },
    {
      "auxiliary_loss_clip": 0.01150304,
      "auxiliary_loss_mlp": 0.01031305,
      "balance_loss_clip": 1.04539502,
      "balance_loss_mlp": 1.02510309,
      "epoch": 0.7660674562616485,
      "flos": 22707430335360.0,
      "grad_norm": 1.6394062540774934,
      "language_loss": 0.87166619,
      "learning_rate": 5.469565173258392e-07,
      "loss": 0.89348227,
      "num_input_tokens_seen": 137132850,
      "step": 6371,
      "time_per_iteration": 2.665968179702759
    },
    {
      "auxiliary_loss_clip": 0.01170207,
      "auxiliary_loss_mlp": 0.01022013,
      "balance_loss_clip": 1.04656506,
      "balance_loss_mlp": 1.0144136,
      "epoch": 0.7661876991522876,
      "flos": 17056455989760.0,
      "grad_norm": 1.6926994880095982,
      "language_loss": 0.63532573,
      "learning_rate": 5.464213622651454e-07,
      "loss": 0.6572479,
      "num_input_tokens_seen": 137150665,
      "step": 6372,
      "time_per_iteration": 2.6460282802581787
    },
    {
      "auxiliary_loss_clip": 0.01131477,
      "auxiliary_loss_mlp": 0.01024372,
      "balance_loss_clip": 1.04250145,
      "balance_loss_mlp": 1.01703799,
      "epoch": 0.7663079420429267,
      "flos": 20084228092800.0,
      "grad_norm": 1.59063295053305,
      "language_loss": 0.84140569,
      "learning_rate": 5.458864277111753e-07,
      "loss": 0.86296421,
      "num_input_tokens_seen": 137168500,
      "step": 6373,
      "time_per_iteration": 2.699895143508911
    },
    {
      "auxiliary_loss_clip": 0.01133515,
      "auxiliary_loss_mlp": 0.00761558,
      "balance_loss_clip": 1.0424211,
      "balance_loss_mlp": 1.00029063,
      "epoch": 0.7664281849335658,
      "flos": 12677473042560.0,
      "grad_norm": 2.531034092465509,
      "language_loss": 0.69403011,
      "learning_rate": 5.453517137450769e-07,
      "loss": 0.71298087,
      "num_input_tokens_seen": 137185075,
      "step": 6374,
      "time_per_iteration": 2.7018518447875977
    },
    {
      "auxiliary_loss_clip": 0.0115284,
      "auxiliary_loss_mlp": 0.01026562,
      "balance_loss_clip": 1.04661727,
      "balance_loss_mlp": 1.01948714,
      "epoch": 0.7665484278242048,
      "flos": 22345271458560.0,
      "grad_norm": 2.129378241758401,
      "language_loss": 0.76058298,
      "learning_rate": 5.448172204479684e-07,
      "loss": 0.782377,
      "num_input_tokens_seen": 137204355,
      "step": 6375,
      "time_per_iteration": 3.6241307258605957
    },
    {
      "auxiliary_loss_clip": 0.01165316,
      "auxiliary_loss_mlp": 0.01026001,
      "balance_loss_clip": 1.04522657,
      "balance_loss_mlp": 1.0193727,
      "epoch": 0.766668670714844,
      "flos": 23617909301760.0,
      "grad_norm": 1.6742809047910137,
      "language_loss": 0.74979305,
      "learning_rate": 5.442829479009294e-07,
      "loss": 0.77170622,
      "num_input_tokens_seen": 137223135,
      "step": 6376,
      "time_per_iteration": 2.64015531539917
    },
    {
      "auxiliary_loss_clip": 0.01160103,
      "auxiliary_loss_mlp": 0.01025826,
      "balance_loss_clip": 1.04567993,
      "balance_loss_mlp": 1.01853693,
      "epoch": 0.7667889136054831,
      "flos": 19427134642560.0,
      "grad_norm": 2.2569588863758177,
      "language_loss": 0.71472394,
      "learning_rate": 5.437488961850103e-07,
      "loss": 0.73658323,
      "num_input_tokens_seen": 137242935,
      "step": 6377,
      "time_per_iteration": 2.7714951038360596
    },
    {
      "auxiliary_loss_clip": 0.01106276,
      "auxiliary_loss_mlp": 0.01020832,
      "balance_loss_clip": 1.03959322,
      "balance_loss_mlp": 1.01406407,
      "epoch": 0.7669091564961221,
      "flos": 26866352609280.0,
      "grad_norm": 1.8524503567415964,
      "language_loss": 0.75385594,
      "learning_rate": 5.432150653812258e-07,
      "loss": 0.77512705,
      "num_input_tokens_seen": 137262970,
      "step": 6378,
      "time_per_iteration": 3.745340347290039
    },
    {
      "auxiliary_loss_clip": 0.01151036,
      "auxiliary_loss_mlp": 0.01027103,
      "balance_loss_clip": 1.0469079,
      "balance_loss_mlp": 1.02034402,
      "epoch": 0.7670293993867613,
      "flos": 12385303816320.0,
      "grad_norm": 2.1143224707120716,
      "language_loss": 0.82675147,
      "learning_rate": 5.42681455570557e-07,
      "loss": 0.8485328,
      "num_input_tokens_seen": 137279500,
      "step": 6379,
      "time_per_iteration": 3.5243401527404785
    },
    {
      "auxiliary_loss_clip": 0.01163705,
      "auxiliary_loss_mlp": 0.01025123,
      "balance_loss_clip": 1.04543662,
      "balance_loss_mlp": 1.01820874,
      "epoch": 0.7671496422774003,
      "flos": 21762944167680.0,
      "grad_norm": 1.6869785052038828,
      "language_loss": 0.64815784,
      "learning_rate": 5.42148066833954e-07,
      "loss": 0.67004609,
      "num_input_tokens_seen": 137298745,
      "step": 6380,
      "time_per_iteration": 2.6245720386505127
    },
    {
      "auxiliary_loss_clip": 0.0116734,
      "auxiliary_loss_mlp": 0.01021444,
      "balance_loss_clip": 1.04773784,
      "balance_loss_mlp": 1.01420236,
      "epoch": 0.7672698851680394,
      "flos": 21069221823360.0,
      "grad_norm": 1.9024696777833305,
      "language_loss": 0.75404704,
      "learning_rate": 5.416148992523289e-07,
      "loss": 0.77593493,
      "num_input_tokens_seen": 137317320,
      "step": 6381,
      "time_per_iteration": 3.683361291885376
    },
    {
      "auxiliary_loss_clip": 0.01083537,
      "auxiliary_loss_mlp": 0.01024279,
      "balance_loss_clip": 1.03886509,
      "balance_loss_mlp": 1.0171324,
      "epoch": 0.7673901280586786,
      "flos": 16976697840000.0,
      "grad_norm": 1.7609533473183545,
      "language_loss": 0.78604978,
      "learning_rate": 5.410819529065644e-07,
      "loss": 0.80712795,
      "num_input_tokens_seen": 137335275,
      "step": 6382,
      "time_per_iteration": 2.828155279159546
    },
    {
      "auxiliary_loss_clip": 0.01109308,
      "auxiliary_loss_mlp": 0.0102408,
      "balance_loss_clip": 1.03997087,
      "balance_loss_mlp": 1.01762819,
      "epoch": 0.7675103709493176,
      "flos": 29242669697280.0,
      "grad_norm": 2.3102895425205507,
      "language_loss": 0.65431333,
      "learning_rate": 5.405492278775079e-07,
      "loss": 0.67564726,
      "num_input_tokens_seen": 137355055,
      "step": 6383,
      "time_per_iteration": 2.824099540710449
    },
    {
      "auxiliary_loss_clip": 0.01139017,
      "auxiliary_loss_mlp": 0.01025331,
      "balance_loss_clip": 1.04169178,
      "balance_loss_mlp": 1.01833665,
      "epoch": 0.7676306138399567,
      "flos": 29023004073600.0,
      "grad_norm": 2.171812742644068,
      "language_loss": 0.79636002,
      "learning_rate": 5.400167242459732e-07,
      "loss": 0.81800354,
      "num_input_tokens_seen": 137374015,
      "step": 6384,
      "time_per_iteration": 2.774479389190674
    },
    {
      "auxiliary_loss_clip": 0.01150202,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.04473054,
      "balance_loss_mlp": 1.01952672,
      "epoch": 0.7677508567305958,
      "flos": 22565116650240.0,
      "grad_norm": 2.424215647708366,
      "language_loss": 0.80994678,
      "learning_rate": 5.394844420927405e-07,
      "loss": 0.83171248,
      "num_input_tokens_seen": 137393625,
      "step": 6385,
      "time_per_iteration": 2.7876338958740234
    },
    {
      "auxiliary_loss_clip": 0.01166727,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.04730129,
      "balance_loss_mlp": 1.01834762,
      "epoch": 0.7678710996212349,
      "flos": 25411432222080.0,
      "grad_norm": 2.3954923024685573,
      "language_loss": 0.73298776,
      "learning_rate": 5.389523814985562e-07,
      "loss": 0.75491077,
      "num_input_tokens_seen": 137413045,
      "step": 6386,
      "time_per_iteration": 2.6423001289367676
    },
    {
      "auxiliary_loss_clip": 0.01108319,
      "auxiliary_loss_mlp": 0.01022558,
      "balance_loss_clip": 1.03986537,
      "balance_loss_mlp": 1.01507735,
      "epoch": 0.767991342511874,
      "flos": 26756825063040.0,
      "grad_norm": 1.7597953076190629,
      "language_loss": 0.76496708,
      "learning_rate": 5.384205425441344e-07,
      "loss": 0.78627586,
      "num_input_tokens_seen": 137433955,
      "step": 6387,
      "time_per_iteration": 2.81948184967041
    },
    {
      "auxiliary_loss_clip": 0.01137589,
      "auxiliary_loss_mlp": 0.01024249,
      "balance_loss_clip": 1.04108226,
      "balance_loss_mlp": 1.01706672,
      "epoch": 0.7681115854025131,
      "flos": 26359509749760.0,
      "grad_norm": 1.6924046247964757,
      "language_loss": 0.84348118,
      "learning_rate": 5.378889253101537e-07,
      "loss": 0.86509955,
      "num_input_tokens_seen": 137454510,
      "step": 6388,
      "time_per_iteration": 2.7051732540130615
    },
    {
      "auxiliary_loss_clip": 0.01149997,
      "auxiliary_loss_mlp": 0.01022863,
      "balance_loss_clip": 1.04168677,
      "balance_loss_mlp": 1.01625931,
      "epoch": 0.7682318282931522,
      "flos": 23257043314560.0,
      "grad_norm": 1.6145039469297509,
      "language_loss": 0.80656451,
      "learning_rate": 5.373575298772617e-07,
      "loss": 0.82829309,
      "num_input_tokens_seen": 137473630,
      "step": 6389,
      "time_per_iteration": 2.653120994567871
    },
    {
      "auxiliary_loss_clip": 0.0105859,
      "auxiliary_loss_mlp": 0.0100172,
      "balance_loss_clip": 1.01212275,
      "balance_loss_mlp": 1.00065947,
      "epoch": 0.7683520711837912,
      "flos": 70072457137920.0,
      "grad_norm": 0.7613138994082183,
      "language_loss": 0.61344755,
      "learning_rate": 5.368263563260689e-07,
      "loss": 0.63405073,
      "num_input_tokens_seen": 137538765,
      "step": 6390,
      "time_per_iteration": 3.2916054725646973
    },
    {
      "auxiliary_loss_clip": 0.01150602,
      "auxiliary_loss_mlp": 0.01022717,
      "balance_loss_clip": 1.04272795,
      "balance_loss_mlp": 1.01535583,
      "epoch": 0.7684723140744304,
      "flos": 18624890332800.0,
      "grad_norm": 1.5440633130100656,
      "language_loss": 0.64154589,
      "learning_rate": 5.362954047371537e-07,
      "loss": 0.66327906,
      "num_input_tokens_seen": 137557875,
      "step": 6391,
      "time_per_iteration": 2.680232524871826
    },
    {
      "auxiliary_loss_clip": 0.01125509,
      "auxiliary_loss_mlp": 0.01027774,
      "balance_loss_clip": 1.0466187,
      "balance_loss_mlp": 1.02024603,
      "epoch": 0.7685925569650695,
      "flos": 27452989532160.0,
      "grad_norm": 1.8580157691117833,
      "language_loss": 0.72026002,
      "learning_rate": 5.357646751910627e-07,
      "loss": 0.74179286,
      "num_input_tokens_seen": 137579055,
      "step": 6392,
      "time_per_iteration": 2.8222649097442627
    },
    {
      "auxiliary_loss_clip": 0.01135123,
      "auxiliary_loss_mlp": 0.01030658,
      "balance_loss_clip": 1.04329586,
      "balance_loss_mlp": 1.02305496,
      "epoch": 0.7687127998557085,
      "flos": 24535714642560.0,
      "grad_norm": 2.3322299227369254,
      "language_loss": 0.80078393,
      "learning_rate": 5.352341677683061e-07,
      "loss": 0.8224417,
      "num_input_tokens_seen": 137600355,
      "step": 6393,
      "time_per_iteration": 2.7269113063812256
    },
    {
      "auxiliary_loss_clip": 0.01132089,
      "auxiliary_loss_mlp": 0.01023129,
      "balance_loss_clip": 1.04273057,
      "balance_loss_mlp": 1.0166446,
      "epoch": 0.7688330427463477,
      "flos": 25155963717120.0,
      "grad_norm": 2.003035981792588,
      "language_loss": 0.79613602,
      "learning_rate": 5.347038825493617e-07,
      "loss": 0.81768823,
      "num_input_tokens_seen": 137621885,
      "step": 6394,
      "time_per_iteration": 2.796241044998169
    },
    {
      "auxiliary_loss_clip": 0.01133463,
      "auxiliary_loss_mlp": 0.01026789,
      "balance_loss_clip": 1.04353929,
      "balance_loss_mlp": 1.02032757,
      "epoch": 0.7689532856369867,
      "flos": 21211284113280.0,
      "grad_norm": 2.122989533800318,
      "language_loss": 0.6884256,
      "learning_rate": 5.341738196146732e-07,
      "loss": 0.71002817,
      "num_input_tokens_seen": 137640230,
      "step": 6395,
      "time_per_iteration": 2.7164947986602783
    },
    {
      "auxiliary_loss_clip": 0.01146092,
      "auxiliary_loss_mlp": 0.01025766,
      "balance_loss_clip": 1.04076064,
      "balance_loss_mlp": 1.01834536,
      "epoch": 0.7690735285276258,
      "flos": 25119083427840.0,
      "grad_norm": 2.261664772624106,
      "language_loss": 0.73397601,
      "learning_rate": 5.336439790446503e-07,
      "loss": 0.75569463,
      "num_input_tokens_seen": 137659330,
      "step": 6396,
      "time_per_iteration": 2.7563045024871826
    },
    {
      "auxiliary_loss_clip": 0.01121477,
      "auxiliary_loss_mlp": 0.01025091,
      "balance_loss_clip": 1.04062676,
      "balance_loss_mlp": 1.01821864,
      "epoch": 0.769193771418265,
      "flos": 54744020640000.0,
      "grad_norm": 2.479228581580639,
      "language_loss": 0.62897038,
      "learning_rate": 5.331143609196711e-07,
      "loss": 0.65043604,
      "num_input_tokens_seen": 137683145,
      "step": 6397,
      "time_per_iteration": 2.9721250534057617
    },
    {
      "auxiliary_loss_clip": 0.01151597,
      "auxiliary_loss_mlp": 0.01024885,
      "balance_loss_clip": 1.04656434,
      "balance_loss_mlp": 1.01786923,
      "epoch": 0.769314014308904,
      "flos": 37341890115840.0,
      "grad_norm": 1.7868253641031386,
      "language_loss": 0.76941401,
      "learning_rate": 5.325849653200758e-07,
      "loss": 0.79117888,
      "num_input_tokens_seen": 137707095,
      "step": 6398,
      "time_per_iteration": 2.771365165710449
    },
    {
      "auxiliary_loss_clip": 0.0116727,
      "auxiliary_loss_mlp": 0.01029693,
      "balance_loss_clip": 1.04718995,
      "balance_loss_mlp": 1.02280867,
      "epoch": 0.7694342571995431,
      "flos": 20631686256000.0,
      "grad_norm": 3.3019738555633853,
      "language_loss": 0.76067495,
      "learning_rate": 5.32055792326175e-07,
      "loss": 0.78264457,
      "num_input_tokens_seen": 137725520,
      "step": 6399,
      "time_per_iteration": 2.649177074432373
    },
    {
      "auxiliary_loss_clip": 0.01141187,
      "auxiliary_loss_mlp": 0.01027716,
      "balance_loss_clip": 1.04605222,
      "balance_loss_mlp": 1.02076578,
      "epoch": 0.7695545000901821,
      "flos": 24207706621440.0,
      "grad_norm": 1.902866483394228,
      "language_loss": 0.72548765,
      "learning_rate": 5.315268420182437e-07,
      "loss": 0.74717665,
      "num_input_tokens_seen": 137744195,
      "step": 6400,
      "time_per_iteration": 2.660510301589966
    },
    {
      "auxiliary_loss_clip": 0.0113101,
      "auxiliary_loss_mlp": 0.0076173,
      "balance_loss_clip": 1.04269159,
      "balance_loss_mlp": 1.00035703,
      "epoch": 0.7696747429808213,
      "flos": 28001273708160.0,
      "grad_norm": 1.681992570682121,
      "language_loss": 0.76208502,
      "learning_rate": 5.309981144765221e-07,
      "loss": 0.78101242,
      "num_input_tokens_seen": 137764340,
      "step": 6401,
      "time_per_iteration": 3.6609530448913574
    },
    {
      "auxiliary_loss_clip": 0.0111497,
      "auxiliary_loss_mlp": 0.01025044,
      "balance_loss_clip": 1.03996408,
      "balance_loss_mlp": 1.01762879,
      "epoch": 0.7697949858714603,
      "flos": 11509550323200.0,
      "grad_norm": 2.6460224070408924,
      "language_loss": 0.75701189,
      "learning_rate": 5.304696097812196e-07,
      "loss": 0.77841198,
      "num_input_tokens_seen": 137780940,
      "step": 6402,
      "time_per_iteration": 2.708095073699951
    },
    {
      "auxiliary_loss_clip": 0.01136198,
      "auxiliary_loss_mlp": 0.01032156,
      "balance_loss_clip": 1.04284167,
      "balance_loss_mlp": 1.02412748,
      "epoch": 0.7699152287620994,
      "flos": 26688271956480.0,
      "grad_norm": 2.63760172135628,
      "language_loss": 0.61183774,
      "learning_rate": 5.299413280125078e-07,
      "loss": 0.63352126,
      "num_input_tokens_seen": 137799250,
      "step": 6403,
      "time_per_iteration": 2.8732190132141113
    },
    {
      "auxiliary_loss_clip": 0.01139578,
      "auxiliary_loss_mlp": 0.01021154,
      "balance_loss_clip": 1.04421771,
      "balance_loss_mlp": 1.01372993,
      "epoch": 0.7700354716527386,
      "flos": 16544944362240.0,
      "grad_norm": 2.296630453131005,
      "language_loss": 0.7309109,
      "learning_rate": 5.294132692505284e-07,
      "loss": 0.7525183,
      "num_input_tokens_seen": 137817660,
      "step": 6404,
      "time_per_iteration": 3.6902973651885986
    },
    {
      "auxiliary_loss_clip": 0.011034,
      "auxiliary_loss_mlp": 0.01026355,
      "balance_loss_clip": 1.04032183,
      "balance_loss_mlp": 1.01910365,
      "epoch": 0.7701557145433776,
      "flos": 19242733196160.0,
      "grad_norm": 1.8800870442130917,
      "language_loss": 0.7916317,
      "learning_rate": 5.288854335753861e-07,
      "loss": 0.81292927,
      "num_input_tokens_seen": 137835920,
      "step": 6405,
      "time_per_iteration": 3.5423619747161865
    },
    {
      "auxiliary_loss_clip": 0.01155107,
      "auxiliary_loss_mlp": 0.01025623,
      "balance_loss_clip": 1.04573441,
      "balance_loss_mlp": 1.01925707,
      "epoch": 0.7702759574340167,
      "flos": 31685744211840.0,
      "grad_norm": 1.8445855877154294,
      "language_loss": 0.75646949,
      "learning_rate": 5.283578210671551e-07,
      "loss": 0.7782768,
      "num_input_tokens_seen": 137858160,
      "step": 6406,
      "time_per_iteration": 2.739253520965576
    },
    {
      "auxiliary_loss_clip": 0.01143392,
      "auxiliary_loss_mlp": 0.01026818,
      "balance_loss_clip": 1.04487705,
      "balance_loss_mlp": 1.01998687,
      "epoch": 0.7703962003246558,
      "flos": 16800089644800.0,
      "grad_norm": 1.9500985859353615,
      "language_loss": 0.76723647,
      "learning_rate": 5.278304318058719e-07,
      "loss": 0.78893858,
      "num_input_tokens_seen": 137876015,
      "step": 6407,
      "time_per_iteration": 3.7025091648101807
    },
    {
      "auxiliary_loss_clip": 0.01094194,
      "auxiliary_loss_mlp": 0.01026279,
      "balance_loss_clip": 1.03667212,
      "balance_loss_mlp": 1.01932287,
      "epoch": 0.7705164432152949,
      "flos": 35736072693120.0,
      "grad_norm": 1.6896424429654362,
      "language_loss": 0.78950554,
      "learning_rate": 5.273032658715411e-07,
      "loss": 0.81071025,
      "num_input_tokens_seen": 137898825,
      "step": 6408,
      "time_per_iteration": 2.891700029373169
    },
    {
      "auxiliary_loss_clip": 0.01104712,
      "auxiliary_loss_mlp": 0.01022416,
      "balance_loss_clip": 1.03783691,
      "balance_loss_mlp": 1.01549911,
      "epoch": 0.7706366861059339,
      "flos": 23365960329600.0,
      "grad_norm": 1.7595666080232313,
      "language_loss": 0.76452291,
      "learning_rate": 5.267763233441347e-07,
      "loss": 0.78579414,
      "num_input_tokens_seen": 137919455,
      "step": 6409,
      "time_per_iteration": 2.836327314376831
    },
    {
      "auxiliary_loss_clip": 0.01156346,
      "auxiliary_loss_mlp": 0.01021902,
      "balance_loss_clip": 1.04629374,
      "balance_loss_mlp": 1.01404619,
      "epoch": 0.7707569289965731,
      "flos": 22929897219840.0,
      "grad_norm": 3.1028401347576526,
      "language_loss": 0.69501936,
      "learning_rate": 5.26249604303588e-07,
      "loss": 0.71680188,
      "num_input_tokens_seen": 137937960,
      "step": 6410,
      "time_per_iteration": 2.6974172592163086
    },
    {
      "auxiliary_loss_clip": 0.01168912,
      "auxiliary_loss_mlp": 0.01030375,
      "balance_loss_clip": 1.04803884,
      "balance_loss_mlp": 1.02307963,
      "epoch": 0.7708771718872122,
      "flos": 17420661941760.0,
      "grad_norm": 3.58273021846403,
      "language_loss": 0.78129029,
      "learning_rate": 5.257231088298057e-07,
      "loss": 0.8032831,
      "num_input_tokens_seen": 137956370,
      "step": 6411,
      "time_per_iteration": 2.6282572746276855
    },
    {
      "auxiliary_loss_clip": 0.01031061,
      "auxiliary_loss_mlp": 0.0100179,
      "balance_loss_clip": 1.01102996,
      "balance_loss_mlp": 1.00065792,
      "epoch": 0.7709974147778512,
      "flos": 72241316248320.0,
      "grad_norm": 0.7910575183030759,
      "language_loss": 0.53945106,
      "learning_rate": 5.25196837002655e-07,
      "loss": 0.55977958,
      "num_input_tokens_seen": 138016080,
      "step": 6412,
      "time_per_iteration": 3.3573949337005615
    },
    {
      "auxiliary_loss_clip": 0.01135268,
      "auxiliary_loss_mlp": 0.01027217,
      "balance_loss_clip": 1.0434736,
      "balance_loss_mlp": 1.01931906,
      "epoch": 0.7711176576684904,
      "flos": 39859694876160.0,
      "grad_norm": 2.27898180571654,
      "language_loss": 0.6877116,
      "learning_rate": 5.24670788901971e-07,
      "loss": 0.7093364,
      "num_input_tokens_seen": 138039170,
      "step": 6413,
      "time_per_iteration": 2.9168944358825684
    },
    {
      "auxiliary_loss_clip": 0.01139074,
      "auxiliary_loss_mlp": 0.01031199,
      "balance_loss_clip": 1.04611027,
      "balance_loss_mlp": 1.0227803,
      "epoch": 0.7712379005591294,
      "flos": 36976391274240.0,
      "grad_norm": 2.2494557467210674,
      "language_loss": 0.6817652,
      "learning_rate": 5.241449646075557e-07,
      "loss": 0.70346797,
      "num_input_tokens_seen": 138062395,
      "step": 6414,
      "time_per_iteration": 2.8658511638641357
    },
    {
      "auxiliary_loss_clip": 0.01160076,
      "auxiliary_loss_mlp": 0.01025015,
      "balance_loss_clip": 1.0453136,
      "balance_loss_mlp": 1.01789558,
      "epoch": 0.7713581434497685,
      "flos": 22776773541120.0,
      "grad_norm": 2.664660940046069,
      "language_loss": 0.72140026,
      "learning_rate": 5.236193641991762e-07,
      "loss": 0.7432512,
      "num_input_tokens_seen": 138080325,
      "step": 6415,
      "time_per_iteration": 2.63266921043396
    },
    {
      "auxiliary_loss_clip": 0.01137436,
      "auxiliary_loss_mlp": 0.01019783,
      "balance_loss_clip": 1.04445648,
      "balance_loss_mlp": 1.01245213,
      "epoch": 0.7714783863404077,
      "flos": 24097460803200.0,
      "grad_norm": 2.0731486619305763,
      "language_loss": 0.70005715,
      "learning_rate": 5.23093987756565e-07,
      "loss": 0.72162926,
      "num_input_tokens_seen": 138099020,
      "step": 6416,
      "time_per_iteration": 2.727823495864868
    },
    {
      "auxiliary_loss_clip": 0.01132973,
      "auxiliary_loss_mlp": 0.01026554,
      "balance_loss_clip": 1.04063606,
      "balance_loss_mlp": 1.01826334,
      "epoch": 0.7715986292310467,
      "flos": 21063655215360.0,
      "grad_norm": 3.7112545916164765,
      "language_loss": 0.75302947,
      "learning_rate": 5.225688353594217e-07,
      "loss": 0.77462476,
      "num_input_tokens_seen": 138118650,
      "step": 6417,
      "time_per_iteration": 2.7487869262695312
    },
    {
      "auxiliary_loss_clip": 0.01147236,
      "auxiliary_loss_mlp": 0.00762056,
      "balance_loss_clip": 1.04795706,
      "balance_loss_mlp": 1.00033951,
      "epoch": 0.7717188721216858,
      "flos": 20594877793920.0,
      "grad_norm": 2.199670325998729,
      "language_loss": 0.77259552,
      "learning_rate": 5.220439070874108e-07,
      "loss": 0.79168844,
      "num_input_tokens_seen": 138137890,
      "step": 6418,
      "time_per_iteration": 2.7488796710968018
    },
    {
      "auxiliary_loss_clip": 0.01153026,
      "auxiliary_loss_mlp": 0.01026234,
      "balance_loss_clip": 1.0466218,
      "balance_loss_mlp": 1.01869369,
      "epoch": 0.7718391150123249,
      "flos": 26250951870720.0,
      "grad_norm": 1.7166794809938501,
      "language_loss": 0.71128881,
      "learning_rate": 5.215192030201652e-07,
      "loss": 0.73308146,
      "num_input_tokens_seen": 138158880,
      "step": 6419,
      "time_per_iteration": 2.719557762145996
    },
    {
      "auxiliary_loss_clip": 0.01109772,
      "auxiliary_loss_mlp": 0.01032763,
      "balance_loss_clip": 1.0366323,
      "balance_loss_mlp": 1.02619743,
      "epoch": 0.771959357902964,
      "flos": 22049762267520.0,
      "grad_norm": 1.772813829255,
      "language_loss": 0.86211777,
      "learning_rate": 5.209947232372798e-07,
      "loss": 0.88354313,
      "num_input_tokens_seen": 138176370,
      "step": 6420,
      "time_per_iteration": 2.753448724746704
    },
    {
      "auxiliary_loss_clip": 0.01155267,
      "auxiliary_loss_mlp": 0.00762554,
      "balance_loss_clip": 1.04376078,
      "balance_loss_mlp": 1.00036001,
      "epoch": 0.772079600793603,
      "flos": 30446000248320.0,
      "grad_norm": 2.5040273237501736,
      "language_loss": 0.81016463,
      "learning_rate": 5.204704678183196e-07,
      "loss": 0.82934284,
      "num_input_tokens_seen": 138195105,
      "step": 6421,
      "time_per_iteration": 2.7588725090026855
    },
    {
      "auxiliary_loss_clip": 0.0116796,
      "auxiliary_loss_mlp": 0.01025801,
      "balance_loss_clip": 1.04771173,
      "balance_loss_mlp": 1.01866031,
      "epoch": 0.7721998436842422,
      "flos": 12969857750400.0,
      "grad_norm": 1.9976498468927195,
      "language_loss": 0.85269761,
      "learning_rate": 5.19946436842813e-07,
      "loss": 0.87463522,
      "num_input_tokens_seen": 138212235,
      "step": 6422,
      "time_per_iteration": 2.7765920162200928
    },
    {
      "auxiliary_loss_clip": 0.01124486,
      "auxiliary_loss_mlp": 0.01020184,
      "balance_loss_clip": 1.04324126,
      "balance_loss_mlp": 1.01340711,
      "epoch": 0.7723200865748813,
      "flos": 32635509678720.0,
      "grad_norm": 1.729676874727167,
      "language_loss": 0.68092155,
      "learning_rate": 5.194226303902546e-07,
      "loss": 0.70236826,
      "num_input_tokens_seen": 138231970,
      "step": 6423,
      "time_per_iteration": 2.8819327354431152
    },
    {
      "auxiliary_loss_clip": 0.01135946,
      "auxiliary_loss_mlp": 0.01028582,
      "balance_loss_clip": 1.04212928,
      "balance_loss_mlp": 1.02133954,
      "epoch": 0.7724403294655203,
      "flos": 21105707063040.0,
      "grad_norm": 2.070220723427855,
      "language_loss": 0.70741105,
      "learning_rate": 5.188990485401072e-07,
      "loss": 0.72905636,
      "num_input_tokens_seen": 138251175,
      "step": 6424,
      "time_per_iteration": 2.7188920974731445
    },
    {
      "auxiliary_loss_clip": 0.01154784,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.04670763,
      "balance_loss_mlp": 1.01873565,
      "epoch": 0.7725605723561595,
      "flos": 22090736707200.0,
      "grad_norm": 1.875834491467513,
      "language_loss": 0.85838175,
      "learning_rate": 5.183756913717954e-07,
      "loss": 0.88019335,
      "num_input_tokens_seen": 138270950,
      "step": 6425,
      "time_per_iteration": 2.8070931434631348
    },
    {
      "auxiliary_loss_clip": 0.01135489,
      "auxiliary_loss_mlp": 0.0102373,
      "balance_loss_clip": 1.04433084,
      "balance_loss_mlp": 1.01660156,
      "epoch": 0.7726808152467985,
      "flos": 34495610457600.0,
      "grad_norm": 1.6632357324463254,
      "language_loss": 0.73441005,
      "learning_rate": 5.178525589647136e-07,
      "loss": 0.75600219,
      "num_input_tokens_seen": 138292590,
      "step": 6426,
      "time_per_iteration": 2.770979881286621
    },
    {
      "auxiliary_loss_clip": 0.01144802,
      "auxiliary_loss_mlp": 0.01026127,
      "balance_loss_clip": 1.04504752,
      "balance_loss_mlp": 1.01901674,
      "epoch": 0.7728010581374376,
      "flos": 22306344094080.0,
      "grad_norm": 1.7742075947940026,
      "language_loss": 0.78883708,
      "learning_rate": 5.173296513982197e-07,
      "loss": 0.8105464,
      "num_input_tokens_seen": 138311115,
      "step": 6427,
      "time_per_iteration": 3.671598196029663
    },
    {
      "auxiliary_loss_clip": 0.01136836,
      "auxiliary_loss_mlp": 0.01026063,
      "balance_loss_clip": 1.04641128,
      "balance_loss_mlp": 1.01867771,
      "epoch": 0.7729213010280768,
      "flos": 27126453968640.0,
      "grad_norm": 1.9817430158183396,
      "language_loss": 0.6527372,
      "learning_rate": 5.168069687516398e-07,
      "loss": 0.67436624,
      "num_input_tokens_seen": 138330885,
      "step": 6428,
      "time_per_iteration": 2.7427639961242676
    },
    {
      "auxiliary_loss_clip": 0.01139192,
      "auxiliary_loss_mlp": 0.01027139,
      "balance_loss_clip": 1.04598558,
      "balance_loss_mlp": 1.01996863,
      "epoch": 0.7730415439187158,
      "flos": 18150223080960.0,
      "grad_norm": 2.0519430786425334,
      "language_loss": 0.71908963,
      "learning_rate": 5.16284511104263e-07,
      "loss": 0.74075294,
      "num_input_tokens_seen": 138350020,
      "step": 6429,
      "time_per_iteration": 3.5955750942230225
    },
    {
      "auxiliary_loss_clip": 0.011372,
      "auxiliary_loss_mlp": 0.01024811,
      "balance_loss_clip": 1.04296303,
      "balance_loss_mlp": 1.01756358,
      "epoch": 0.7731617868093549,
      "flos": 11947480940160.0,
      "grad_norm": 2.44053418087118,
      "language_loss": 0.81044734,
      "learning_rate": 5.157622785353457e-07,
      "loss": 0.83206737,
      "num_input_tokens_seen": 138368135,
      "step": 6430,
      "time_per_iteration": 3.499960422515869
    },
    {
      "auxiliary_loss_clip": 0.01059354,
      "auxiliary_loss_mlp": 0.01001041,
      "balance_loss_clip": 1.01350415,
      "balance_loss_mlp": 0.99990886,
      "epoch": 0.7732820296999939,
      "flos": 64201027069440.0,
      "grad_norm": 0.6459628709422021,
      "language_loss": 0.60392249,
      "learning_rate": 5.152402711241113e-07,
      "loss": 0.62452644,
      "num_input_tokens_seen": 138436040,
      "step": 6431,
      "time_per_iteration": 3.3474714756011963
    },
    {
      "auxiliary_loss_clip": 0.01121319,
      "auxiliary_loss_mlp": 0.01024232,
      "balance_loss_clip": 1.04037702,
      "balance_loss_mlp": 1.01734161,
      "epoch": 0.7734022725906331,
      "flos": 25302191984640.0,
      "grad_norm": 1.7783610017447533,
      "language_loss": 0.82917446,
      "learning_rate": 5.147184889497465e-07,
      "loss": 0.85062993,
      "num_input_tokens_seen": 138455510,
      "step": 6432,
      "time_per_iteration": 2.7860913276672363
    },
    {
      "auxiliary_loss_clip": 0.01115844,
      "auxiliary_loss_mlp": 0.01029796,
      "balance_loss_clip": 1.04077435,
      "balance_loss_mlp": 1.02166593,
      "epoch": 0.7735225154812722,
      "flos": 17347440067200.0,
      "grad_norm": 4.056209079891061,
      "language_loss": 0.80328608,
      "learning_rate": 5.141969320914072e-07,
      "loss": 0.82474244,
      "num_input_tokens_seen": 138473015,
      "step": 6433,
      "time_per_iteration": 3.8188586235046387
    },
    {
      "auxiliary_loss_clip": 0.01169918,
      "auxiliary_loss_mlp": 0.01030235,
      "balance_loss_clip": 1.04704857,
      "balance_loss_mlp": 1.02224803,
      "epoch": 0.7736427583719112,
      "flos": 32630086725120.0,
      "grad_norm": 22.157868158695997,
      "language_loss": 0.62548733,
      "learning_rate": 5.136756006282113e-07,
      "loss": 0.64748883,
      "num_input_tokens_seen": 138491680,
      "step": 6434,
      "time_per_iteration": 2.678572177886963
    },
    {
      "auxiliary_loss_clip": 0.01169601,
      "auxiliary_loss_mlp": 0.01025116,
      "balance_loss_clip": 1.0476625,
      "balance_loss_mlp": 1.01793087,
      "epoch": 0.7737630012625504,
      "flos": 19860073269120.0,
      "grad_norm": 2.1620797324315184,
      "language_loss": 0.84767663,
      "learning_rate": 5.131544946392446e-07,
      "loss": 0.86962378,
      "num_input_tokens_seen": 138506960,
      "step": 6435,
      "time_per_iteration": 2.6757941246032715
    },
    {
      "auxiliary_loss_clip": 0.01136752,
      "auxiliary_loss_mlp": 0.0102883,
      "balance_loss_clip": 1.04515851,
      "balance_loss_mlp": 1.02134633,
      "epoch": 0.7738832441531894,
      "flos": 36022639397760.0,
      "grad_norm": 2.3707517759969563,
      "language_loss": 0.63956642,
      "learning_rate": 5.126336142035592e-07,
      "loss": 0.66122222,
      "num_input_tokens_seen": 138526995,
      "step": 6436,
      "time_per_iteration": 2.858335256576538
    },
    {
      "auxiliary_loss_clip": 0.01138007,
      "auxiliary_loss_mlp": 0.01025221,
      "balance_loss_clip": 1.04168022,
      "balance_loss_mlp": 1.01741242,
      "epoch": 0.7740034870438285,
      "flos": 13405274415360.0,
      "grad_norm": 2.404318442976146,
      "language_loss": 0.71885681,
      "learning_rate": 5.121129594001721e-07,
      "loss": 0.74048907,
      "num_input_tokens_seen": 138541260,
      "step": 6437,
      "time_per_iteration": 2.679598331451416
    },
    {
      "auxiliary_loss_clip": 0.01151714,
      "auxiliary_loss_mlp": 0.01023939,
      "balance_loss_clip": 1.04493713,
      "balance_loss_mlp": 1.01681638,
      "epoch": 0.7741237299344677,
      "flos": 22086714384000.0,
      "grad_norm": 1.562178010772168,
      "language_loss": 0.81174451,
      "learning_rate": 5.115925303080661e-07,
      "loss": 0.8335011,
      "num_input_tokens_seen": 138560970,
      "step": 6438,
      "time_per_iteration": 2.6316263675689697
    },
    {
      "auxiliary_loss_clip": 0.01138886,
      "auxiliary_loss_mlp": 0.01026225,
      "balance_loss_clip": 1.0429461,
      "balance_loss_mlp": 1.01920366,
      "epoch": 0.7742439728251067,
      "flos": 19864777950720.0,
      "grad_norm": 2.204659357807789,
      "language_loss": 0.79076701,
      "learning_rate": 5.110723270061899e-07,
      "loss": 0.8124181,
      "num_input_tokens_seen": 138577460,
      "step": 6439,
      "time_per_iteration": 2.6606359481811523
    },
    {
      "auxiliary_loss_clip": 0.01164415,
      "auxiliary_loss_mlp": 0.01024201,
      "balance_loss_clip": 1.04537988,
      "balance_loss_mlp": 1.01727772,
      "epoch": 0.7743642157157458,
      "flos": 16690167048960.0,
      "grad_norm": 1.9023463630591777,
      "language_loss": 0.79318559,
      "learning_rate": 5.105523495734572e-07,
      "loss": 0.8150717,
      "num_input_tokens_seen": 138594860,
      "step": 6440,
      "time_per_iteration": 2.6125144958496094
    },
    {
      "auxiliary_loss_clip": 0.01163638,
      "auxiliary_loss_mlp": 0.01026098,
      "balance_loss_clip": 1.04316127,
      "balance_loss_mlp": 1.01902902,
      "epoch": 0.7744844586063849,
      "flos": 20304360593280.0,
      "grad_norm": 1.6854903294054728,
      "language_loss": 0.74844944,
      "learning_rate": 5.100325980887499e-07,
      "loss": 0.77034676,
      "num_input_tokens_seen": 138614785,
      "step": 6441,
      "time_per_iteration": 2.6861441135406494
    },
    {
      "auxiliary_loss_clip": 0.01148197,
      "auxiliary_loss_mlp": 0.01026856,
      "balance_loss_clip": 1.04539585,
      "balance_loss_mlp": 1.01976919,
      "epoch": 0.774604701497024,
      "flos": 22966705681920.0,
      "grad_norm": 1.8463180804138655,
      "language_loss": 0.83195031,
      "learning_rate": 5.095130726309116e-07,
      "loss": 0.85370088,
      "num_input_tokens_seen": 138634960,
      "step": 6442,
      "time_per_iteration": 2.7292332649230957
    },
    {
      "auxiliary_loss_clip": 0.0106823,
      "auxiliary_loss_mlp": 0.01001697,
      "balance_loss_clip": 1.01348519,
      "balance_loss_mlp": 1.0005765,
      "epoch": 0.774724944387663,
      "flos": 60288523073280.0,
      "grad_norm": 0.7941688999338065,
      "language_loss": 0.59057462,
      "learning_rate": 5.089937732787559e-07,
      "loss": 0.61127388,
      "num_input_tokens_seen": 138699520,
      "step": 6443,
      "time_per_iteration": 3.2514429092407227
    },
    {
      "auxiliary_loss_clip": 0.01126128,
      "auxiliary_loss_mlp": 0.01025154,
      "balance_loss_clip": 1.04298973,
      "balance_loss_mlp": 1.01760232,
      "epoch": 0.7748451872783022,
      "flos": 26761026954240.0,
      "grad_norm": 3.9225416558798964,
      "language_loss": 0.66530561,
      "learning_rate": 5.084747001110592e-07,
      "loss": 0.68681842,
      "num_input_tokens_seen": 138719145,
      "step": 6444,
      "time_per_iteration": 2.8492958545684814
    },
    {
      "auxiliary_loss_clip": 0.0114794,
      "auxiliary_loss_mlp": 0.00761955,
      "balance_loss_clip": 1.04622149,
      "balance_loss_mlp": 1.00038099,
      "epoch": 0.7749654301689413,
      "flos": 30338627518080.0,
      "grad_norm": 1.822902785445455,
      "language_loss": 0.69928116,
      "learning_rate": 5.07955853206564e-07,
      "loss": 0.71838009,
      "num_input_tokens_seen": 138743850,
      "step": 6445,
      "time_per_iteration": 2.746682643890381
    },
    {
      "auxiliary_loss_clip": 0.0115542,
      "auxiliary_loss_mlp": 0.01026645,
      "balance_loss_clip": 1.04482055,
      "balance_loss_mlp": 1.01948404,
      "epoch": 0.7750856730595803,
      "flos": 43179851687040.0,
      "grad_norm": 1.6007121356050662,
      "language_loss": 0.70974147,
      "learning_rate": 5.074372326439807e-07,
      "loss": 0.73156214,
      "num_input_tokens_seen": 138766860,
      "step": 6446,
      "time_per_iteration": 2.8201136589050293
    },
    {
      "auxiliary_loss_clip": 0.01124826,
      "auxiliary_loss_mlp": 0.01021973,
      "balance_loss_clip": 1.04125595,
      "balance_loss_mlp": 1.01509202,
      "epoch": 0.7752059159502195,
      "flos": 17640040256640.0,
      "grad_norm": 2.618848042312786,
      "language_loss": 0.73940384,
      "learning_rate": 5.069188385019814e-07,
      "loss": 0.76087183,
      "num_input_tokens_seen": 138784560,
      "step": 6447,
      "time_per_iteration": 2.6787567138671875
    },
    {
      "auxiliary_loss_clip": 0.01114355,
      "auxiliary_loss_mlp": 0.01026693,
      "balance_loss_clip": 1.0391345,
      "balance_loss_mlp": 1.0196414,
      "epoch": 0.7753261588408585,
      "flos": 12677688524160.0,
      "grad_norm": 3.0016443703734415,
      "language_loss": 0.61175913,
      "learning_rate": 5.064006708592077e-07,
      "loss": 0.63316953,
      "num_input_tokens_seen": 138800805,
      "step": 6448,
      "time_per_iteration": 2.6894946098327637
    },
    {
      "auxiliary_loss_clip": 0.01131387,
      "auxiliary_loss_mlp": 0.01026369,
      "balance_loss_clip": 1.04410458,
      "balance_loss_mlp": 1.01965165,
      "epoch": 0.7754464017314976,
      "flos": 16690741666560.0,
      "grad_norm": 2.478813443954334,
      "language_loss": 0.75778174,
      "learning_rate": 5.058827297942641e-07,
      "loss": 0.77935928,
      "num_input_tokens_seen": 138815910,
      "step": 6449,
      "time_per_iteration": 2.6561739444732666
    },
    {
      "auxiliary_loss_clip": 0.01145461,
      "auxiliary_loss_mlp": 0.01024964,
      "balance_loss_clip": 1.04478037,
      "balance_loss_mlp": 1.01799667,
      "epoch": 0.7755666446221368,
      "flos": 19718944732800.0,
      "grad_norm": 1.8846018491981815,
      "language_loss": 0.75187117,
      "learning_rate": 5.053650153857237e-07,
      "loss": 0.77357543,
      "num_input_tokens_seen": 138834920,
      "step": 6450,
      "time_per_iteration": 2.659416675567627
    },
    {
      "auxiliary_loss_clip": 0.011523,
      "auxiliary_loss_mlp": 0.01021156,
      "balance_loss_clip": 1.04715598,
      "balance_loss_mlp": 1.0139823,
      "epoch": 0.7756868875127758,
      "flos": 18693623007360.0,
      "grad_norm": 1.5774534707436472,
      "language_loss": 0.6981768,
      "learning_rate": 5.048475277121214e-07,
      "loss": 0.71991134,
      "num_input_tokens_seen": 138852135,
      "step": 6451,
      "time_per_iteration": 2.7153401374816895
    },
    {
      "auxiliary_loss_clip": 0.01155226,
      "auxiliary_loss_mlp": 0.01020876,
      "balance_loss_clip": 1.04583013,
      "balance_loss_mlp": 1.01400697,
      "epoch": 0.7758071304034149,
      "flos": 28404191543040.0,
      "grad_norm": 1.8933194641412132,
      "language_loss": 0.76703584,
      "learning_rate": 5.043302668519598e-07,
      "loss": 0.7887969,
      "num_input_tokens_seen": 138871470,
      "step": 6452,
      "time_per_iteration": 2.6996614933013916
    },
    {
      "auxiliary_loss_clip": 0.01155932,
      "auxiliary_loss_mlp": 0.01024086,
      "balance_loss_clip": 1.04545593,
      "balance_loss_mlp": 1.01740408,
      "epoch": 0.775927373294054,
      "flos": 20595344670720.0,
      "grad_norm": 2.048114495790253,
      "language_loss": 0.72242427,
      "learning_rate": 5.038132328837079e-07,
      "loss": 0.74422443,
      "num_input_tokens_seen": 138889860,
      "step": 6453,
      "time_per_iteration": 3.617781400680542
    },
    {
      "auxiliary_loss_clip": 0.01154341,
      "auxiliary_loss_mlp": 0.01025515,
      "balance_loss_clip": 1.04534531,
      "balance_loss_mlp": 1.0185951,
      "epoch": 0.7760476161846931,
      "flos": 22526368853760.0,
      "grad_norm": 2.1897326045199863,
      "language_loss": 0.7391476,
      "learning_rate": 5.032964258857993e-07,
      "loss": 0.76094621,
      "num_input_tokens_seen": 138909955,
      "step": 6454,
      "time_per_iteration": 2.610060453414917
    },
    {
      "auxiliary_loss_clip": 0.01148891,
      "auxiliary_loss_mlp": 0.01024288,
      "balance_loss_clip": 1.03994656,
      "balance_loss_mlp": 1.0168792,
      "epoch": 0.7761678590753321,
      "flos": 48651488403840.0,
      "grad_norm": 1.4498786617429609,
      "language_loss": 0.68275189,
      "learning_rate": 5.027798459366329e-07,
      "loss": 0.70448375,
      "num_input_tokens_seen": 138935320,
      "step": 6455,
      "time_per_iteration": 3.817877769470215
    },
    {
      "auxiliary_loss_clip": 0.01155112,
      "auxiliary_loss_mlp": 0.01025822,
      "balance_loss_clip": 1.04394746,
      "balance_loss_mlp": 1.01855659,
      "epoch": 0.7762881019659713,
      "flos": 26177047637760.0,
      "grad_norm": 1.4302560581559935,
      "language_loss": 0.63766849,
      "learning_rate": 5.02263493114573e-07,
      "loss": 0.65947783,
      "num_input_tokens_seen": 138957115,
      "step": 6456,
      "time_per_iteration": 2.5985450744628906
    },
    {
      "auxiliary_loss_clip": 0.0116429,
      "auxiliary_loss_mlp": 0.01024018,
      "balance_loss_clip": 1.04462838,
      "balance_loss_mlp": 1.01693726,
      "epoch": 0.7764083448566104,
      "flos": 20588341518720.0,
      "grad_norm": 2.2602438638820526,
      "language_loss": 0.76780146,
      "learning_rate": 5.017473674979502e-07,
      "loss": 0.78968453,
      "num_input_tokens_seen": 138973140,
      "step": 6457,
      "time_per_iteration": 3.5946390628814697
    },
    {
      "auxiliary_loss_clip": 0.01031843,
      "auxiliary_loss_mlp": 0.01002603,
      "balance_loss_clip": 1.01597202,
      "balance_loss_mlp": 1.00156021,
      "epoch": 0.7765285877472494,
      "flos": 67293078560640.0,
      "grad_norm": 0.752214283597663,
      "language_loss": 0.5831582,
      "learning_rate": 5.01231469165061e-07,
      "loss": 0.60350269,
      "num_input_tokens_seen": 139028965,
      "step": 6458,
      "time_per_iteration": 3.1834661960601807
    },
    {
      "auxiliary_loss_clip": 0.01059607,
      "auxiliary_loss_mlp": 0.0100093,
      "balance_loss_clip": 1.01384521,
      "balance_loss_mlp": 0.99983895,
      "epoch": 0.7766488306378886,
      "flos": 61344476121600.0,
      "grad_norm": 0.822561254599888,
      "language_loss": 0.56862414,
      "learning_rate": 5.007157981941663e-07,
      "loss": 0.58922958,
      "num_input_tokens_seen": 139094325,
      "step": 6459,
      "time_per_iteration": 4.1516313552856445
    },
    {
      "auxiliary_loss_clip": 0.01047194,
      "auxiliary_loss_mlp": 0.01002471,
      "balance_loss_clip": 1.01288843,
      "balance_loss_mlp": 1.00127339,
      "epoch": 0.7767690735285276,
      "flos": 62946199393920.0,
      "grad_norm": 0.8755056117427497,
      "language_loss": 0.67480373,
      "learning_rate": 5.002003546634928e-07,
      "loss": 0.69530046,
      "num_input_tokens_seen": 139150425,
      "step": 6460,
      "time_per_iteration": 3.181318998336792
    },
    {
      "auxiliary_loss_clip": 0.01108705,
      "auxiliary_loss_mlp": 0.01024395,
      "balance_loss_clip": 1.04324031,
      "balance_loss_mlp": 1.01789248,
      "epoch": 0.7768893164191667,
      "flos": 20886400575360.0,
      "grad_norm": 1.7949676875286518,
      "language_loss": 0.76194465,
      "learning_rate": 4.996851386512331e-07,
      "loss": 0.78327572,
      "num_input_tokens_seen": 139169130,
      "step": 6461,
      "time_per_iteration": 2.714665412902832
    },
    {
      "auxiliary_loss_clip": 0.01137453,
      "auxiliary_loss_mlp": 0.01029596,
      "balance_loss_clip": 1.04218972,
      "balance_loss_mlp": 1.02151418,
      "epoch": 0.7770095593098058,
      "flos": 20704584908160.0,
      "grad_norm": 1.7281912233807137,
      "language_loss": 0.83168024,
      "learning_rate": 4.991701502355444e-07,
      "loss": 0.85335076,
      "num_input_tokens_seen": 139189595,
      "step": 6462,
      "time_per_iteration": 2.6480584144592285
    },
    {
      "auxiliary_loss_clip": 0.01155295,
      "auxiliary_loss_mlp": 0.01025194,
      "balance_loss_clip": 1.04463816,
      "balance_loss_mlp": 1.01868773,
      "epoch": 0.7771298022004449,
      "flos": 24717709877760.0,
      "grad_norm": 1.5997942560185956,
      "language_loss": 0.76085317,
      "learning_rate": 4.986553894945518e-07,
      "loss": 0.7826581,
      "num_input_tokens_seen": 139210805,
      "step": 6463,
      "time_per_iteration": 2.6461575031280518
    },
    {
      "auxiliary_loss_clip": 0.01111241,
      "auxiliary_loss_mlp": 0.01024649,
      "balance_loss_clip": 1.03996766,
      "balance_loss_mlp": 1.01824188,
      "epoch": 0.777250045091084,
      "flos": 25009232659200.0,
      "grad_norm": 2.0799850288031463,
      "language_loss": 0.86169779,
      "learning_rate": 4.981408565063416e-07,
      "loss": 0.88305676,
      "num_input_tokens_seen": 139230750,
      "step": 6464,
      "time_per_iteration": 2.814223527908325
    },
    {
      "auxiliary_loss_clip": 0.01168085,
      "auxiliary_loss_mlp": 0.01022604,
      "balance_loss_clip": 1.04642463,
      "balance_loss_mlp": 1.01550794,
      "epoch": 0.777370287981723,
      "flos": 20119887319680.0,
      "grad_norm": 2.0715943766377056,
      "language_loss": 0.75847101,
      "learning_rate": 4.976265513489701e-07,
      "loss": 0.78037786,
      "num_input_tokens_seen": 139250720,
      "step": 6465,
      "time_per_iteration": 2.6299853324890137
    },
    {
      "auxiliary_loss_clip": 0.01150618,
      "auxiliary_loss_mlp": 0.01019793,
      "balance_loss_clip": 1.04261351,
      "balance_loss_mlp": 1.01306105,
      "epoch": 0.7774905308723622,
      "flos": 21718809331200.0,
      "grad_norm": 1.6822820347852592,
      "language_loss": 0.80462503,
      "learning_rate": 4.971124741004562e-07,
      "loss": 0.82632917,
      "num_input_tokens_seen": 139269720,
      "step": 6466,
      "time_per_iteration": 2.635477066040039
    },
    {
      "auxiliary_loss_clip": 0.01152104,
      "auxiliary_loss_mlp": 0.01020257,
      "balance_loss_clip": 1.04461551,
      "balance_loss_mlp": 1.01299095,
      "epoch": 0.7776107737630013,
      "flos": 16034115093120.0,
      "grad_norm": 1.796220729026924,
      "language_loss": 0.76739693,
      "learning_rate": 4.965986248387846e-07,
      "loss": 0.78912055,
      "num_input_tokens_seen": 139288035,
      "step": 6467,
      "time_per_iteration": 2.665466547012329
    },
    {
      "auxiliary_loss_clip": 0.01140994,
      "auxiliary_loss_mlp": 0.01024096,
      "balance_loss_clip": 1.04372537,
      "balance_loss_mlp": 1.01738214,
      "epoch": 0.7777310166536403,
      "flos": 24790895838720.0,
      "grad_norm": 1.6545632798177887,
      "language_loss": 0.7725265,
      "learning_rate": 4.960850036419073e-07,
      "loss": 0.79417741,
      "num_input_tokens_seen": 139307135,
      "step": 6468,
      "time_per_iteration": 2.6672067642211914
    },
    {
      "auxiliary_loss_clip": 0.01132018,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.04175663,
      "balance_loss_mlp": 1.0186379,
      "epoch": 0.7778512595442795,
      "flos": 17272530253440.0,
      "grad_norm": 1.7540391274234641,
      "language_loss": 0.78676045,
      "learning_rate": 4.955716105877378e-07,
      "loss": 0.80833769,
      "num_input_tokens_seen": 139325905,
      "step": 6469,
      "time_per_iteration": 2.7395763397216797
    },
    {
      "auxiliary_loss_clip": 0.01154851,
      "auxiliary_loss_mlp": 0.0076235,
      "balance_loss_clip": 1.04376721,
      "balance_loss_mlp": 1.00029695,
      "epoch": 0.7779715024349185,
      "flos": 17748418567680.0,
      "grad_norm": 1.8017653196162637,
      "language_loss": 0.83470154,
      "learning_rate": 4.950584457541598e-07,
      "loss": 0.85387355,
      "num_input_tokens_seen": 139344370,
      "step": 6470,
      "time_per_iteration": 2.6993916034698486
    },
    {
      "auxiliary_loss_clip": 0.01155753,
      "auxiliary_loss_mlp": 0.01025722,
      "balance_loss_clip": 1.04485857,
      "balance_loss_mlp": 1.01876044,
      "epoch": 0.7780917453255576,
      "flos": 24316875031680.0,
      "grad_norm": 1.4160150360587893,
      "language_loss": 0.82013917,
      "learning_rate": 4.945455092190183e-07,
      "loss": 0.84195393,
      "num_input_tokens_seen": 139365625,
      "step": 6471,
      "time_per_iteration": 2.6903605461120605
    },
    {
      "auxiliary_loss_clip": 0.01067996,
      "auxiliary_loss_mlp": 0.01001653,
      "balance_loss_clip": 1.01324511,
      "balance_loss_mlp": 1.00057971,
      "epoch": 0.7782119882161967,
      "flos": 56364601530240.0,
      "grad_norm": 0.6856111128833383,
      "language_loss": 0.55956912,
      "learning_rate": 4.940328010601271e-07,
      "loss": 0.58026558,
      "num_input_tokens_seen": 139430540,
      "step": 6472,
      "time_per_iteration": 3.1728127002716064
    },
    {
      "auxiliary_loss_clip": 0.01148671,
      "auxiliary_loss_mlp": 0.01029824,
      "balance_loss_clip": 1.04724932,
      "balance_loss_mlp": 1.02228153,
      "epoch": 0.7783322311068358,
      "flos": 46789986994560.0,
      "grad_norm": 1.7836299176374093,
      "language_loss": 0.76575935,
      "learning_rate": 4.935203213552621e-07,
      "loss": 0.78754431,
      "num_input_tokens_seen": 139454280,
      "step": 6473,
      "time_per_iteration": 2.8815653324127197
    },
    {
      "auxiliary_loss_clip": 0.01139833,
      "auxiliary_loss_mlp": 0.01027316,
      "balance_loss_clip": 1.04394925,
      "balance_loss_mlp": 1.02044106,
      "epoch": 0.7784524739974749,
      "flos": 19057864872960.0,
      "grad_norm": 2.2381279709341255,
      "language_loss": 0.6688993,
      "learning_rate": 4.930080701821662e-07,
      "loss": 0.69057077,
      "num_input_tokens_seen": 139471745,
      "step": 6474,
      "time_per_iteration": 2.6783008575439453
    },
    {
      "auxiliary_loss_clip": 0.01140352,
      "auxiliary_loss_mlp": 0.01023055,
      "balance_loss_clip": 1.04386652,
      "balance_loss_mlp": 1.01621211,
      "epoch": 0.778572716888114,
      "flos": 24791111320320.0,
      "grad_norm": 6.263478103286799,
      "language_loss": 0.77012241,
      "learning_rate": 4.92496047618548e-07,
      "loss": 0.79175651,
      "num_input_tokens_seen": 139491505,
      "step": 6475,
      "time_per_iteration": 2.72338604927063
    },
    {
      "auxiliary_loss_clip": 0.01155148,
      "auxiliary_loss_mlp": 0.01023402,
      "balance_loss_clip": 1.04690385,
      "balance_loss_mlp": 1.01625812,
      "epoch": 0.7786929597787531,
      "flos": 20078086867200.0,
      "grad_norm": 1.9834186106298857,
      "language_loss": 0.78220427,
      "learning_rate": 4.919842537420811e-07,
      "loss": 0.80398977,
      "num_input_tokens_seen": 139508620,
      "step": 6476,
      "time_per_iteration": 2.637674570083618
    },
    {
      "auxiliary_loss_clip": 0.01139471,
      "auxiliary_loss_mlp": 0.0102583,
      "balance_loss_clip": 1.04591179,
      "balance_loss_mlp": 1.01898694,
      "epoch": 0.7788132026693921,
      "flos": 21872220318720.0,
      "grad_norm": 1.6878529302025571,
      "language_loss": 0.79502815,
      "learning_rate": 4.91472688630404e-07,
      "loss": 0.81668115,
      "num_input_tokens_seen": 139529360,
      "step": 6477,
      "time_per_iteration": 2.6739304065704346
    },
    {
      "auxiliary_loss_clip": 0.01163541,
      "auxiliary_loss_mlp": 0.01024472,
      "balance_loss_clip": 1.04494286,
      "balance_loss_mlp": 1.01751947,
      "epoch": 0.7789334455600313,
      "flos": 11181937351680.0,
      "grad_norm": 1.8480577570858563,
      "language_loss": 0.74438775,
      "learning_rate": 4.909613523611202e-07,
      "loss": 0.7662679,
      "num_input_tokens_seen": 139546240,
      "step": 6478,
      "time_per_iteration": 2.6428725719451904
    },
    {
      "auxiliary_loss_clip": 0.01107386,
      "auxiliary_loss_mlp": 0.00762858,
      "balance_loss_clip": 1.03770041,
      "balance_loss_mlp": 1.00032306,
      "epoch": 0.7790536884506704,
      "flos": 28695427015680.0,
      "grad_norm": 1.7562949178445104,
      "language_loss": 0.74715012,
      "learning_rate": 4.904502450117991e-07,
      "loss": 0.76585257,
      "num_input_tokens_seen": 139567200,
      "step": 6479,
      "time_per_iteration": 3.8457181453704834
    },
    {
      "auxiliary_loss_clip": 0.01135774,
      "auxiliary_loss_mlp": 0.01028014,
      "balance_loss_clip": 1.04673386,
      "balance_loss_mlp": 1.02095652,
      "epoch": 0.7791739313413094,
      "flos": 11072302064640.0,
      "grad_norm": 2.455798611845857,
      "language_loss": 0.7288183,
      "learning_rate": 4.899393666599762e-07,
      "loss": 0.75045609,
      "num_input_tokens_seen": 139583775,
      "step": 6480,
      "time_per_iteration": 2.7839150428771973
    },
    {
      "auxiliary_loss_clip": 0.0116489,
      "auxiliary_loss_mlp": 0.01030578,
      "balance_loss_clip": 1.04376745,
      "balance_loss_mlp": 1.02408743,
      "epoch": 0.7792941742319486,
      "flos": 14679276975360.0,
      "grad_norm": 2.2856088138509976,
      "language_loss": 0.72678024,
      "learning_rate": 4.894287173831506e-07,
      "loss": 0.74873495,
      "num_input_tokens_seen": 139599735,
      "step": 6481,
      "time_per_iteration": 3.635589599609375
    },
    {
      "auxiliary_loss_clip": 0.01140352,
      "auxiliary_loss_mlp": 0.01029414,
      "balance_loss_clip": 1.04149628,
      "balance_loss_mlp": 1.02192187,
      "epoch": 0.7794144171225876,
      "flos": 23258874908160.0,
      "grad_norm": 1.9367567173263704,
      "language_loss": 0.84623909,
      "learning_rate": 4.889182972587877e-07,
      "loss": 0.86793679,
      "num_input_tokens_seen": 139619030,
      "step": 6482,
      "time_per_iteration": 2.7233073711395264
    },
    {
      "auxiliary_loss_clip": 0.01134007,
      "auxiliary_loss_mlp": 0.01025289,
      "balance_loss_clip": 1.04406214,
      "balance_loss_mlp": 1.01844668,
      "epoch": 0.7795346600132267,
      "flos": 21507080613120.0,
      "grad_norm": 1.9498208743933556,
      "language_loss": 0.66406345,
      "learning_rate": 4.884081063643177e-07,
      "loss": 0.68565637,
      "num_input_tokens_seen": 139637690,
      "step": 6483,
      "time_per_iteration": 3.57716965675354
    },
    {
      "auxiliary_loss_clip": 0.01038897,
      "auxiliary_loss_mlp": 0.01002983,
      "balance_loss_clip": 1.01105428,
      "balance_loss_mlp": 1.00189805,
      "epoch": 0.7796549029038659,
      "flos": 70052273694720.0,
      "grad_norm": 0.8726391741822886,
      "language_loss": 0.52546024,
      "learning_rate": 4.878981447771353e-07,
      "loss": 0.54587901,
      "num_input_tokens_seen": 139692070,
      "step": 6484,
      "time_per_iteration": 3.2176403999328613
    },
    {
      "auxiliary_loss_clip": 0.01118769,
      "auxiliary_loss_mlp": 0.01025413,
      "balance_loss_clip": 1.04198575,
      "balance_loss_mlp": 1.01792932,
      "epoch": 0.7797751457945049,
      "flos": 23989405714560.0,
      "grad_norm": 1.618157020421602,
      "language_loss": 0.73182797,
      "learning_rate": 4.873884125746035e-07,
      "loss": 0.75326979,
      "num_input_tokens_seen": 139713745,
      "step": 6485,
      "time_per_iteration": 3.6669607162475586
    },
    {
      "auxiliary_loss_clip": 0.01133003,
      "auxiliary_loss_mlp": 0.01025056,
      "balance_loss_clip": 1.0414927,
      "balance_loss_mlp": 1.0176053,
      "epoch": 0.779895388685144,
      "flos": 22674751937280.0,
      "grad_norm": 2.3527004457314376,
      "language_loss": 0.72136891,
      "learning_rate": 4.868789098340456e-07,
      "loss": 0.74294949,
      "num_input_tokens_seen": 139731650,
      "step": 6486,
      "time_per_iteration": 2.82139253616333
    },
    {
      "auxiliary_loss_clip": 0.01126032,
      "auxiliary_loss_mlp": 0.01023583,
      "balance_loss_clip": 1.04270244,
      "balance_loss_mlp": 1.0164957,
      "epoch": 0.7800156315757831,
      "flos": 23768698596480.0,
      "grad_norm": 2.394968700562239,
      "language_loss": 0.72802091,
      "learning_rate": 4.863696366327543e-07,
      "loss": 0.74951708,
      "num_input_tokens_seen": 139750820,
      "step": 6487,
      "time_per_iteration": 2.697099447250366
    },
    {
      "auxiliary_loss_clip": 0.01152109,
      "auxiliary_loss_mlp": 0.01020556,
      "balance_loss_clip": 1.04224277,
      "balance_loss_mlp": 1.01325476,
      "epoch": 0.7801358744664222,
      "flos": 26429714881920.0,
      "grad_norm": 1.6723277964858903,
      "language_loss": 0.77978295,
      "learning_rate": 4.85860593047986e-07,
      "loss": 0.80150962,
      "num_input_tokens_seen": 139770885,
      "step": 6488,
      "time_per_iteration": 2.652677059173584
    },
    {
      "auxiliary_loss_clip": 0.01116286,
      "auxiliary_loss_mlp": 0.01025155,
      "balance_loss_clip": 1.03719664,
      "balance_loss_mlp": 1.01813996,
      "epoch": 0.7802561173570612,
      "flos": 26322162583680.0,
      "grad_norm": 1.5805377911966227,
      "language_loss": 0.74991924,
      "learning_rate": 4.853517791569613e-07,
      "loss": 0.77133363,
      "num_input_tokens_seen": 139793065,
      "step": 6489,
      "time_per_iteration": 2.7648510932922363
    },
    {
      "auxiliary_loss_clip": 0.01141426,
      "auxiliary_loss_mlp": 0.00762949,
      "balance_loss_clip": 1.04247558,
      "balance_loss_mlp": 1.00031161,
      "epoch": 0.7803763602477004,
      "flos": 40333751596800.0,
      "grad_norm": 1.775369385709843,
      "language_loss": 0.66525555,
      "learning_rate": 4.848431950368684e-07,
      "loss": 0.68429923,
      "num_input_tokens_seen": 139815625,
      "step": 6490,
      "time_per_iteration": 2.832223653793335
    },
    {
      "auxiliary_loss_clip": 0.01067782,
      "auxiliary_loss_mlp": 0.00753583,
      "balance_loss_clip": 1.01309848,
      "balance_loss_mlp": 1.00021553,
      "epoch": 0.7804966031383395,
      "flos": 67001448038400.0,
      "grad_norm": 0.7416839678124272,
      "language_loss": 0.55685437,
      "learning_rate": 4.843348407648569e-07,
      "loss": 0.575068,
      "num_input_tokens_seen": 139876905,
      "step": 6491,
      "time_per_iteration": 3.132840633392334
    },
    {
      "auxiliary_loss_clip": 0.01154032,
      "auxiliary_loss_mlp": 0.01024296,
      "balance_loss_clip": 1.04165101,
      "balance_loss_mlp": 1.01688743,
      "epoch": 0.7806168460289785,
      "flos": 17740733057280.0,
      "grad_norm": 2.3123721522641643,
      "language_loss": 0.82773179,
      "learning_rate": 4.838267164180457e-07,
      "loss": 0.84951508,
      "num_input_tokens_seen": 139892575,
      "step": 6492,
      "time_per_iteration": 2.595829963684082
    },
    {
      "auxiliary_loss_clip": 0.0116886,
      "auxiliary_loss_mlp": 0.01026919,
      "balance_loss_clip": 1.0465076,
      "balance_loss_mlp": 1.01954639,
      "epoch": 0.7807370889196176,
      "flos": 23946240545280.0,
      "grad_norm": 1.979848602662472,
      "language_loss": 0.83643359,
      "learning_rate": 4.833188220735156e-07,
      "loss": 0.8583914,
      "num_input_tokens_seen": 139912245,
      "step": 6493,
      "time_per_iteration": 2.6299400329589844
    },
    {
      "auxiliary_loss_clip": 0.01148356,
      "auxiliary_loss_mlp": 0.01025825,
      "balance_loss_clip": 1.04276729,
      "balance_loss_mlp": 1.01899457,
      "epoch": 0.7808573318102567,
      "flos": 18989024457600.0,
      "grad_norm": 2.2131090536186164,
      "language_loss": 0.7434932,
      "learning_rate": 4.828111578083152e-07,
      "loss": 0.76523507,
      "num_input_tokens_seen": 139929150,
      "step": 6494,
      "time_per_iteration": 2.635835886001587
    },
    {
      "auxiliary_loss_clip": 0.01138668,
      "auxiliary_loss_mlp": 0.0102479,
      "balance_loss_clip": 1.04576337,
      "balance_loss_mlp": 1.01814365,
      "epoch": 0.7809775747008958,
      "flos": 23980750536960.0,
      "grad_norm": 2.005280747149861,
      "language_loss": 0.80942309,
      "learning_rate": 4.823037236994556e-07,
      "loss": 0.83105767,
      "num_input_tokens_seen": 139947315,
      "step": 6495,
      "time_per_iteration": 2.706821918487549
    },
    {
      "auxiliary_loss_clip": 0.01058834,
      "auxiliary_loss_mlp": 0.0100208,
      "balance_loss_clip": 1.01273489,
      "balance_loss_mlp": 1.00101936,
      "epoch": 0.7810978175915348,
      "flos": 68535875180160.0,
      "grad_norm": 0.7164227959020683,
      "language_loss": 0.56315351,
      "learning_rate": 4.817965198239136e-07,
      "loss": 0.58376265,
      "num_input_tokens_seen": 140013775,
      "step": 6496,
      "time_per_iteration": 3.2237048149108887
    },
    {
      "auxiliary_loss_clip": 0.01122918,
      "auxiliary_loss_mlp": 0.01027089,
      "balance_loss_clip": 1.04022777,
      "balance_loss_mlp": 1.02027369,
      "epoch": 0.781218060482174,
      "flos": 19642131498240.0,
      "grad_norm": 2.004018700319703,
      "language_loss": 0.74465728,
      "learning_rate": 4.812895462586331e-07,
      "loss": 0.76615733,
      "num_input_tokens_seen": 140031600,
      "step": 6497,
      "time_per_iteration": 2.754545211791992
    },
    {
      "auxiliary_loss_clip": 0.01128106,
      "auxiliary_loss_mlp": 0.01025823,
      "balance_loss_clip": 1.04246426,
      "balance_loss_mlp": 1.01907921,
      "epoch": 0.7813383033728131,
      "flos": 25627865621760.0,
      "grad_norm": 2.161221674182993,
      "language_loss": 0.82125139,
      "learning_rate": 4.807828030805207e-07,
      "loss": 0.84279072,
      "num_input_tokens_seen": 140050590,
      "step": 6498,
      "time_per_iteration": 2.741144895553589
    },
    {
      "auxiliary_loss_clip": 0.0115266,
      "auxiliary_loss_mlp": 0.01026925,
      "balance_loss_clip": 1.04734945,
      "balance_loss_mlp": 1.01968932,
      "epoch": 0.7814585462634521,
      "flos": 20485924865280.0,
      "grad_norm": 2.7148601169117326,
      "language_loss": 0.67761749,
      "learning_rate": 4.802762903664495e-07,
      "loss": 0.69941336,
      "num_input_tokens_seen": 140069770,
      "step": 6499,
      "time_per_iteration": 2.628065586090088
    },
    {
      "auxiliary_loss_clip": 0.01141958,
      "auxiliary_loss_mlp": 0.01024596,
      "balance_loss_clip": 1.04427123,
      "balance_loss_mlp": 1.0172056,
      "epoch": 0.7815787891540913,
      "flos": 22304297018880.0,
      "grad_norm": 2.4614267420154814,
      "language_loss": 0.73741066,
      "learning_rate": 4.797700081932565e-07,
      "loss": 0.75907624,
      "num_input_tokens_seen": 140087635,
      "step": 6500,
      "time_per_iteration": 2.685960292816162
    },
    {
      "auxiliary_loss_clip": 0.0109231,
      "auxiliary_loss_mlp": 0.01029253,
      "balance_loss_clip": 1.03771329,
      "balance_loss_mlp": 1.02210057,
      "epoch": 0.7816990320447303,
      "flos": 22600668136320.0,
      "grad_norm": 2.7368271045160637,
      "language_loss": 0.8148374,
      "learning_rate": 4.792639566377442e-07,
      "loss": 0.83605301,
      "num_input_tokens_seen": 140105045,
      "step": 6501,
      "time_per_iteration": 2.7704479694366455
    },
    {
      "auxiliary_loss_clip": 0.01144911,
      "auxiliary_loss_mlp": 0.01025459,
      "balance_loss_clip": 1.0402112,
      "balance_loss_mlp": 1.01784742,
      "epoch": 0.7818192749353694,
      "flos": 24935974871040.0,
      "grad_norm": 1.6336279600859456,
      "language_loss": 0.77348495,
      "learning_rate": 4.78758135776681e-07,
      "loss": 0.79518861,
      "num_input_tokens_seen": 140124900,
      "step": 6502,
      "time_per_iteration": 2.713123321533203
    },
    {
      "auxiliary_loss_clip": 0.01143787,
      "auxiliary_loss_mlp": 0.01026481,
      "balance_loss_clip": 1.04750299,
      "balance_loss_mlp": 1.01901817,
      "epoch": 0.7819395178260086,
      "flos": 23733039369600.0,
      "grad_norm": 2.137510511301502,
      "language_loss": 0.78671944,
      "learning_rate": 4.782525456867989e-07,
      "loss": 0.80842209,
      "num_input_tokens_seen": 140143755,
      "step": 6503,
      "time_per_iteration": 2.7096970081329346
    },
    {
      "auxiliary_loss_clip": 0.01128704,
      "auxiliary_loss_mlp": 0.01025174,
      "balance_loss_clip": 1.04381442,
      "balance_loss_mlp": 1.01729465,
      "epoch": 0.7820597607166476,
      "flos": 23221671396480.0,
      "grad_norm": 1.87347471338828,
      "language_loss": 0.83443689,
      "learning_rate": 4.777471864447959e-07,
      "loss": 0.85597569,
      "num_input_tokens_seen": 140164495,
      "step": 6504,
      "time_per_iteration": 2.7644662857055664
    },
    {
      "auxiliary_loss_clip": 0.01141083,
      "auxiliary_loss_mlp": 0.01026369,
      "balance_loss_clip": 1.04327416,
      "balance_loss_mlp": 1.01905584,
      "epoch": 0.7821800036072867,
      "flos": 22309540404480.0,
      "grad_norm": 2.0412351970810523,
      "language_loss": 0.80360079,
      "learning_rate": 4.772420581273344e-07,
      "loss": 0.8252753,
      "num_input_tokens_seen": 140181980,
      "step": 6505,
      "time_per_iteration": 3.649690866470337
    },
    {
      "auxiliary_loss_clip": 0.01149532,
      "auxiliary_loss_mlp": 0.01022908,
      "balance_loss_clip": 1.0452559,
      "balance_loss_mlp": 1.01576114,
      "epoch": 0.7823002464979258,
      "flos": 21544176384000.0,
      "grad_norm": 2.02620607906541,
      "language_loss": 0.7657876,
      "learning_rate": 4.7673716081104134e-07,
      "loss": 0.78751194,
      "num_input_tokens_seen": 140202155,
      "step": 6506,
      "time_per_iteration": 2.739870309829712
    },
    {
      "auxiliary_loss_clip": 0.01151386,
      "auxiliary_loss_mlp": 0.01024134,
      "balance_loss_clip": 1.04672098,
      "balance_loss_mlp": 1.01754165,
      "epoch": 0.7824204893885649,
      "flos": 24535642815360.0,
      "grad_norm": 1.7339916883145055,
      "language_loss": 0.84749007,
      "learning_rate": 4.762324945725109e-07,
      "loss": 0.86924529,
      "num_input_tokens_seen": 140221600,
      "step": 6507,
      "time_per_iteration": 3.708195447921753
    },
    {
      "auxiliary_loss_clip": 0.01136269,
      "auxiliary_loss_mlp": 0.01028393,
      "balance_loss_clip": 1.04586339,
      "balance_loss_mlp": 1.02120435,
      "epoch": 0.782540732279204,
      "flos": 27415211402880.0,
      "grad_norm": 1.6707353165861734,
      "language_loss": 0.75711691,
      "learning_rate": 4.7572805948829844e-07,
      "loss": 0.77876353,
      "num_input_tokens_seen": 140241860,
      "step": 6508,
      "time_per_iteration": 3.6458725929260254
    },
    {
      "auxiliary_loss_clip": 0.01115593,
      "auxiliary_loss_mlp": 0.01022723,
      "balance_loss_clip": 1.04082417,
      "balance_loss_mlp": 1.01644051,
      "epoch": 0.7826609751698431,
      "flos": 24353216616960.0,
      "grad_norm": 1.7284796602822707,
      "language_loss": 0.70950717,
      "learning_rate": 4.7522385563492795e-07,
      "loss": 0.73089033,
      "num_input_tokens_seen": 140262160,
      "step": 6509,
      "time_per_iteration": 2.7927610874176025
    },
    {
      "auxiliary_loss_clip": 0.01129312,
      "auxiliary_loss_mlp": 0.01027456,
      "balance_loss_clip": 1.04428709,
      "balance_loss_mlp": 1.02054191,
      "epoch": 0.7827812180604822,
      "flos": 23988543788160.0,
      "grad_norm": 2.131050588258103,
      "language_loss": 0.70212454,
      "learning_rate": 4.747198830888863e-07,
      "loss": 0.72369218,
      "num_input_tokens_seen": 140282030,
      "step": 6510,
      "time_per_iteration": 2.6927080154418945
    },
    {
      "auxiliary_loss_clip": 0.01134704,
      "auxiliary_loss_mlp": 0.01032845,
      "balance_loss_clip": 1.04248953,
      "balance_loss_mlp": 1.02527809,
      "epoch": 0.7829014609511212,
      "flos": 27454318335360.0,
      "grad_norm": 2.4557858302003663,
      "language_loss": 0.68832856,
      "learning_rate": 4.742161419266251e-07,
      "loss": 0.71000403,
      "num_input_tokens_seen": 140301190,
      "step": 6511,
      "time_per_iteration": 3.5797083377838135
    },
    {
      "auxiliary_loss_clip": 0.01156082,
      "auxiliary_loss_mlp": 0.01025296,
      "balance_loss_clip": 1.04555655,
      "balance_loss_mlp": 1.01780701,
      "epoch": 0.7830217038417604,
      "flos": 29204532432000.0,
      "grad_norm": 2.362803190038171,
      "language_loss": 0.65238452,
      "learning_rate": 4.7371263222456304e-07,
      "loss": 0.67419827,
      "num_input_tokens_seen": 140318510,
      "step": 6512,
      "time_per_iteration": 2.850525379180908
    },
    {
      "auxiliary_loss_clip": 0.01052182,
      "auxiliary_loss_mlp": 0.01002249,
      "balance_loss_clip": 1.01057458,
      "balance_loss_mlp": 1.00121212,
      "epoch": 0.7831419467323995,
      "flos": 60950895822720.0,
      "grad_norm": 0.8035402461808099,
      "language_loss": 0.61378908,
      "learning_rate": 4.7320935405908004e-07,
      "loss": 0.63433337,
      "num_input_tokens_seen": 140379380,
      "step": 6513,
      "time_per_iteration": 3.212474822998047
    },
    {
      "auxiliary_loss_clip": 0.01169045,
      "auxiliary_loss_mlp": 0.01022777,
      "balance_loss_clip": 1.04531562,
      "balance_loss_mlp": 1.01534402,
      "epoch": 0.7832621896230385,
      "flos": 19682531320320.0,
      "grad_norm": 2.646457061383335,
      "language_loss": 0.84070718,
      "learning_rate": 4.7270630750652475e-07,
      "loss": 0.86262548,
      "num_input_tokens_seen": 140395335,
      "step": 6514,
      "time_per_iteration": 2.5818753242492676
    },
    {
      "auxiliary_loss_clip": 0.01148263,
      "auxiliary_loss_mlp": 0.01024398,
      "balance_loss_clip": 1.04314542,
      "balance_loss_mlp": 1.01729023,
      "epoch": 0.7833824325136777,
      "flos": 25009232659200.0,
      "grad_norm": 1.7583176131906804,
      "language_loss": 0.80288374,
      "learning_rate": 4.7220349264320746e-07,
      "loss": 0.82461035,
      "num_input_tokens_seen": 140414420,
      "step": 6515,
      "time_per_iteration": 2.736367702484131
    },
    {
      "auxiliary_loss_clip": 0.01056253,
      "auxiliary_loss_mlp": 0.01002319,
      "balance_loss_clip": 1.01291633,
      "balance_loss_mlp": 1.00118685,
      "epoch": 0.7835026754043167,
      "flos": 68800142517120.0,
      "grad_norm": 0.7381798755897361,
      "language_loss": 0.54850918,
      "learning_rate": 4.71700909545407e-07,
      "loss": 0.5690949,
      "num_input_tokens_seen": 140477365,
      "step": 6516,
      "time_per_iteration": 3.216501235961914
    },
    {
      "auxiliary_loss_clip": 0.01155581,
      "auxiliary_loss_mlp": 0.01023199,
      "balance_loss_clip": 1.04490209,
      "balance_loss_mlp": 1.01616311,
      "epoch": 0.7836229182949558,
      "flos": 19864598382720.0,
      "grad_norm": 1.7435618883171538,
      "language_loss": 0.76963866,
      "learning_rate": 4.711985582893627e-07,
      "loss": 0.79142648,
      "num_input_tokens_seen": 140495885,
      "step": 6517,
      "time_per_iteration": 2.604694128036499
    },
    {
      "auxiliary_loss_clip": 0.01113355,
      "auxiliary_loss_mlp": 0.01022746,
      "balance_loss_clip": 1.03974986,
      "balance_loss_mlp": 1.01538157,
      "epoch": 0.783743161185595,
      "flos": 22965843755520.0,
      "grad_norm": 1.702977391361563,
      "language_loss": 0.71579391,
      "learning_rate": 4.706964389512811e-07,
      "loss": 0.73715496,
      "num_input_tokens_seen": 140515920,
      "step": 6518,
      "time_per_iteration": 2.7204086780548096
    },
    {
      "auxiliary_loss_clip": 0.01168441,
      "auxiliary_loss_mlp": 0.01029594,
      "balance_loss_clip": 1.04993188,
      "balance_loss_mlp": 1.02176809,
      "epoch": 0.783863404076234,
      "flos": 12458489777280.0,
      "grad_norm": 1.929812012400059,
      "language_loss": 0.87469423,
      "learning_rate": 4.701945516073345e-07,
      "loss": 0.89667463,
      "num_input_tokens_seen": 140533395,
      "step": 6519,
      "time_per_iteration": 2.597367525100708
    },
    {
      "auxiliary_loss_clip": 0.01121524,
      "auxiliary_loss_mlp": 0.01024113,
      "balance_loss_clip": 1.04178286,
      "balance_loss_mlp": 1.01675785,
      "epoch": 0.7839836469668731,
      "flos": 24243940465920.0,
      "grad_norm": 1.937925012022678,
      "language_loss": 0.75177819,
      "learning_rate": 4.696928963336577e-07,
      "loss": 0.77323449,
      "num_input_tokens_seen": 140552825,
      "step": 6520,
      "time_per_iteration": 2.6953039169311523
    },
    {
      "auxiliary_loss_clip": 0.01052418,
      "auxiliary_loss_mlp": 0.01002183,
      "balance_loss_clip": 1.01071143,
      "balance_loss_mlp": 1.00112247,
      "epoch": 0.7841038898575122,
      "flos": 62121978938880.0,
      "grad_norm": 0.8514675553265723,
      "language_loss": 0.61002588,
      "learning_rate": 4.6919147320635224e-07,
      "loss": 0.63057184,
      "num_input_tokens_seen": 140615535,
      "step": 6521,
      "time_per_iteration": 3.2307302951812744
    },
    {
      "auxiliary_loss_clip": 0.01153539,
      "auxiliary_loss_mlp": 0.01030485,
      "balance_loss_clip": 1.04268312,
      "balance_loss_mlp": 1.02346635,
      "epoch": 0.7842241327481513,
      "flos": 20193899293440.0,
      "grad_norm": 2.3458921208731462,
      "language_loss": 0.72891557,
      "learning_rate": 4.6869028230148286e-07,
      "loss": 0.75075579,
      "num_input_tokens_seen": 140633330,
      "step": 6522,
      "time_per_iteration": 2.634263038635254
    },
    {
      "auxiliary_loss_clip": 0.01117245,
      "auxiliary_loss_mlp": 0.01025528,
      "balance_loss_clip": 1.03807116,
      "balance_loss_mlp": 1.01718295,
      "epoch": 0.7843443756387903,
      "flos": 28074531496320.0,
      "grad_norm": 2.188934808958748,
      "language_loss": 0.59796172,
      "learning_rate": 4.6818932369507957e-07,
      "loss": 0.61938941,
      "num_input_tokens_seen": 140652830,
      "step": 6523,
      "time_per_iteration": 2.81190824508667
    },
    {
      "auxiliary_loss_clip": 0.01154284,
      "auxiliary_loss_mlp": 0.01027161,
      "balance_loss_clip": 1.04728448,
      "balance_loss_mlp": 1.02010727,
      "epoch": 0.7844646185294295,
      "flos": 21323397438720.0,
      "grad_norm": 2.2289330118390343,
      "language_loss": 0.89438963,
      "learning_rate": 4.676885974631386e-07,
      "loss": 0.91620398,
      "num_input_tokens_seen": 140671190,
      "step": 6524,
      "time_per_iteration": 2.6719560623168945
    },
    {
      "auxiliary_loss_clip": 0.01150863,
      "auxiliary_loss_mlp": 0.01028565,
      "balance_loss_clip": 1.04421163,
      "balance_loss_mlp": 1.02145684,
      "epoch": 0.7845848614200686,
      "flos": 23656585271040.0,
      "grad_norm": 1.9566310791697055,
      "language_loss": 0.81413531,
      "learning_rate": 4.67188103681619e-07,
      "loss": 0.83592963,
      "num_input_tokens_seen": 140690975,
      "step": 6525,
      "time_per_iteration": 2.6624062061309814
    },
    {
      "auxiliary_loss_clip": 0.01148986,
      "auxiliary_loss_mlp": 0.00762025,
      "balance_loss_clip": 1.04680371,
      "balance_loss_mlp": 1.00033152,
      "epoch": 0.7847051043107076,
      "flos": 23402194174080.0,
      "grad_norm": 2.05880972361832,
      "language_loss": 0.69340259,
      "learning_rate": 4.666878424264453e-07,
      "loss": 0.71251267,
      "num_input_tokens_seen": 140710930,
      "step": 6526,
      "time_per_iteration": 2.6948983669281006
    },
    {
      "auxiliary_loss_clip": 0.01130694,
      "auxiliary_loss_mlp": 0.01022389,
      "balance_loss_clip": 1.04281449,
      "balance_loss_mlp": 1.01589537,
      "epoch": 0.7848253472013467,
      "flos": 19022277473280.0,
      "grad_norm": 1.5703116274649527,
      "language_loss": 0.73804486,
      "learning_rate": 4.661878137735069e-07,
      "loss": 0.75957566,
      "num_input_tokens_seen": 140729120,
      "step": 6527,
      "time_per_iteration": 2.6417665481567383
    },
    {
      "auxiliary_loss_clip": 0.011386,
      "auxiliary_loss_mlp": 0.01024925,
      "balance_loss_clip": 1.044608,
      "balance_loss_mlp": 1.01849961,
      "epoch": 0.7849455900919858,
      "flos": 21179180332800.0,
      "grad_norm": 1.8874763645142707,
      "language_loss": 0.75004268,
      "learning_rate": 4.656880177986571e-07,
      "loss": 0.77167785,
      "num_input_tokens_seen": 140747665,
      "step": 6528,
      "time_per_iteration": 2.7247180938720703
    },
    {
      "auxiliary_loss_clip": 0.01140943,
      "auxiliary_loss_mlp": 0.01031669,
      "balance_loss_clip": 1.04186249,
      "balance_loss_mlp": 1.02439106,
      "epoch": 0.7850658329826249,
      "flos": 19536482620800.0,
      "grad_norm": 1.8036284286668014,
      "language_loss": 0.8152957,
      "learning_rate": 4.6518845457771607e-07,
      "loss": 0.83702177,
      "num_input_tokens_seen": 140766525,
      "step": 6529,
      "time_per_iteration": 2.621004819869995
    },
    {
      "auxiliary_loss_clip": 0.01146764,
      "auxiliary_loss_mlp": 0.00761985,
      "balance_loss_clip": 1.04415882,
      "balance_loss_mlp": 1.00033748,
      "epoch": 0.7851860758732639,
      "flos": 12495334152960.0,
      "grad_norm": 1.8650112955659495,
      "language_loss": 0.79178798,
      "learning_rate": 4.646891241864652e-07,
      "loss": 0.81087542,
      "num_input_tokens_seen": 140785090,
      "step": 6530,
      "time_per_iteration": 2.6424753665924072
    },
    {
      "auxiliary_loss_clip": 0.0114799,
      "auxiliary_loss_mlp": 0.01029023,
      "balance_loss_clip": 1.04147577,
      "balance_loss_mlp": 1.02162933,
      "epoch": 0.7853063187639031,
      "flos": 22960959505920.0,
      "grad_norm": 2.2584725423911847,
      "language_loss": 0.73293316,
      "learning_rate": 4.6419002670065397e-07,
      "loss": 0.75470328,
      "num_input_tokens_seen": 140804670,
      "step": 6531,
      "time_per_iteration": 3.6408517360687256
    },
    {
      "auxiliary_loss_clip": 0.01128906,
      "auxiliary_loss_mlp": 0.01021468,
      "balance_loss_clip": 1.04347396,
      "balance_loss_mlp": 1.01436865,
      "epoch": 0.7854265616545422,
      "flos": 17347260499200.0,
      "grad_norm": 2.0470471342488112,
      "language_loss": 0.87063831,
      "learning_rate": 4.6369116219599445e-07,
      "loss": 0.89214206,
      "num_input_tokens_seen": 140820655,
      "step": 6532,
      "time_per_iteration": 2.634772777557373
    },
    {
      "auxiliary_loss_clip": 0.01123772,
      "auxiliary_loss_mlp": 0.01025043,
      "balance_loss_clip": 1.04215527,
      "balance_loss_mlp": 1.01783085,
      "epoch": 0.7855468045451812,
      "flos": 23838293197440.0,
      "grad_norm": 1.8346912960306412,
      "language_loss": 0.7917468,
      "learning_rate": 4.631925307481637e-07,
      "loss": 0.81323493,
      "num_input_tokens_seen": 140840470,
      "step": 6533,
      "time_per_iteration": 3.7489569187164307
    },
    {
      "auxiliary_loss_clip": 0.01137624,
      "auxiliary_loss_mlp": 0.0102293,
      "balance_loss_clip": 1.04461479,
      "balance_loss_mlp": 1.01648378,
      "epoch": 0.7856670474358204,
      "flos": 25666792986240.0,
      "grad_norm": 2.351012759212642,
      "language_loss": 0.7571286,
      "learning_rate": 4.6269413243280533e-07,
      "loss": 0.77873409,
      "num_input_tokens_seen": 140859890,
      "step": 6534,
      "time_per_iteration": 3.6823599338531494
    },
    {
      "auxiliary_loss_clip": 0.01142354,
      "auxiliary_loss_mlp": 0.01025216,
      "balance_loss_clip": 1.04485023,
      "balance_loss_mlp": 1.01776004,
      "epoch": 0.7857872903264594,
      "flos": 18144656472960.0,
      "grad_norm": 2.643578500758141,
      "language_loss": 0.74123603,
      "learning_rate": 4.621959673255236e-07,
      "loss": 0.76291174,
      "num_input_tokens_seen": 140876190,
      "step": 6535,
      "time_per_iteration": 2.740847110748291
    },
    {
      "auxiliary_loss_clip": 0.01107053,
      "auxiliary_loss_mlp": 0.01023079,
      "balance_loss_clip": 1.03989673,
      "balance_loss_mlp": 1.015944,
      "epoch": 0.7859075332170985,
      "flos": 14386138081920.0,
      "grad_norm": 1.9743967525569641,
      "language_loss": 0.90623361,
      "learning_rate": 4.6169803550189135e-07,
      "loss": 0.92753488,
      "num_input_tokens_seen": 140891885,
      "step": 6536,
      "time_per_iteration": 2.7829766273498535
    },
    {
      "auxiliary_loss_clip": 0.01103936,
      "auxiliary_loss_mlp": 0.01022835,
      "balance_loss_clip": 1.04109764,
      "balance_loss_mlp": 1.01564121,
      "epoch": 0.7860277761077377,
      "flos": 19864059678720.0,
      "grad_norm": 2.7165200151715707,
      "language_loss": 0.77474707,
      "learning_rate": 4.6120033703744355e-07,
      "loss": 0.79601473,
      "num_input_tokens_seen": 140910780,
      "step": 6537,
      "time_per_iteration": 3.644303560256958
    },
    {
      "auxiliary_loss_clip": 0.01128731,
      "auxiliary_loss_mlp": 0.01026819,
      "balance_loss_clip": 1.04113722,
      "balance_loss_mlp": 1.01999998,
      "epoch": 0.7861480189983767,
      "flos": 26396174557440.0,
      "grad_norm": 2.1626602532023886,
      "language_loss": 0.78468531,
      "learning_rate": 4.607028720076822e-07,
      "loss": 0.8062408,
      "num_input_tokens_seen": 140927460,
      "step": 6538,
      "time_per_iteration": 2.719326972961426
    },
    {
      "auxiliary_loss_clip": 0.01151668,
      "auxiliary_loss_mlp": 0.01022403,
      "balance_loss_clip": 1.04510629,
      "balance_loss_mlp": 1.01529825,
      "epoch": 0.7862682618890158,
      "flos": 24236578177920.0,
      "grad_norm": 2.009243439209818,
      "language_loss": 0.73821133,
      "learning_rate": 4.6020564048807074e-07,
      "loss": 0.75995207,
      "num_input_tokens_seen": 140945135,
      "step": 6539,
      "time_per_iteration": 2.649303913116455
    },
    {
      "auxiliary_loss_clip": 0.0115164,
      "auxiliary_loss_mlp": 0.01023582,
      "balance_loss_clip": 1.04424453,
      "balance_loss_mlp": 1.01654232,
      "epoch": 0.7863885047796549,
      "flos": 47551508259840.0,
      "grad_norm": 2.0476943998704877,
      "language_loss": 0.72336972,
      "learning_rate": 4.5970864255403883e-07,
      "loss": 0.74512196,
      "num_input_tokens_seen": 140966660,
      "step": 6540,
      "time_per_iteration": 2.8656511306762695
    },
    {
      "auxiliary_loss_clip": 0.01142652,
      "auxiliary_loss_mlp": 0.01022248,
      "balance_loss_clip": 1.04286623,
      "balance_loss_mlp": 1.01573014,
      "epoch": 0.786508747670294,
      "flos": 24389234979840.0,
      "grad_norm": 2.138915679509711,
      "language_loss": 0.82185066,
      "learning_rate": 4.59211878280982e-07,
      "loss": 0.84349966,
      "num_input_tokens_seen": 140986175,
      "step": 6541,
      "time_per_iteration": 2.7298736572265625
    },
    {
      "auxiliary_loss_clip": 0.01141152,
      "auxiliary_loss_mlp": 0.010269,
      "balance_loss_clip": 1.04435539,
      "balance_loss_mlp": 1.01957464,
      "epoch": 0.786628990560933,
      "flos": 18041234238720.0,
      "grad_norm": 2.3896103460293086,
      "language_loss": 0.70372742,
      "learning_rate": 4.587153477442578e-07,
      "loss": 0.7254079,
      "num_input_tokens_seen": 141002490,
      "step": 6542,
      "time_per_iteration": 2.6405084133148193
    },
    {
      "auxiliary_loss_clip": 0.01170431,
      "auxiliary_loss_mlp": 0.01036391,
      "balance_loss_clip": 1.04768038,
      "balance_loss_mlp": 1.02875245,
      "epoch": 0.7867492334515722,
      "flos": 25848860048640.0,
      "grad_norm": 2.1652174479655764,
      "language_loss": 0.81551778,
      "learning_rate": 4.582190510191899e-07,
      "loss": 0.83758593,
      "num_input_tokens_seen": 141021150,
      "step": 6543,
      "time_per_iteration": 2.7101471424102783
    },
    {
      "auxiliary_loss_clip": 0.01121926,
      "auxiliary_loss_mlp": 0.01027424,
      "balance_loss_clip": 1.04262877,
      "balance_loss_mlp": 1.02078438,
      "epoch": 0.7868694763422113,
      "flos": 16580819070720.0,
      "grad_norm": 2.0619018079848894,
      "language_loss": 0.87624788,
      "learning_rate": 4.5772298818106625e-07,
      "loss": 0.89774138,
      "num_input_tokens_seen": 141036940,
      "step": 6544,
      "time_per_iteration": 2.655843496322632
    },
    {
      "auxiliary_loss_clip": 0.0113278,
      "auxiliary_loss_mlp": 0.01028974,
      "balance_loss_clip": 1.04543138,
      "balance_loss_mlp": 1.02148795,
      "epoch": 0.7869897192328503,
      "flos": 29386276272000.0,
      "grad_norm": 4.663540698872355,
      "language_loss": 0.7158742,
      "learning_rate": 4.572271593051384e-07,
      "loss": 0.73749173,
      "num_input_tokens_seen": 141054295,
      "step": 6545,
      "time_per_iteration": 2.718353033065796
    },
    {
      "auxiliary_loss_clip": 0.0110284,
      "auxiliary_loss_mlp": 0.01029868,
      "balance_loss_clip": 1.04103041,
      "balance_loss_mlp": 1.02253985,
      "epoch": 0.7871099621234895,
      "flos": 17128923678720.0,
      "grad_norm": 1.5556034695510152,
      "language_loss": 0.78277373,
      "learning_rate": 4.567315644666245e-07,
      "loss": 0.80410075,
      "num_input_tokens_seen": 141073090,
      "step": 6546,
      "time_per_iteration": 2.7910475730895996
    },
    {
      "auxiliary_loss_clip": 0.01117721,
      "auxiliary_loss_mlp": 0.01031127,
      "balance_loss_clip": 1.04316914,
      "balance_loss_mlp": 1.02477646,
      "epoch": 0.7872302050141285,
      "flos": 23440187784960.0,
      "grad_norm": 1.8621165870092966,
      "language_loss": 0.84666979,
      "learning_rate": 4.5623620374070507e-07,
      "loss": 0.86815828,
      "num_input_tokens_seen": 141092405,
      "step": 6547,
      "time_per_iteration": 2.7623181343078613
    },
    {
      "auxiliary_loss_clip": 0.0103324,
      "auxiliary_loss_mlp": 0.01002075,
      "balance_loss_clip": 1.01081991,
      "balance_loss_mlp": 1.00098419,
      "epoch": 0.7873504479047676,
      "flos": 65959752689280.0,
      "grad_norm": 0.9866431168711856,
      "language_loss": 0.58348221,
      "learning_rate": 4.557410772025263e-07,
      "loss": 0.60383534,
      "num_input_tokens_seen": 141154355,
      "step": 6548,
      "time_per_iteration": 3.3675191402435303
    },
    {
      "auxiliary_loss_clip": 0.01135339,
      "auxiliary_loss_mlp": 0.01025255,
      "balance_loss_clip": 1.04200256,
      "balance_loss_mlp": 1.01813495,
      "epoch": 0.7874706907954068,
      "flos": 23258336204160.0,
      "grad_norm": 2.066006259179901,
      "language_loss": 0.66360658,
      "learning_rate": 4.5524618492719803e-07,
      "loss": 0.68521249,
      "num_input_tokens_seen": 141173575,
      "step": 6549,
      "time_per_iteration": 2.6847474575042725
    },
    {
      "auxiliary_loss_clip": 0.01151731,
      "auxiliary_loss_mlp": 0.01022695,
      "balance_loss_clip": 1.04375494,
      "balance_loss_mlp": 1.01572442,
      "epoch": 0.7875909336860458,
      "flos": 28767786963840.0,
      "grad_norm": 1.5294075871146862,
      "language_loss": 0.79010028,
      "learning_rate": 4.54751526989795e-07,
      "loss": 0.81184453,
      "num_input_tokens_seen": 141195415,
      "step": 6550,
      "time_per_iteration": 2.824073314666748
    },
    {
      "auxiliary_loss_clip": 0.01156235,
      "auxiliary_loss_mlp": 0.01022516,
      "balance_loss_clip": 1.04496276,
      "balance_loss_mlp": 1.01473737,
      "epoch": 0.7877111765766849,
      "flos": 18697286194560.0,
      "grad_norm": 1.9889939881377474,
      "language_loss": 0.7950635,
      "learning_rate": 4.5425710346535775e-07,
      "loss": 0.81685096,
      "num_input_tokens_seen": 141213360,
      "step": 6551,
      "time_per_iteration": 2.6355676651000977
    },
    {
      "auxiliary_loss_clip": 0.01156237,
      "auxiliary_loss_mlp": 0.01026441,
      "balance_loss_clip": 1.04492104,
      "balance_loss_mlp": 1.0196346,
      "epoch": 0.787831419467324,
      "flos": 27592968833280.0,
      "grad_norm": 1.9672284385624146,
      "language_loss": 0.81780326,
      "learning_rate": 4.537629144288877e-07,
      "loss": 0.83963013,
      "num_input_tokens_seen": 141230815,
      "step": 6552,
      "time_per_iteration": 2.662985324859619
    },
    {
      "auxiliary_loss_clip": 0.01117042,
      "auxiliary_loss_mlp": 0.01025404,
      "balance_loss_clip": 1.03891873,
      "balance_loss_mlp": 1.01873112,
      "epoch": 0.7879516623579631,
      "flos": 18150187167360.0,
      "grad_norm": 1.8666838725698773,
      "language_loss": 0.7503131,
      "learning_rate": 4.5326895995535477e-07,
      "loss": 0.77173758,
      "num_input_tokens_seen": 141249715,
      "step": 6553,
      "time_per_iteration": 2.7002344131469727
    },
    {
      "auxiliary_loss_clip": 0.01150124,
      "auxiliary_loss_mlp": 0.01023505,
      "balance_loss_clip": 1.04413927,
      "balance_loss_mlp": 1.01643038,
      "epoch": 0.7880719052486022,
      "flos": 20339193807360.0,
      "grad_norm": 2.1727990837310536,
      "language_loss": 0.84388721,
      "learning_rate": 4.527752401196907e-07,
      "loss": 0.86562359,
      "num_input_tokens_seen": 141267730,
      "step": 6554,
      "time_per_iteration": 2.6406917572021484
    },
    {
      "auxiliary_loss_clip": 0.01132671,
      "auxiliary_loss_mlp": 0.01023317,
      "balance_loss_clip": 1.04241252,
      "balance_loss_mlp": 1.01650119,
      "epoch": 0.7881921481392413,
      "flos": 21653237053440.0,
      "grad_norm": 1.8793056401779376,
      "language_loss": 0.66952729,
      "learning_rate": 4.5228175499679254e-07,
      "loss": 0.69108713,
      "num_input_tokens_seen": 141287315,
      "step": 6555,
      "time_per_iteration": 2.619337797164917
    },
    {
      "auxiliary_loss_clip": 0.01056679,
      "auxiliary_loss_mlp": 0.01001159,
      "balance_loss_clip": 1.01170564,
      "balance_loss_mlp": 1.00004995,
      "epoch": 0.7883123910298804,
      "flos": 68565860058240.0,
      "grad_norm": 0.8149922310657656,
      "language_loss": 0.54506767,
      "learning_rate": 4.5178850466152174e-07,
      "loss": 0.56564605,
      "num_input_tokens_seen": 141346145,
      "step": 6556,
      "time_per_iteration": 3.3280327320098877
    },
    {
      "auxiliary_loss_clip": 0.01133662,
      "auxiliary_loss_mlp": 0.01029994,
      "balance_loss_clip": 1.04099798,
      "balance_loss_mlp": 1.02297235,
      "epoch": 0.7884326339205194,
      "flos": 19318217627520.0,
      "grad_norm": 1.817638445008885,
      "language_loss": 0.81955147,
      "learning_rate": 4.512954891887031e-07,
      "loss": 0.84118807,
      "num_input_tokens_seen": 141364445,
      "step": 6557,
      "time_per_iteration": 3.548931360244751
    },
    {
      "auxiliary_loss_clip": 0.01131631,
      "auxiliary_loss_mlp": 0.01029984,
      "balance_loss_clip": 1.04338217,
      "balance_loss_mlp": 1.02232528,
      "epoch": 0.7885528768111585,
      "flos": 17784903807360.0,
      "grad_norm": 2.447038106571842,
      "language_loss": 0.83515429,
      "learning_rate": 4.5080270865312806e-07,
      "loss": 0.8567704,
      "num_input_tokens_seen": 141381640,
      "step": 6558,
      "time_per_iteration": 3.563401222229004
    },
    {
      "auxiliary_loss_clip": 0.0115118,
      "auxiliary_loss_mlp": 0.01034699,
      "balance_loss_clip": 1.04364157,
      "balance_loss_mlp": 1.02793384,
      "epoch": 0.7886731197017977,
      "flos": 18807639753600.0,
      "grad_norm": 1.8815182859597837,
      "language_loss": 0.71232963,
      "learning_rate": 4.5031016312954985e-07,
      "loss": 0.73418844,
      "num_input_tokens_seen": 141399955,
      "step": 6559,
      "time_per_iteration": 2.6551735401153564
    },
    {
      "auxiliary_loss_clip": 0.01162075,
      "auxiliary_loss_mlp": 0.01025009,
      "balance_loss_clip": 1.04860187,
      "balance_loss_mlp": 1.01706338,
      "epoch": 0.7887933625924367,
      "flos": 33365358126720.0,
      "grad_norm": 2.286705896198435,
      "language_loss": 0.74358279,
      "learning_rate": 4.498178526926886e-07,
      "loss": 0.76545364,
      "num_input_tokens_seen": 141420820,
      "step": 6560,
      "time_per_iteration": 3.534046173095703
    },
    {
      "auxiliary_loss_clip": 0.01164854,
      "auxiliary_loss_mlp": 0.01034973,
      "balance_loss_clip": 1.04638696,
      "balance_loss_mlp": 1.02810609,
      "epoch": 0.7889136054830758,
      "flos": 17019360218880.0,
      "grad_norm": 2.1882539855749945,
      "language_loss": 0.72542822,
      "learning_rate": 4.4932577741722635e-07,
      "loss": 0.74742651,
      "num_input_tokens_seen": 141439350,
      "step": 6561,
      "time_per_iteration": 2.611114978790283
    },
    {
      "auxiliary_loss_clip": 0.01136127,
      "auxiliary_loss_mlp": 0.01021298,
      "balance_loss_clip": 1.04255438,
      "balance_loss_mlp": 1.01417542,
      "epoch": 0.7890338483737149,
      "flos": 29424629018880.0,
      "grad_norm": 1.6073754913268334,
      "language_loss": 0.74488145,
      "learning_rate": 4.4883393737780985e-07,
      "loss": 0.76645565,
      "num_input_tokens_seen": 141460300,
      "step": 6562,
      "time_per_iteration": 3.726135015487671
    },
    {
      "auxiliary_loss_clip": 0.01146444,
      "auxiliary_loss_mlp": 0.01026647,
      "balance_loss_clip": 1.04271138,
      "balance_loss_mlp": 1.01967072,
      "epoch": 0.789154091264354,
      "flos": 19971576063360.0,
      "grad_norm": 2.3646061175033113,
      "language_loss": 0.78332841,
      "learning_rate": 4.4834233264905254e-07,
      "loss": 0.80505931,
      "num_input_tokens_seen": 141477315,
      "step": 6563,
      "time_per_iteration": 2.656501293182373
    },
    {
      "auxiliary_loss_clip": 0.01118198,
      "auxiliary_loss_mlp": 0.01027295,
      "balance_loss_clip": 1.0412817,
      "balance_loss_mlp": 1.01908731,
      "epoch": 0.789274334154993,
      "flos": 14537825216640.0,
      "grad_norm": 2.761064206611322,
      "language_loss": 0.72229016,
      "learning_rate": 4.478509633055294e-07,
      "loss": 0.74374509,
      "num_input_tokens_seen": 141495025,
      "step": 6564,
      "time_per_iteration": 2.691890239715576
    },
    {
      "auxiliary_loss_clip": 0.01141516,
      "auxiliary_loss_mlp": 0.01032135,
      "balance_loss_clip": 1.04269898,
      "balance_loss_mlp": 1.02426434,
      "epoch": 0.7893945770456322,
      "flos": 21827403123840.0,
      "grad_norm": 3.7766899694245737,
      "language_loss": 0.79584867,
      "learning_rate": 4.473598294217813e-07,
      "loss": 0.81758523,
      "num_input_tokens_seen": 141510450,
      "step": 6565,
      "time_per_iteration": 2.7199668884277344
    },
    {
      "auxiliary_loss_clip": 0.0115106,
      "auxiliary_loss_mlp": 0.01019835,
      "balance_loss_clip": 1.04574788,
      "balance_loss_mlp": 1.01328707,
      "epoch": 0.7895148199362713,
      "flos": 20740639184640.0,
      "grad_norm": 2.0829669840529634,
      "language_loss": 0.71817136,
      "learning_rate": 4.468689310723124e-07,
      "loss": 0.73988026,
      "num_input_tokens_seen": 141528265,
      "step": 6566,
      "time_per_iteration": 2.5981650352478027
    },
    {
      "auxiliary_loss_clip": 0.0112801,
      "auxiliary_loss_mlp": 0.01027744,
      "balance_loss_clip": 1.04102254,
      "balance_loss_mlp": 1.02102959,
      "epoch": 0.7896350628269103,
      "flos": 16690669839360.0,
      "grad_norm": 1.8233368040251934,
      "language_loss": 0.78604758,
      "learning_rate": 4.463782683315913e-07,
      "loss": 0.80760515,
      "num_input_tokens_seen": 141547270,
      "step": 6567,
      "time_per_iteration": 2.736029863357544
    },
    {
      "auxiliary_loss_clip": 0.01165053,
      "auxiliary_loss_mlp": 0.01022007,
      "balance_loss_clip": 1.04595351,
      "balance_loss_mlp": 1.01503932,
      "epoch": 0.7897553057175495,
      "flos": 22638374438400.0,
      "grad_norm": 1.6596794948853546,
      "language_loss": 0.73316944,
      "learning_rate": 4.458878412740523e-07,
      "loss": 0.75504005,
      "num_input_tokens_seen": 141566050,
      "step": 6568,
      "time_per_iteration": 2.5992326736450195
    },
    {
      "auxiliary_loss_clip": 0.01151287,
      "auxiliary_loss_mlp": 0.01026369,
      "balance_loss_clip": 1.04574108,
      "balance_loss_mlp": 1.01936805,
      "epoch": 0.7898755486081885,
      "flos": 14537573821440.0,
      "grad_norm": 2.5278020281825526,
      "language_loss": 0.77608633,
      "learning_rate": 4.453976499740919e-07,
      "loss": 0.79786289,
      "num_input_tokens_seen": 141583695,
      "step": 6569,
      "time_per_iteration": 2.7377381324768066
    },
    {
      "auxiliary_loss_clip": 0.01151164,
      "auxiliary_loss_mlp": 0.01024438,
      "balance_loss_clip": 1.04572093,
      "balance_loss_mlp": 1.0171659,
      "epoch": 0.7899957914988276,
      "flos": 17238487138560.0,
      "grad_norm": 1.9272554374408004,
      "language_loss": 0.778036,
      "learning_rate": 4.4490769450607215e-07,
      "loss": 0.79979205,
      "num_input_tokens_seen": 141601320,
      "step": 6570,
      "time_per_iteration": 2.5805087089538574
    },
    {
      "auxiliary_loss_clip": 0.01120397,
      "auxiliary_loss_mlp": 0.01032536,
      "balance_loss_clip": 1.03802979,
      "balance_loss_mlp": 1.02529407,
      "epoch": 0.7901160343894668,
      "flos": 41279351086080.0,
      "grad_norm": 2.1183495770997682,
      "language_loss": 0.72867256,
      "learning_rate": 4.4441797494431845e-07,
      "loss": 0.75020194,
      "num_input_tokens_seen": 141623125,
      "step": 6571,
      "time_per_iteration": 2.885620355606079
    },
    {
      "auxiliary_loss_clip": 0.01149752,
      "auxiliary_loss_mlp": 0.01025505,
      "balance_loss_clip": 1.04517829,
      "balance_loss_mlp": 1.01845074,
      "epoch": 0.7902362772801058,
      "flos": 16837005847680.0,
      "grad_norm": 2.2991869277648824,
      "language_loss": 0.77602798,
      "learning_rate": 4.439284913631207e-07,
      "loss": 0.79778051,
      "num_input_tokens_seen": 141640335,
      "step": 6572,
      "time_per_iteration": 2.612126350402832
    },
    {
      "auxiliary_loss_clip": 0.01131071,
      "auxiliary_loss_mlp": 0.01029151,
      "balance_loss_clip": 1.04621196,
      "balance_loss_mlp": 1.0221951,
      "epoch": 0.7903565201707449,
      "flos": 27125987091840.0,
      "grad_norm": 2.1405212629013697,
      "language_loss": 0.83670473,
      "learning_rate": 4.434392438367347e-07,
      "loss": 0.858307,
      "num_input_tokens_seen": 141659760,
      "step": 6573,
      "time_per_iteration": 2.744536876678467
    },
    {
      "auxiliary_loss_clip": 0.01154702,
      "auxiliary_loss_mlp": 0.01020204,
      "balance_loss_clip": 1.04215395,
      "balance_loss_mlp": 1.01346552,
      "epoch": 0.790476763061384,
      "flos": 31025167142400.0,
      "grad_norm": 1.8112933648399423,
      "language_loss": 0.74432015,
      "learning_rate": 4.4295023243937677e-07,
      "loss": 0.76606917,
      "num_input_tokens_seen": 141679965,
      "step": 6574,
      "time_per_iteration": 2.67392635345459
    },
    {
      "auxiliary_loss_clip": 0.01154005,
      "auxiliary_loss_mlp": 0.01033672,
      "balance_loss_clip": 1.04677653,
      "balance_loss_mlp": 1.02540469,
      "epoch": 0.7905970059520231,
      "flos": 22089084681600.0,
      "grad_norm": 1.6691020590721684,
      "language_loss": 0.80659276,
      "learning_rate": 4.4246145724523123e-07,
      "loss": 0.82846951,
      "num_input_tokens_seen": 141697710,
      "step": 6575,
      "time_per_iteration": 2.6139297485351562
    },
    {
      "auxiliary_loss_clip": 0.01127273,
      "auxiliary_loss_mlp": 0.0102455,
      "balance_loss_clip": 1.04504657,
      "balance_loss_mlp": 1.01792252,
      "epoch": 0.7907172488426621,
      "flos": 20558141159040.0,
      "grad_norm": 2.231793773301612,
      "language_loss": 0.77643675,
      "learning_rate": 4.41972918328444e-07,
      "loss": 0.79795498,
      "num_input_tokens_seen": 141715145,
      "step": 6576,
      "time_per_iteration": 2.6796176433563232
    },
    {
      "auxiliary_loss_clip": 0.01151101,
      "auxiliary_loss_mlp": 0.01028562,
      "balance_loss_clip": 1.04685283,
      "balance_loss_mlp": 1.0212307,
      "epoch": 0.7908374917333013,
      "flos": 30081542901120.0,
      "grad_norm": 2.0559465054088375,
      "language_loss": 0.77554029,
      "learning_rate": 4.4148461576312646e-07,
      "loss": 0.79733694,
      "num_input_tokens_seen": 141734810,
      "step": 6577,
      "time_per_iteration": 2.7399675846099854
    },
    {
      "auxiliary_loss_clip": 0.011548,
      "auxiliary_loss_mlp": 0.01024901,
      "balance_loss_clip": 1.04763997,
      "balance_loss_mlp": 1.01827884,
      "epoch": 0.7909577346239404,
      "flos": 20996359084800.0,
      "grad_norm": 1.4259078603368212,
      "language_loss": 0.7441473,
      "learning_rate": 4.4099654962335343e-07,
      "loss": 0.76594436,
      "num_input_tokens_seen": 141755260,
      "step": 6578,
      "time_per_iteration": 2.6793501377105713
    },
    {
      "auxiliary_loss_clip": 0.01144472,
      "auxiliary_loss_mlp": 0.0102662,
      "balance_loss_clip": 1.04413962,
      "balance_loss_mlp": 1.0184598,
      "epoch": 0.7910779775145794,
      "flos": 26247935128320.0,
      "grad_norm": 1.8932801837412665,
      "language_loss": 0.75036913,
      "learning_rate": 4.405087199831636e-07,
      "loss": 0.77208,
      "num_input_tokens_seen": 141775500,
      "step": 6579,
      "time_per_iteration": 2.662370443344116
    },
    {
      "auxiliary_loss_clip": 0.01141344,
      "auxiliary_loss_mlp": 0.00761808,
      "balance_loss_clip": 1.04423678,
      "balance_loss_mlp": 1.00031042,
      "epoch": 0.7911982204052186,
      "flos": 22564434291840.0,
      "grad_norm": 2.4907727341526873,
      "language_loss": 0.67167008,
      "learning_rate": 4.400211269165619e-07,
      "loss": 0.6907016,
      "num_input_tokens_seen": 141791955,
      "step": 6580,
      "time_per_iteration": 2.6002676486968994
    },
    {
      "auxiliary_loss_clip": 0.01167511,
      "auxiliary_loss_mlp": 0.01027333,
      "balance_loss_clip": 1.04833972,
      "balance_loss_mlp": 1.02069855,
      "epoch": 0.7913184632958576,
      "flos": 23112538899840.0,
      "grad_norm": 1.471968634431362,
      "language_loss": 0.76648951,
      "learning_rate": 4.3953377049751416e-07,
      "loss": 0.7884379,
      "num_input_tokens_seen": 141812380,
      "step": 6581,
      "time_per_iteration": 2.4867498874664307
    },
    {
      "auxiliary_loss_clip": 0.0114613,
      "auxiliary_loss_mlp": 0.01024431,
      "balance_loss_clip": 1.04663146,
      "balance_loss_mlp": 1.01778245,
      "epoch": 0.7914387061864967,
      "flos": 12311758719360.0,
      "grad_norm": 2.835134876846789,
      "language_loss": 0.77872956,
      "learning_rate": 4.390466507999537e-07,
      "loss": 0.80043519,
      "num_input_tokens_seen": 141828130,
      "step": 6582,
      "time_per_iteration": 2.559220314025879
    },
    {
      "auxiliary_loss_clip": 0.01122812,
      "auxiliary_loss_mlp": 0.01024,
      "balance_loss_clip": 1.04213488,
      "balance_loss_mlp": 1.0170207,
      "epoch": 0.7915589490771359,
      "flos": 17603267708160.0,
      "grad_norm": 2.0871129673683924,
      "language_loss": 0.75623381,
      "learning_rate": 4.385597678977748e-07,
      "loss": 0.77770191,
      "num_input_tokens_seen": 141846965,
      "step": 6583,
      "time_per_iteration": 3.3919739723205566
    },
    {
      "auxiliary_loss_clip": 0.01136033,
      "auxiliary_loss_mlp": 0.01023604,
      "balance_loss_clip": 1.04259777,
      "balance_loss_mlp": 1.01632023,
      "epoch": 0.7916791919677749,
      "flos": 25591272641280.0,
      "grad_norm": 1.688755111005333,
      "language_loss": 0.75415182,
      "learning_rate": 4.3807312186483726e-07,
      "loss": 0.77574825,
      "num_input_tokens_seen": 141867685,
      "step": 6584,
      "time_per_iteration": 3.6593778133392334
    },
    {
      "auxiliary_loss_clip": 0.01151325,
      "auxiliary_loss_mlp": 0.01025702,
      "balance_loss_clip": 1.04845881,
      "balance_loss_mlp": 1.01820409,
      "epoch": 0.791799434858414,
      "flos": 18844340474880.0,
      "grad_norm": 2.067439016947969,
      "language_loss": 0.78546363,
      "learning_rate": 4.375867127749655e-07,
      "loss": 0.80723393,
      "num_input_tokens_seen": 141885960,
      "step": 6585,
      "time_per_iteration": 2.5773589611053467
    },
    {
      "auxiliary_loss_clip": 0.01124761,
      "auxiliary_loss_mlp": 0.01030391,
      "balance_loss_clip": 1.0449053,
      "balance_loss_mlp": 1.02321208,
      "epoch": 0.7919196777490531,
      "flos": 25812015672960.0,
      "grad_norm": 2.6259510177194083,
      "language_loss": 0.67132556,
      "learning_rate": 4.3710054070194744e-07,
      "loss": 0.69287705,
      "num_input_tokens_seen": 141905655,
      "step": 6586,
      "time_per_iteration": 3.5546681880950928
    },
    {
      "auxiliary_loss_clip": 0.01166712,
      "auxiliary_loss_mlp": 0.00762717,
      "balance_loss_clip": 1.04552519,
      "balance_loss_mlp": 1.00033295,
      "epoch": 0.7920399206396922,
      "flos": 11947624594560.0,
      "grad_norm": 3.010499486649441,
      "language_loss": 0.67067397,
      "learning_rate": 4.3661460571953455e-07,
      "loss": 0.68996823,
      "num_input_tokens_seen": 141922390,
      "step": 6587,
      "time_per_iteration": 2.565969944000244
    },
    {
      "auxiliary_loss_clip": 0.01155014,
      "auxiliary_loss_mlp": 0.01022379,
      "balance_loss_clip": 1.04440284,
      "balance_loss_mlp": 1.0151366,
      "epoch": 0.7921601635303313,
      "flos": 21579907438080.0,
      "grad_norm": 1.6232356796912386,
      "language_loss": 0.68545932,
      "learning_rate": 4.36128907901443e-07,
      "loss": 0.70723325,
      "num_input_tokens_seen": 141941985,
      "step": 6588,
      "time_per_iteration": 3.5271337032318115
    },
    {
      "auxiliary_loss_clip": 0.01129008,
      "auxiliary_loss_mlp": 0.0102648,
      "balance_loss_clip": 1.04249501,
      "balance_loss_mlp": 1.01930332,
      "epoch": 0.7922804064209703,
      "flos": 18113989236480.0,
      "grad_norm": 2.268056303636022,
      "language_loss": 0.72763604,
      "learning_rate": 4.356434473213519e-07,
      "loss": 0.74919093,
      "num_input_tokens_seen": 141959435,
      "step": 6589,
      "time_per_iteration": 2.6921334266662598
    },
    {
      "auxiliary_loss_clip": 0.01138042,
      "auxiliary_loss_mlp": 0.01025676,
      "balance_loss_clip": 1.04596138,
      "balance_loss_mlp": 1.01907218,
      "epoch": 0.7924006493116095,
      "flos": 21652806090240.0,
      "grad_norm": 1.6742639432437334,
      "language_loss": 0.79861051,
      "learning_rate": 4.351582240529068e-07,
      "loss": 0.82024765,
      "num_input_tokens_seen": 141980265,
      "step": 6590,
      "time_per_iteration": 2.700547456741333
    },
    {
      "auxiliary_loss_clip": 0.01048599,
      "auxiliary_loss_mlp": 0.01003617,
      "balance_loss_clip": 1.01125717,
      "balance_loss_mlp": 1.0025084,
      "epoch": 0.7925208922022485,
      "flos": 64242755694720.0,
      "grad_norm": 0.6803916454915758,
      "language_loss": 0.58201158,
      "learning_rate": 4.346732381697149e-07,
      "loss": 0.6025337,
      "num_input_tokens_seen": 142044395,
      "step": 6591,
      "time_per_iteration": 3.2797157764434814
    },
    {
      "auxiliary_loss_clip": 0.01132943,
      "auxiliary_loss_mlp": 0.01027058,
      "balance_loss_clip": 1.04304218,
      "balance_loss_mlp": 1.02052498,
      "epoch": 0.7926411350928876,
      "flos": 16941541403520.0,
      "grad_norm": 3.899620780568765,
      "language_loss": 0.80946338,
      "learning_rate": 4.3418848974534825e-07,
      "loss": 0.83106339,
      "num_input_tokens_seen": 142061335,
      "step": 6592,
      "time_per_iteration": 2.714526414871216
    },
    {
      "auxiliary_loss_clip": 0.0113174,
      "auxiliary_loss_mlp": 0.01028676,
      "balance_loss_clip": 1.0441947,
      "balance_loss_mlp": 1.02150536,
      "epoch": 0.7927613779835267,
      "flos": 34459987144320.0,
      "grad_norm": 1.7027060133792158,
      "language_loss": 0.68533641,
      "learning_rate": 4.3370397885334276e-07,
      "loss": 0.70694065,
      "num_input_tokens_seen": 142081965,
      "step": 6593,
      "time_per_iteration": 2.817122459411621
    },
    {
      "auxiliary_loss_clip": 0.01146755,
      "auxiliary_loss_mlp": 0.01024909,
      "balance_loss_clip": 1.0451901,
      "balance_loss_mlp": 1.01772666,
      "epoch": 0.7928816208741658,
      "flos": 18951174501120.0,
      "grad_norm": 1.842052047482128,
      "language_loss": 0.75661194,
      "learning_rate": 4.3321970556719777e-07,
      "loss": 0.77832854,
      "num_input_tokens_seen": 142100260,
      "step": 6594,
      "time_per_iteration": 2.6958248615264893
    },
    {
      "auxiliary_loss_clip": 0.01168985,
      "auxiliary_loss_mlp": 0.01029868,
      "balance_loss_clip": 1.04831183,
      "balance_loss_mlp": 1.02226877,
      "epoch": 0.7930018637648049,
      "flos": 18623022825600.0,
      "grad_norm": 2.4484711774387358,
      "language_loss": 0.72057086,
      "learning_rate": 4.3273566996037856e-07,
      "loss": 0.74255937,
      "num_input_tokens_seen": 142116955,
      "step": 6595,
      "time_per_iteration": 2.6819543838500977
    },
    {
      "auxiliary_loss_clip": 0.01137205,
      "auxiliary_loss_mlp": 0.0102242,
      "balance_loss_clip": 1.04269814,
      "balance_loss_mlp": 1.01595235,
      "epoch": 0.793122106655444,
      "flos": 24530650824960.0,
      "grad_norm": 1.941070734157617,
      "language_loss": 0.80763304,
      "learning_rate": 4.322518721063113e-07,
      "loss": 0.82922924,
      "num_input_tokens_seen": 142135505,
      "step": 6596,
      "time_per_iteration": 2.6704723834991455
    },
    {
      "auxiliary_loss_clip": 0.01153819,
      "auxiliary_loss_mlp": 0.0102467,
      "balance_loss_clip": 1.0477159,
      "balance_loss_mlp": 1.01732111,
      "epoch": 0.7932423495460831,
      "flos": 34421203434240.0,
      "grad_norm": 1.8224453360823383,
      "language_loss": 0.70263267,
      "learning_rate": 4.3176831207838906e-07,
      "loss": 0.72441757,
      "num_input_tokens_seen": 142158915,
      "step": 6597,
      "time_per_iteration": 2.738457679748535
    },
    {
      "auxiliary_loss_clip": 0.01152293,
      "auxiliary_loss_mlp": 0.0102466,
      "balance_loss_clip": 1.04740357,
      "balance_loss_mlp": 1.01736426,
      "epoch": 0.7933625924367221,
      "flos": 26980333441920.0,
      "grad_norm": 2.3053334392508975,
      "language_loss": 0.74474257,
      "learning_rate": 4.3128498994996685e-07,
      "loss": 0.76651216,
      "num_input_tokens_seen": 142178390,
      "step": 6598,
      "time_per_iteration": 2.6537108421325684
    },
    {
      "auxiliary_loss_clip": 0.01158147,
      "auxiliary_loss_mlp": 0.01025388,
      "balance_loss_clip": 1.04699552,
      "balance_loss_mlp": 1.01781261,
      "epoch": 0.7934828353273613,
      "flos": 29568630643200.0,
      "grad_norm": 2.304702368534531,
      "language_loss": 0.71498388,
      "learning_rate": 4.308019057943646e-07,
      "loss": 0.73681927,
      "num_input_tokens_seen": 142200115,
      "step": 6599,
      "time_per_iteration": 2.681518077850342
    },
    {
      "auxiliary_loss_clip": 0.01117741,
      "auxiliary_loss_mlp": 0.0102126,
      "balance_loss_clip": 1.04100549,
      "balance_loss_mlp": 1.01419103,
      "epoch": 0.7936030782180004,
      "flos": 28615381557120.0,
      "grad_norm": 1.8786489081817974,
      "language_loss": 0.74474412,
      "learning_rate": 4.3031905968486535e-07,
      "loss": 0.76613414,
      "num_input_tokens_seen": 142220945,
      "step": 6600,
      "time_per_iteration": 2.747075319290161
    },
    {
      "auxiliary_loss_clip": 0.01109768,
      "auxiliary_loss_mlp": 0.0102267,
      "balance_loss_clip": 1.04509687,
      "balance_loss_mlp": 1.01518393,
      "epoch": 0.7937233211086394,
      "flos": 16392574869120.0,
      "grad_norm": 2.087317650598343,
      "language_loss": 0.69005632,
      "learning_rate": 4.298364516947162e-07,
      "loss": 0.71138072,
      "num_input_tokens_seen": 142238175,
      "step": 6601,
      "time_per_iteration": 2.6921565532684326
    },
    {
      "auxiliary_loss_clip": 0.01107157,
      "auxiliary_loss_mlp": 0.01022351,
      "balance_loss_clip": 1.03985536,
      "balance_loss_mlp": 1.015082,
      "epoch": 0.7938435639992786,
      "flos": 22013420682240.0,
      "grad_norm": 2.232783682800349,
      "language_loss": 0.65724939,
      "learning_rate": 4.293540818971295e-07,
      "loss": 0.67854446,
      "num_input_tokens_seen": 142255980,
      "step": 6602,
      "time_per_iteration": 2.707388162612915
    },
    {
      "auxiliary_loss_clip": 0.0115636,
      "auxiliary_loss_mlp": 0.01028712,
      "balance_loss_clip": 1.04408228,
      "balance_loss_mlp": 1.02179515,
      "epoch": 0.7939638068899176,
      "flos": 22197032029440.0,
      "grad_norm": 2.2792970929418073,
      "language_loss": 0.76566613,
      "learning_rate": 4.2887195036527934e-07,
      "loss": 0.78751689,
      "num_input_tokens_seen": 142274785,
      "step": 6603,
      "time_per_iteration": 2.6386613845825195
    },
    {
      "auxiliary_loss_clip": 0.01144115,
      "auxiliary_loss_mlp": 0.01026335,
      "balance_loss_clip": 1.04141831,
      "balance_loss_mlp": 1.01875901,
      "epoch": 0.7940840497805567,
      "flos": 17745186343680.0,
      "grad_norm": 2.6881563889119175,
      "language_loss": 0.73623884,
      "learning_rate": 4.28390057172306e-07,
      "loss": 0.75794339,
      "num_input_tokens_seen": 142291290,
      "step": 6604,
      "time_per_iteration": 2.595839262008667
    },
    {
      "auxiliary_loss_clip": 0.01119241,
      "auxiliary_loss_mlp": 0.01025596,
      "balance_loss_clip": 1.03985262,
      "balance_loss_mlp": 1.01801074,
      "epoch": 0.7942042926711959,
      "flos": 23805435231360.0,
      "grad_norm": 2.1837075443046152,
      "language_loss": 0.71968406,
      "learning_rate": 4.279084023913111e-07,
      "loss": 0.74113244,
      "num_input_tokens_seen": 142309165,
      "step": 6605,
      "time_per_iteration": 2.7436530590057373
    },
    {
      "auxiliary_loss_clip": 0.01152123,
      "auxiliary_loss_mlp": 0.01028948,
      "balance_loss_clip": 1.04638004,
      "balance_loss_mlp": 1.02161026,
      "epoch": 0.7943245355618349,
      "flos": 19244959839360.0,
      "grad_norm": 1.8848455650486993,
      "language_loss": 0.69322759,
      "learning_rate": 4.2742698609536096e-07,
      "loss": 0.7150383,
      "num_input_tokens_seen": 142327475,
      "step": 6606,
      "time_per_iteration": 2.637544870376587
    },
    {
      "auxiliary_loss_clip": 0.01144007,
      "auxiliary_loss_mlp": 0.01028253,
      "balance_loss_clip": 1.0466063,
      "balance_loss_mlp": 1.02082586,
      "epoch": 0.794444778452474,
      "flos": 25007616547200.0,
      "grad_norm": 1.7130174245728278,
      "language_loss": 0.78467387,
      "learning_rate": 4.2694580835748706e-07,
      "loss": 0.80639642,
      "num_input_tokens_seen": 142347335,
      "step": 6607,
      "time_per_iteration": 2.7070252895355225
    },
    {
      "auxiliary_loss_clip": 0.0113929,
      "auxiliary_loss_mlp": 0.01026575,
      "balance_loss_clip": 1.04224896,
      "balance_loss_mlp": 1.01998019,
      "epoch": 0.7945650213431131,
      "flos": 23221491828480.0,
      "grad_norm": 2.099407377652163,
      "language_loss": 0.7390883,
      "learning_rate": 4.264648692506836e-07,
      "loss": 0.76074696,
      "num_input_tokens_seen": 142366125,
      "step": 6608,
      "time_per_iteration": 3.6462104320526123
    },
    {
      "auxiliary_loss_clip": 0.0113366,
      "auxiliary_loss_mlp": 0.01020289,
      "balance_loss_clip": 1.04245579,
      "balance_loss_mlp": 1.0125258,
      "epoch": 0.7946852642337522,
      "flos": 26062887237120.0,
      "grad_norm": 1.8052307590889696,
      "language_loss": 0.72130656,
      "learning_rate": 4.2598416884790824e-07,
      "loss": 0.74284613,
      "num_input_tokens_seen": 142385175,
      "step": 6609,
      "time_per_iteration": 2.6625540256500244
    },
    {
      "auxiliary_loss_clip": 0.01147592,
      "auxiliary_loss_mlp": 0.01025646,
      "balance_loss_clip": 1.04356539,
      "balance_loss_mlp": 1.01858544,
      "epoch": 0.7948055071243912,
      "flos": 23769704177280.0,
      "grad_norm": 2.0954804433810468,
      "language_loss": 0.80744004,
      "learning_rate": 4.255037072220828e-07,
      "loss": 0.82917237,
      "num_input_tokens_seen": 142406545,
      "step": 6610,
      "time_per_iteration": 3.6090331077575684
    },
    {
      "auxiliary_loss_clip": 0.01163737,
      "auxiliary_loss_mlp": 0.01020264,
      "balance_loss_clip": 1.04579067,
      "balance_loss_mlp": 1.01369858,
      "epoch": 0.7949257500150304,
      "flos": 21980814111360.0,
      "grad_norm": 2.2444320613222146,
      "language_loss": 0.71758533,
      "learning_rate": 4.2502348444609293e-07,
      "loss": 0.73942542,
      "num_input_tokens_seen": 142426165,
      "step": 6611,
      "time_per_iteration": 2.5966131687164307
    },
    {
      "auxiliary_loss_clip": 0.01104111,
      "auxiliary_loss_mlp": 0.0102738,
      "balance_loss_clip": 1.03645778,
      "balance_loss_mlp": 1.01999247,
      "epoch": 0.7950459929056695,
      "flos": 25774129802880.0,
      "grad_norm": 1.7992212715604992,
      "language_loss": 0.69166189,
      "learning_rate": 4.2454350059278844e-07,
      "loss": 0.71297681,
      "num_input_tokens_seen": 142447225,
      "step": 6612,
      "time_per_iteration": 3.7075371742248535
    },
    {
      "auxiliary_loss_clip": 0.01129987,
      "auxiliary_loss_mlp": 0.01030656,
      "balance_loss_clip": 1.03784871,
      "balance_loss_mlp": 1.02360177,
      "epoch": 0.7951662357963085,
      "flos": 22158068751360.0,
      "grad_norm": 1.8395695269170795,
      "language_loss": 0.84035015,
      "learning_rate": 4.240637557349824e-07,
      "loss": 0.8619566,
      "num_input_tokens_seen": 142464440,
      "step": 6613,
      "time_per_iteration": 2.6705801486968994
    },
    {
      "auxiliary_loss_clip": 0.01128205,
      "auxiliary_loss_mlp": 0.01029596,
      "balance_loss_clip": 1.04359269,
      "balance_loss_mlp": 1.02234757,
      "epoch": 0.7952864786869477,
      "flos": 24641938137600.0,
      "grad_norm": 4.372283167560864,
      "language_loss": 0.66712534,
      "learning_rate": 4.235842499454516e-07,
      "loss": 0.68870342,
      "num_input_tokens_seen": 142484355,
      "step": 6614,
      "time_per_iteration": 3.544501781463623
    },
    {
      "auxiliary_loss_clip": 0.01137659,
      "auxiliary_loss_mlp": 0.01031768,
      "balance_loss_clip": 1.04305339,
      "balance_loss_mlp": 1.02443326,
      "epoch": 0.7954067215775867,
      "flos": 21830922656640.0,
      "grad_norm": 1.702732078796595,
      "language_loss": 0.83062238,
      "learning_rate": 4.2310498329693687e-07,
      "loss": 0.85231662,
      "num_input_tokens_seen": 142505255,
      "step": 6615,
      "time_per_iteration": 2.7796730995178223
    },
    {
      "auxiliary_loss_clip": 0.01153501,
      "auxiliary_loss_mlp": 0.01028395,
      "balance_loss_clip": 1.04555619,
      "balance_loss_mlp": 1.02092612,
      "epoch": 0.7955269644682258,
      "flos": 24060652341120.0,
      "grad_norm": 1.9520837655629037,
      "language_loss": 0.81253648,
      "learning_rate": 4.2262595586214164e-07,
      "loss": 0.83435541,
      "num_input_tokens_seen": 142526350,
      "step": 6616,
      "time_per_iteration": 2.7931745052337646
    },
    {
      "auxiliary_loss_clip": 0.01156637,
      "auxiliary_loss_mlp": 0.01024413,
      "balance_loss_clip": 1.04625058,
      "balance_loss_mlp": 1.01654482,
      "epoch": 0.795647207358865,
      "flos": 25010741030400.0,
      "grad_norm": 1.7078938912867316,
      "language_loss": 0.76930273,
      "learning_rate": 4.221471677137358e-07,
      "loss": 0.7911132,
      "num_input_tokens_seen": 142547165,
      "step": 6617,
      "time_per_iteration": 2.717799663543701
    },
    {
      "auxiliary_loss_clip": 0.01127945,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 1.04291022,
      "balance_loss_mlp": 1.02163339,
      "epoch": 0.795767450249504,
      "flos": 14648358343680.0,
      "grad_norm": 1.5765809689404129,
      "language_loss": 0.70008832,
      "learning_rate": 4.216686189243492e-07,
      "loss": 0.72164714,
      "num_input_tokens_seen": 142565955,
      "step": 6618,
      "time_per_iteration": 2.7128725051879883
    },
    {
      "auxiliary_loss_clip": 0.01121928,
      "auxiliary_loss_mlp": 0.0102681,
      "balance_loss_clip": 1.04343247,
      "balance_loss_mlp": 1.01901376,
      "epoch": 0.7958876931401431,
      "flos": 18547897530240.0,
      "grad_norm": 1.6533442996143157,
      "language_loss": 0.72681165,
      "learning_rate": 4.211903095665785e-07,
      "loss": 0.748299,
      "num_input_tokens_seen": 142585340,
      "step": 6619,
      "time_per_iteration": 2.72342586517334
    },
    {
      "auxiliary_loss_clip": 0.01147253,
      "auxiliary_loss_mlp": 0.01023265,
      "balance_loss_clip": 1.04300714,
      "balance_loss_mlp": 1.01664567,
      "epoch": 0.7960079360307821,
      "flos": 21543960902400.0,
      "grad_norm": 1.933912079442407,
      "language_loss": 0.75134587,
      "learning_rate": 4.2071223971298277e-07,
      "loss": 0.77305108,
      "num_input_tokens_seen": 142602525,
      "step": 6620,
      "time_per_iteration": 2.660771131515503
    },
    {
      "auxiliary_loss_clip": 0.01153009,
      "auxiliary_loss_mlp": 0.01022202,
      "balance_loss_clip": 1.04366338,
      "balance_loss_mlp": 1.01469207,
      "epoch": 0.7961281789214213,
      "flos": 25481745095040.0,
      "grad_norm": 2.42667822648509,
      "language_loss": 0.61178529,
      "learning_rate": 4.2023440943608433e-07,
      "loss": 0.63353741,
      "num_input_tokens_seen": 142622490,
      "step": 6621,
      "time_per_iteration": 2.6651463508605957
    },
    {
      "auxiliary_loss_clip": 0.0115114,
      "auxiliary_loss_mlp": 0.01024614,
      "balance_loss_clip": 1.0433557,
      "balance_loss_mlp": 1.01798654,
      "epoch": 0.7962484218120603,
      "flos": 21944436612480.0,
      "grad_norm": 1.6892435322744372,
      "language_loss": 0.78271627,
      "learning_rate": 4.1975681880837023e-07,
      "loss": 0.80447382,
      "num_input_tokens_seen": 142642495,
      "step": 6622,
      "time_per_iteration": 2.701033592224121
    },
    {
      "auxiliary_loss_clip": 0.01118987,
      "auxiliary_loss_mlp": 0.01025705,
      "balance_loss_clip": 1.03900933,
      "balance_loss_mlp": 1.01862729,
      "epoch": 0.7963686647026994,
      "flos": 18876264687360.0,
      "grad_norm": 1.7421357881806323,
      "language_loss": 0.82522047,
      "learning_rate": 4.192794679022895e-07,
      "loss": 0.84666735,
      "num_input_tokens_seen": 142660820,
      "step": 6623,
      "time_per_iteration": 2.8209164142608643
    },
    {
      "auxiliary_loss_clip": 0.01151061,
      "auxiliary_loss_mlp": 0.0102591,
      "balance_loss_clip": 1.04136157,
      "balance_loss_mlp": 1.01936889,
      "epoch": 0.7964889075933386,
      "flos": 29716582763520.0,
      "grad_norm": 1.7459344701964405,
      "language_loss": 0.72072279,
      "learning_rate": 4.1880235679025743e-07,
      "loss": 0.74249244,
      "num_input_tokens_seen": 142680915,
      "step": 6624,
      "time_per_iteration": 2.7281858921051025
    },
    {
      "auxiliary_loss_clip": 0.01096769,
      "auxiliary_loss_mlp": 0.01026236,
      "balance_loss_clip": 1.03850508,
      "balance_loss_mlp": 1.01868987,
      "epoch": 0.7966091504839776,
      "flos": 29491458272640.0,
      "grad_norm": 1.8098663370232195,
      "language_loss": 0.6398896,
      "learning_rate": 4.1832548554464986e-07,
      "loss": 0.6611197,
      "num_input_tokens_seen": 142699210,
      "step": 6625,
      "time_per_iteration": 2.7737364768981934
    },
    {
      "auxiliary_loss_clip": 0.01050562,
      "auxiliary_loss_mlp": 0.01003174,
      "balance_loss_clip": 1.01047742,
      "balance_loss_mlp": 1.00208306,
      "epoch": 0.7967293933746167,
      "flos": 67288697101440.0,
      "grad_norm": 0.7389845512539909,
      "language_loss": 0.58728892,
      "learning_rate": 4.178488542378098e-07,
      "loss": 0.60782623,
      "num_input_tokens_seen": 142756790,
      "step": 6626,
      "time_per_iteration": 3.131082057952881
    },
    {
      "auxiliary_loss_clip": 0.01170185,
      "auxiliary_loss_mlp": 0.01025291,
      "balance_loss_clip": 1.04675663,
      "balance_loss_mlp": 1.01732206,
      "epoch": 0.7968496362652558,
      "flos": 25554679660800.0,
      "grad_norm": 2.1396027904525288,
      "language_loss": 0.89241338,
      "learning_rate": 4.173724629420401e-07,
      "loss": 0.91436815,
      "num_input_tokens_seen": 142778150,
      "step": 6627,
      "time_per_iteration": 2.68062686920166
    },
    {
      "auxiliary_loss_clip": 0.0114327,
      "auxiliary_loss_mlp": 0.0102512,
      "balance_loss_clip": 1.04390192,
      "balance_loss_mlp": 1.01717436,
      "epoch": 0.7969698791558949,
      "flos": 14501088581760.0,
      "grad_norm": 2.7968331721948316,
      "language_loss": 0.6829145,
      "learning_rate": 4.168963117296087e-07,
      "loss": 0.70459837,
      "num_input_tokens_seen": 142795485,
      "step": 6628,
      "time_per_iteration": 2.6608705520629883
    },
    {
      "auxiliary_loss_clip": 0.01167098,
      "auxiliary_loss_mlp": 0.01020548,
      "balance_loss_clip": 1.04657626,
      "balance_loss_mlp": 1.01330888,
      "epoch": 0.797090122046534,
      "flos": 22127545169280.0,
      "grad_norm": 2.1799370837378804,
      "language_loss": 0.75819182,
      "learning_rate": 4.1642040067274876e-07,
      "loss": 0.78006828,
      "num_input_tokens_seen": 142815155,
      "step": 6629,
      "time_per_iteration": 2.6247172355651855
    },
    {
      "auxiliary_loss_clip": 0.01143317,
      "auxiliary_loss_mlp": 0.01022652,
      "balance_loss_clip": 1.04322028,
      "balance_loss_mlp": 1.01581573,
      "epoch": 0.7972103649371731,
      "flos": 19897671830400.0,
      "grad_norm": 2.5477718161510037,
      "language_loss": 0.72408128,
      "learning_rate": 4.1594472984365493e-07,
      "loss": 0.74574101,
      "num_input_tokens_seen": 142833840,
      "step": 6630,
      "time_per_iteration": 2.6708824634552
    },
    {
      "auxiliary_loss_clip": 0.01148682,
      "auxiliary_loss_mlp": 0.01023632,
      "balance_loss_clip": 1.04490709,
      "balance_loss_mlp": 1.01634264,
      "epoch": 0.7973306078278122,
      "flos": 36058621847040.0,
      "grad_norm": 1.8757672118327924,
      "language_loss": 0.7760886,
      "learning_rate": 4.154692993144862e-07,
      "loss": 0.79781175,
      "num_input_tokens_seen": 142853610,
      "step": 6631,
      "time_per_iteration": 2.7654147148132324
    },
    {
      "auxiliary_loss_clip": 0.0116578,
      "auxiliary_loss_mlp": 0.00761995,
      "balance_loss_clip": 1.04559827,
      "balance_loss_mlp": 1.00034165,
      "epoch": 0.7974508507184512,
      "flos": 21360600950400.0,
      "grad_norm": 2.218169043351546,
      "language_loss": 0.71765107,
      "learning_rate": 4.1499410915736476e-07,
      "loss": 0.73692882,
      "num_input_tokens_seen": 142872540,
      "step": 6632,
      "time_per_iteration": 2.586106538772583
    },
    {
      "auxiliary_loss_clip": 0.01056106,
      "auxiliary_loss_mlp": 0.01001677,
      "balance_loss_clip": 1.01096582,
      "balance_loss_mlp": 1.00061655,
      "epoch": 0.7975710936090904,
      "flos": 68253115317120.0,
      "grad_norm": 0.772278231300321,
      "language_loss": 0.64280772,
      "learning_rate": 4.145191594443762e-07,
      "loss": 0.66338563,
      "num_input_tokens_seen": 142936895,
      "step": 6633,
      "time_per_iteration": 3.3651533126831055
    },
    {
      "auxiliary_loss_clip": 0.0112014,
      "auxiliary_loss_mlp": 0.01023464,
      "balance_loss_clip": 1.0427376,
      "balance_loss_mlp": 1.01614463,
      "epoch": 0.7976913364997295,
      "flos": 22492433479680.0,
      "grad_norm": 1.9221053805459567,
      "language_loss": 0.70264852,
      "learning_rate": 4.140444502475713e-07,
      "loss": 0.7240845,
      "num_input_tokens_seen": 142956445,
      "step": 6634,
      "time_per_iteration": 3.744769811630249
    },
    {
      "auxiliary_loss_clip": 0.01148017,
      "auxiliary_loss_mlp": 0.01027908,
      "balance_loss_clip": 1.04295862,
      "balance_loss_mlp": 1.02057409,
      "epoch": 0.7978115793903685,
      "flos": 15263220378240.0,
      "grad_norm": 2.204993520311742,
      "language_loss": 0.69948012,
      "learning_rate": 4.1356998163896216e-07,
      "loss": 0.72123933,
      "num_input_tokens_seen": 142973495,
      "step": 6635,
      "time_per_iteration": 2.584517002105713
    },
    {
      "auxiliary_loss_clip": 0.01130287,
      "auxiliary_loss_mlp": 0.01024034,
      "balance_loss_clip": 1.04370642,
      "balance_loss_mlp": 1.01689351,
      "epoch": 0.7979318222810077,
      "flos": 19719232041600.0,
      "grad_norm": 2.2650138702848337,
      "language_loss": 0.74848908,
      "learning_rate": 4.130957536905255e-07,
      "loss": 0.77003229,
      "num_input_tokens_seen": 142991510,
      "step": 6636,
      "time_per_iteration": 3.7424020767211914
    },
    {
      "auxiliary_loss_clip": 0.01148713,
      "auxiliary_loss_mlp": 0.01022475,
      "balance_loss_clip": 1.04510033,
      "balance_loss_mlp": 1.01525378,
      "epoch": 0.7980520651716467,
      "flos": 15560273854080.0,
      "grad_norm": 2.1155481574875807,
      "language_loss": 0.71889347,
      "learning_rate": 4.1262176647420134e-07,
      "loss": 0.74060535,
      "num_input_tokens_seen": 143009675,
      "step": 6637,
      "time_per_iteration": 2.6771535873413086
    },
    {
      "auxiliary_loss_clip": 0.01143596,
      "auxiliary_loss_mlp": 0.0102627,
      "balance_loss_clip": 1.04423714,
      "balance_loss_mlp": 1.01931989,
      "epoch": 0.7981723080622858,
      "flos": 22309432663680.0,
      "grad_norm": 1.7828988147090705,
      "language_loss": 0.79890704,
      "learning_rate": 4.121480200618923e-07,
      "loss": 0.8206057,
      "num_input_tokens_seen": 143029330,
      "step": 6638,
      "time_per_iteration": 3.58729887008667
    },
    {
      "auxiliary_loss_clip": 0.01134274,
      "auxiliary_loss_mlp": 0.01030838,
      "balance_loss_clip": 1.04290044,
      "balance_loss_mlp": 1.02367103,
      "epoch": 0.798292550952925,
      "flos": 22929573997440.0,
      "grad_norm": 1.7560359721772547,
      "language_loss": 0.80230844,
      "learning_rate": 4.116745145254674e-07,
      "loss": 0.82395959,
      "num_input_tokens_seen": 143048865,
      "step": 6639,
      "time_per_iteration": 2.764585256576538
    },
    {
      "auxiliary_loss_clip": 0.01038556,
      "auxiliary_loss_mlp": 0.01000227,
      "balance_loss_clip": 1.0095582,
      "balance_loss_mlp": 0.99916619,
      "epoch": 0.798412793843564,
      "flos": 64497936890880.0,
      "grad_norm": 0.765878016361562,
      "language_loss": 0.58009958,
      "learning_rate": 4.1120124993675476e-07,
      "loss": 0.60048747,
      "num_input_tokens_seen": 143113295,
      "step": 6640,
      "time_per_iteration": 4.163068056106567
    },
    {
      "auxiliary_loss_clip": 0.01146323,
      "auxiliary_loss_mlp": 0.01025695,
      "balance_loss_clip": 1.04511511,
      "balance_loss_mlp": 1.01853311,
      "epoch": 0.7985330367342031,
      "flos": 13586910514560.0,
      "grad_norm": 1.8750743392976605,
      "language_loss": 0.62057328,
      "learning_rate": 4.107282263675498e-07,
      "loss": 0.64229345,
      "num_input_tokens_seen": 143130965,
      "step": 6641,
      "time_per_iteration": 2.6209092140197754
    },
    {
      "auxiliary_loss_clip": 0.01043504,
      "auxiliary_loss_mlp": 0.00753948,
      "balance_loss_clip": 1.01348042,
      "balance_loss_mlp": 1.00032485,
      "epoch": 0.7986532796248422,
      "flos": 67698797656320.0,
      "grad_norm": 0.7719501805341424,
      "language_loss": 0.52439427,
      "learning_rate": 4.1025544388960907e-07,
      "loss": 0.54236877,
      "num_input_tokens_seen": 143192005,
      "step": 6642,
      "time_per_iteration": 3.2523021697998047
    },
    {
      "auxiliary_loss_clip": 0.01154395,
      "auxiliary_loss_mlp": 0.0102572,
      "balance_loss_clip": 1.04630303,
      "balance_loss_mlp": 1.01868939,
      "epoch": 0.7987735225154813,
      "flos": 22455373622400.0,
      "grad_norm": 3.764465394838307,
      "language_loss": 0.71760416,
      "learning_rate": 4.097829025746538e-07,
      "loss": 0.73940539,
      "num_input_tokens_seen": 143213550,
      "step": 6643,
      "time_per_iteration": 2.6863629817962646
    },
    {
      "auxiliary_loss_clip": 0.01053809,
      "auxiliary_loss_mlp": 0.01002492,
      "balance_loss_clip": 1.01017475,
      "balance_loss_mlp": 1.0013833,
      "epoch": 0.7988937654061203,
      "flos": 68864098682880.0,
      "grad_norm": 0.657498188933818,
      "language_loss": 0.6100446,
      "learning_rate": 4.0931060249436757e-07,
      "loss": 0.6306076,
      "num_input_tokens_seen": 143277390,
      "step": 6644,
      "time_per_iteration": 3.2721004486083984
    },
    {
      "auxiliary_loss_clip": 0.01151821,
      "auxiliary_loss_mlp": 0.010291,
      "balance_loss_clip": 1.04664159,
      "balance_loss_mlp": 1.02166152,
      "epoch": 0.7990140082967595,
      "flos": 20806893820800.0,
      "grad_norm": 1.9901529004076401,
      "language_loss": 0.6951946,
      "learning_rate": 4.088385437203978e-07,
      "loss": 0.71700382,
      "num_input_tokens_seen": 143294400,
      "step": 6645,
      "time_per_iteration": 2.5758488178253174
    },
    {
      "auxiliary_loss_clip": 0.01166981,
      "auxiliary_loss_mlp": 0.01026221,
      "balance_loss_clip": 1.04625666,
      "balance_loss_mlp": 1.01885378,
      "epoch": 0.7991342511873986,
      "flos": 18985289443200.0,
      "grad_norm": 2.0831947701096554,
      "language_loss": 0.77237403,
      "learning_rate": 4.083667263243564e-07,
      "loss": 0.79430604,
      "num_input_tokens_seen": 143312745,
      "step": 6646,
      "time_per_iteration": 2.6265087127685547
    },
    {
      "auxiliary_loss_clip": 0.01147306,
      "auxiliary_loss_mlp": 0.01023143,
      "balance_loss_clip": 1.04510188,
      "balance_loss_mlp": 1.01668739,
      "epoch": 0.7992544940780376,
      "flos": 20816805974400.0,
      "grad_norm": 1.6840189143213304,
      "language_loss": 0.7187891,
      "learning_rate": 4.0789515037781653e-07,
      "loss": 0.74049366,
      "num_input_tokens_seen": 143333470,
      "step": 6647,
      "time_per_iteration": 2.6656789779663086
    },
    {
      "auxiliary_loss_clip": 0.01153527,
      "auxiliary_loss_mlp": 0.01026603,
      "balance_loss_clip": 1.0429256,
      "balance_loss_mlp": 1.01872909,
      "epoch": 0.7993747369686768,
      "flos": 12640772321280.0,
      "grad_norm": 1.8958411360164624,
      "language_loss": 0.82700378,
      "learning_rate": 4.0742381595231755e-07,
      "loss": 0.84880507,
      "num_input_tokens_seen": 143350195,
      "step": 6648,
      "time_per_iteration": 2.6367146968841553
    },
    {
      "auxiliary_loss_clip": 0.01129939,
      "auxiliary_loss_mlp": 0.01023216,
      "balance_loss_clip": 1.04398274,
      "balance_loss_mlp": 1.01624846,
      "epoch": 0.7994949798593158,
      "flos": 20078769225600.0,
      "grad_norm": 1.9573830764149018,
      "language_loss": 0.78186178,
      "learning_rate": 4.06952723119359e-07,
      "loss": 0.8033933,
      "num_input_tokens_seen": 143370070,
      "step": 6649,
      "time_per_iteration": 2.756636381149292
    },
    {
      "auxiliary_loss_clip": 0.01131829,
      "auxiliary_loss_mlp": 0.01027184,
      "balance_loss_clip": 1.04419208,
      "balance_loss_mlp": 1.01968575,
      "epoch": 0.7996152227499549,
      "flos": 38654209509120.0,
      "grad_norm": 1.918896244060421,
      "language_loss": 0.67443299,
      "learning_rate": 4.0648187195040504e-07,
      "loss": 0.69602311,
      "num_input_tokens_seen": 143392275,
      "step": 6650,
      "time_per_iteration": 2.7790627479553223
    },
    {
      "auxiliary_loss_clip": 0.01050058,
      "auxiliary_loss_mlp": 0.010009,
      "balance_loss_clip": 1.01004803,
      "balance_loss_mlp": 0.99973184,
      "epoch": 0.799735465640594,
      "flos": 70243821947520.0,
      "grad_norm": 0.8119900651899423,
      "language_loss": 0.67610514,
      "learning_rate": 4.060112625168848e-07,
      "loss": 0.69661474,
      "num_input_tokens_seen": 143457385,
      "step": 6651,
      "time_per_iteration": 3.415618658065796
    },
    {
      "auxiliary_loss_clip": 0.01170947,
      "auxiliary_loss_mlp": 0.01020788,
      "balance_loss_clip": 1.05035496,
      "balance_loss_mlp": 1.01345682,
      "epoch": 0.7998557085312331,
      "flos": 24240995550720.0,
      "grad_norm": 2.1154873670917764,
      "language_loss": 0.74255818,
      "learning_rate": 4.055408948901886e-07,
      "loss": 0.76447558,
      "num_input_tokens_seen": 143478785,
      "step": 6652,
      "time_per_iteration": 2.6214535236358643
    },
    {
      "auxiliary_loss_clip": 0.01157811,
      "auxiliary_loss_mlp": 0.01021885,
      "balance_loss_clip": 1.04539061,
      "balance_loss_mlp": 1.01413655,
      "epoch": 0.7999759514218722,
      "flos": 27564025449600.0,
      "grad_norm": 1.724964186143899,
      "language_loss": 0.71275151,
      "learning_rate": 4.050707691416708e-07,
      "loss": 0.73454845,
      "num_input_tokens_seen": 143500095,
      "step": 6653,
      "time_per_iteration": 2.7239198684692383
    },
    {
      "auxiliary_loss_clip": 0.01050168,
      "auxiliary_loss_mlp": 0.0100051,
      "balance_loss_clip": 1.0099349,
      "balance_loss_mlp": 0.99940729,
      "epoch": 0.8000961943125112,
      "flos": 67337428878720.0,
      "grad_norm": 0.6714850581057558,
      "language_loss": 0.59780562,
      "learning_rate": 4.046008853426495e-07,
      "loss": 0.61831236,
      "num_input_tokens_seen": 143563410,
      "step": 6654,
      "time_per_iteration": 3.318298578262329
    },
    {
      "auxiliary_loss_clip": 0.01121327,
      "auxiliary_loss_mlp": 0.01024978,
      "balance_loss_clip": 1.04183161,
      "balance_loss_mlp": 1.01764083,
      "epoch": 0.8002164372031504,
      "flos": 28733815676160.0,
      "grad_norm": 1.9017568002732907,
      "language_loss": 0.6245991,
      "learning_rate": 4.0413124356440464e-07,
      "loss": 0.64606214,
      "num_input_tokens_seen": 143587455,
      "step": 6655,
      "time_per_iteration": 2.7620980739593506
    },
    {
      "auxiliary_loss_clip": 0.01113668,
      "auxiliary_loss_mlp": 0.01023177,
      "balance_loss_clip": 1.04051781,
      "balance_loss_mlp": 1.01546419,
      "epoch": 0.8003366800937894,
      "flos": 17639429725440.0,
      "grad_norm": 2.2930991992347467,
      "language_loss": 0.82098979,
      "learning_rate": 4.0366184387818223e-07,
      "loss": 0.84235823,
      "num_input_tokens_seen": 143605915,
      "step": 6656,
      "time_per_iteration": 2.737867832183838
    },
    {
      "auxiliary_loss_clip": 0.01170641,
      "auxiliary_loss_mlp": 0.0102768,
      "balance_loss_clip": 1.04659057,
      "balance_loss_mlp": 1.01997864,
      "epoch": 0.8004569229844285,
      "flos": 25995303797760.0,
      "grad_norm": 1.83130701499229,
      "language_loss": 0.85740399,
      "learning_rate": 4.0319268635518797e-07,
      "loss": 0.8793872,
      "num_input_tokens_seen": 143626490,
      "step": 6657,
      "time_per_iteration": 2.6952006816864014
    },
    {
      "auxiliary_loss_clip": 0.01153349,
      "auxiliary_loss_mlp": 0.01022722,
      "balance_loss_clip": 1.04441631,
      "balance_loss_mlp": 1.0156796,
      "epoch": 0.8005771658750677,
      "flos": 20812352688000.0,
      "grad_norm": 1.5267786824322473,
      "language_loss": 0.74763185,
      "learning_rate": 4.027237710665943e-07,
      "loss": 0.76939255,
      "num_input_tokens_seen": 143644955,
      "step": 6658,
      "time_per_iteration": 2.695887327194214
    },
    {
      "auxiliary_loss_clip": 0.0112783,
      "auxiliary_loss_mlp": 0.01026104,
      "balance_loss_clip": 1.04070568,
      "balance_loss_mlp": 1.01833117,
      "epoch": 0.8006974087657067,
      "flos": 25812626204160.0,
      "grad_norm": 1.792931699090747,
      "language_loss": 0.69331968,
      "learning_rate": 4.022550980835344e-07,
      "loss": 0.71485913,
      "num_input_tokens_seen": 143667200,
      "step": 6659,
      "time_per_iteration": 2.7515246868133545
    },
    {
      "auxiliary_loss_clip": 0.01122326,
      "auxiliary_loss_mlp": 0.0102029,
      "balance_loss_clip": 1.03961229,
      "balance_loss_mlp": 1.01287508,
      "epoch": 0.8008176516563458,
      "flos": 17164690646400.0,
      "grad_norm": 2.9960552429431964,
      "language_loss": 0.7979877,
      "learning_rate": 4.017866674771051e-07,
      "loss": 0.8194139,
      "num_input_tokens_seen": 143684685,
      "step": 6660,
      "time_per_iteration": 3.6636250019073486
    },
    {
      "auxiliary_loss_clip": 0.01103583,
      "auxiliary_loss_mlp": 0.01029242,
      "balance_loss_clip": 1.0380764,
      "balance_loss_mlp": 1.02210176,
      "epoch": 0.8009378945469849,
      "flos": 24207311571840.0,
      "grad_norm": 1.8592722494133118,
      "language_loss": 0.74507761,
      "learning_rate": 4.013184793183688e-07,
      "loss": 0.76640588,
      "num_input_tokens_seen": 143706780,
      "step": 6661,
      "time_per_iteration": 3.7798209190368652
    },
    {
      "auxiliary_loss_clip": 0.01149743,
      "auxiliary_loss_mlp": 0.01019786,
      "balance_loss_clip": 1.04176569,
      "balance_loss_mlp": 1.01274645,
      "epoch": 0.801058137437624,
      "flos": 19787318271360.0,
      "grad_norm": 2.168932486002685,
      "language_loss": 0.7247718,
      "learning_rate": 4.008505336783472e-07,
      "loss": 0.74646711,
      "num_input_tokens_seen": 143724505,
      "step": 6662,
      "time_per_iteration": 2.7066030502319336
    },
    {
      "auxiliary_loss_clip": 0.01142856,
      "auxiliary_loss_mlp": 0.01020144,
      "balance_loss_clip": 1.04217696,
      "balance_loss_mlp": 1.01384974,
      "epoch": 0.801178380328263,
      "flos": 18659400324480.0,
      "grad_norm": 1.9810694453285715,
      "language_loss": 0.80543631,
      "learning_rate": 4.003828306280284e-07,
      "loss": 0.8270663,
      "num_input_tokens_seen": 143742180,
      "step": 6663,
      "time_per_iteration": 2.7109992504119873
    },
    {
      "auxiliary_loss_clip": 0.01153562,
      "auxiliary_loss_mlp": 0.0102653,
      "balance_loss_clip": 1.04398632,
      "balance_loss_mlp": 1.01992035,
      "epoch": 0.8012986232189022,
      "flos": 15706573948800.0,
      "grad_norm": 1.8856192083850938,
      "language_loss": 0.77977014,
      "learning_rate": 3.999153702383626e-07,
      "loss": 0.80157101,
      "num_input_tokens_seen": 143760070,
      "step": 6664,
      "time_per_iteration": 3.676591396331787
    },
    {
      "auxiliary_loss_clip": 0.01157035,
      "auxiliary_loss_mlp": 0.01021981,
      "balance_loss_clip": 1.04482079,
      "balance_loss_mlp": 1.01474488,
      "epoch": 0.8014188661095413,
      "flos": 28584139703040.0,
      "grad_norm": 2.0374664609452426,
      "language_loss": 0.73997349,
      "learning_rate": 3.9944815258026263e-07,
      "loss": 0.76176357,
      "num_input_tokens_seen": 143781890,
      "step": 6665,
      "time_per_iteration": 3.7532827854156494
    },
    {
      "auxiliary_loss_clip": 0.01157038,
      "auxiliary_loss_mlp": 0.01028747,
      "balance_loss_clip": 1.04589391,
      "balance_loss_mlp": 1.02123094,
      "epoch": 0.8015391090001803,
      "flos": 29310360877440.0,
      "grad_norm": 1.9699411345399531,
      "language_loss": 0.83173484,
      "learning_rate": 3.989811777246057e-07,
      "loss": 0.85359263,
      "num_input_tokens_seen": 143802060,
      "step": 6666,
      "time_per_iteration": 2.6741790771484375
    },
    {
      "auxiliary_loss_clip": 0.01064715,
      "auxiliary_loss_mlp": 0.01003182,
      "balance_loss_clip": 1.01025105,
      "balance_loss_mlp": 1.00205553,
      "epoch": 0.8016593518908195,
      "flos": 70397340675840.0,
      "grad_norm": 0.8571874662979818,
      "language_loss": 0.66202956,
      "learning_rate": 3.985144457422305e-07,
      "loss": 0.68270856,
      "num_input_tokens_seen": 143856345,
      "step": 6667,
      "time_per_iteration": 3.1302645206451416
    },
    {
      "auxiliary_loss_clip": 0.01167299,
      "auxiliary_loss_mlp": 0.0102592,
      "balance_loss_clip": 1.04591012,
      "balance_loss_mlp": 1.01871073,
      "epoch": 0.8017795947814585,
      "flos": 26026114688640.0,
      "grad_norm": 1.8282722862344438,
      "language_loss": 0.76581407,
      "learning_rate": 3.9804795670394096e-07,
      "loss": 0.78774631,
      "num_input_tokens_seen": 143876470,
      "step": 6668,
      "time_per_iteration": 2.5995113849639893
    },
    {
      "auxiliary_loss_clip": 0.01127793,
      "auxiliary_loss_mlp": 0.01021173,
      "balance_loss_clip": 1.04074919,
      "balance_loss_mlp": 1.01430368,
      "epoch": 0.8018998376720976,
      "flos": 22087181260800.0,
      "grad_norm": 1.63931249636291,
      "language_loss": 0.70664251,
      "learning_rate": 3.975817106805022e-07,
      "loss": 0.72813219,
      "num_input_tokens_seen": 143895170,
      "step": 6669,
      "time_per_iteration": 2.704141139984131
    },
    {
      "auxiliary_loss_clip": 0.01125367,
      "auxiliary_loss_mlp": 0.01025197,
      "balance_loss_clip": 1.0431006,
      "balance_loss_mlp": 1.01786613,
      "epoch": 0.8020200805627368,
      "flos": 34568545023360.0,
      "grad_norm": 3.314096642308104,
      "language_loss": 0.65041065,
      "learning_rate": 3.97115707742645e-07,
      "loss": 0.67191625,
      "num_input_tokens_seen": 143915845,
      "step": 6670,
      "time_per_iteration": 2.8983476161956787
    },
    {
      "auxiliary_loss_clip": 0.01142608,
      "auxiliary_loss_mlp": 0.01022773,
      "balance_loss_clip": 1.04613709,
      "balance_loss_mlp": 1.01595688,
      "epoch": 0.8021403234533758,
      "flos": 20120354196480.0,
      "grad_norm": 2.0501038992173344,
      "language_loss": 0.64725602,
      "learning_rate": 3.966499479610599e-07,
      "loss": 0.66890985,
      "num_input_tokens_seen": 143933940,
      "step": 6671,
      "time_per_iteration": 2.8100409507751465
    },
    {
      "auxiliary_loss_clip": 0.01122202,
      "auxiliary_loss_mlp": 0.01025394,
      "balance_loss_clip": 1.04365039,
      "balance_loss_mlp": 1.0186466,
      "epoch": 0.8022605663440149,
      "flos": 27746200252800.0,
      "grad_norm": 2.0580553244467317,
      "language_loss": 0.64835286,
      "learning_rate": 3.9618443140640225e-07,
      "loss": 0.66982883,
      "num_input_tokens_seen": 143952850,
      "step": 6672,
      "time_per_iteration": 2.8200061321258545
    },
    {
      "auxiliary_loss_clip": 0.01019357,
      "auxiliary_loss_mlp": 0.00999761,
      "balance_loss_clip": 1.00868118,
      "balance_loss_mlp": 0.998658,
      "epoch": 0.802380809234654,
      "flos": 60244998768000.0,
      "grad_norm": 0.9514401170998399,
      "language_loss": 0.51319474,
      "learning_rate": 3.957191581492918e-07,
      "loss": 0.53338593,
      "num_input_tokens_seen": 144013610,
      "step": 6673,
      "time_per_iteration": 3.3444461822509766
    },
    {
      "auxiliary_loss_clip": 0.01133597,
      "auxiliary_loss_mlp": 0.01028432,
      "balance_loss_clip": 1.04279375,
      "balance_loss_mlp": 1.02079701,
      "epoch": 0.8025010521252931,
      "flos": 15080722352640.0,
      "grad_norm": 3.042229916381196,
      "language_loss": 0.71183473,
      "learning_rate": 3.952541282603097e-07,
      "loss": 0.73345506,
      "num_input_tokens_seen": 144028715,
      "step": 6674,
      "time_per_iteration": 2.603052854537964
    },
    {
      "auxiliary_loss_clip": 0.01148643,
      "auxiliary_loss_mlp": 0.01023713,
      "balance_loss_clip": 1.04405761,
      "balance_loss_mlp": 1.01621485,
      "epoch": 0.8026212950159322,
      "flos": 22163527618560.0,
      "grad_norm": 1.976011786906928,
      "language_loss": 0.83579731,
      "learning_rate": 3.9478934181000013e-07,
      "loss": 0.85752082,
      "num_input_tokens_seen": 144048740,
      "step": 6675,
      "time_per_iteration": 2.678072214126587
    },
    {
      "auxiliary_loss_clip": 0.01169617,
      "auxiliary_loss_mlp": 0.01028351,
      "balance_loss_clip": 1.04609609,
      "balance_loss_mlp": 1.02086437,
      "epoch": 0.8027415379065713,
      "flos": 17675986792320.0,
      "grad_norm": 2.6470521414139885,
      "language_loss": 0.84771991,
      "learning_rate": 3.943247988688714e-07,
      "loss": 0.86969954,
      "num_input_tokens_seen": 144067435,
      "step": 6676,
      "time_per_iteration": 2.6244890689849854
    },
    {
      "auxiliary_loss_clip": 0.01151195,
      "auxiliary_loss_mlp": 0.01023232,
      "balance_loss_clip": 1.04206729,
      "balance_loss_mlp": 1.01641655,
      "epoch": 0.8028617807972104,
      "flos": 21979593048960.0,
      "grad_norm": 1.621980457380252,
      "language_loss": 0.72034276,
      "learning_rate": 3.938604995073933e-07,
      "loss": 0.74208701,
      "num_input_tokens_seen": 144085905,
      "step": 6677,
      "time_per_iteration": 2.666801929473877
    },
    {
      "auxiliary_loss_clip": 0.0114165,
      "auxiliary_loss_mlp": 0.01026746,
      "balance_loss_clip": 1.04435122,
      "balance_loss_mlp": 1.01930141,
      "epoch": 0.8029820236878494,
      "flos": 26428457905920.0,
      "grad_norm": 1.896965134520913,
      "language_loss": 0.65158582,
      "learning_rate": 3.9339644379600157e-07,
      "loss": 0.67326975,
      "num_input_tokens_seen": 144105735,
      "step": 6678,
      "time_per_iteration": 2.738889217376709
    },
    {
      "auxiliary_loss_clip": 0.01157797,
      "auxiliary_loss_mlp": 0.01025104,
      "balance_loss_clip": 1.04745066,
      "balance_loss_mlp": 1.01746583,
      "epoch": 0.8031022665784886,
      "flos": 17676489582720.0,
      "grad_norm": 2.318460260470543,
      "language_loss": 0.7157408,
      "learning_rate": 3.929326318050907e-07,
      "loss": 0.73756987,
      "num_input_tokens_seen": 144123405,
      "step": 6679,
      "time_per_iteration": 2.5747742652893066
    },
    {
      "auxiliary_loss_clip": 0.0116356,
      "auxiliary_loss_mlp": 0.01021488,
      "balance_loss_clip": 1.04461241,
      "balance_loss_mlp": 1.01428211,
      "epoch": 0.8032225094691277,
      "flos": 15450279431040.0,
      "grad_norm": 2.0258329194147024,
      "language_loss": 0.78872329,
      "learning_rate": 3.924690636050225e-07,
      "loss": 0.81057382,
      "num_input_tokens_seen": 144140815,
      "step": 6680,
      "time_per_iteration": 2.5574538707733154
    },
    {
      "auxiliary_loss_clip": 0.01154745,
      "auxiliary_loss_mlp": 0.01022324,
      "balance_loss_clip": 1.04568195,
      "balance_loss_mlp": 1.01467371,
      "epoch": 0.8033427523597667,
      "flos": 26179202453760.0,
      "grad_norm": 2.0826822266057534,
      "language_loss": 0.72938728,
      "learning_rate": 3.9200573926611915e-07,
      "loss": 0.75115794,
      "num_input_tokens_seen": 144162230,
      "step": 6681,
      "time_per_iteration": 2.6499826908111572
    },
    {
      "auxiliary_loss_clip": 0.01151886,
      "auxiliary_loss_mlp": 0.01026539,
      "balance_loss_clip": 1.0474658,
      "balance_loss_mlp": 1.01960444,
      "epoch": 0.8034629952504058,
      "flos": 21324905809920.0,
      "grad_norm": 1.8789600024984061,
      "language_loss": 0.73186529,
      "learning_rate": 3.9154265885866613e-07,
      "loss": 0.75364959,
      "num_input_tokens_seen": 144181540,
      "step": 6682,
      "time_per_iteration": 2.6929173469543457
    },
    {
      "auxiliary_loss_clip": 0.01153499,
      "auxiliary_loss_mlp": 0.0102535,
      "balance_loss_clip": 1.04752469,
      "balance_loss_mlp": 1.01787543,
      "epoch": 0.8035832381410449,
      "flos": 21651585027840.0,
      "grad_norm": 2.6816917408072922,
      "language_loss": 0.75016391,
      "learning_rate": 3.9107982245291394e-07,
      "loss": 0.77195239,
      "num_input_tokens_seen": 144199665,
      "step": 6683,
      "time_per_iteration": 2.5866661071777344
    },
    {
      "auxiliary_loss_clip": 0.0112534,
      "auxiliary_loss_mlp": 0.01026076,
      "balance_loss_clip": 1.04474306,
      "balance_loss_mlp": 1.0186553,
      "epoch": 0.803703481031684,
      "flos": 20518818744960.0,
      "grad_norm": 2.026007095002717,
      "language_loss": 0.77216923,
      "learning_rate": 3.9061723011907245e-07,
      "loss": 0.79368341,
      "num_input_tokens_seen": 144219020,
      "step": 6684,
      "time_per_iteration": 2.7521634101867676
    },
    {
      "auxiliary_loss_clip": 0.01138673,
      "auxiliary_loss_mlp": 0.01030288,
      "balance_loss_clip": 1.04344726,
      "balance_loss_mlp": 1.02258158,
      "epoch": 0.803823723922323,
      "flos": 22854807838080.0,
      "grad_norm": 1.677040812408557,
      "language_loss": 0.79368579,
      "learning_rate": 3.901548819273179e-07,
      "loss": 0.81537545,
      "num_input_tokens_seen": 144239035,
      "step": 6685,
      "time_per_iteration": 2.6913387775421143
    },
    {
      "auxiliary_loss_clip": 0.01152069,
      "auxiliary_loss_mlp": 0.01025199,
      "balance_loss_clip": 1.04500258,
      "balance_loss_mlp": 1.01807976,
      "epoch": 0.8039439668129622,
      "flos": 21362145235200.0,
      "grad_norm": 1.930173022697641,
      "language_loss": 0.69341147,
      "learning_rate": 3.896927779477881e-07,
      "loss": 0.71518415,
      "num_input_tokens_seen": 144258295,
      "step": 6686,
      "time_per_iteration": 3.53283429145813
    },
    {
      "auxiliary_loss_clip": 0.01126043,
      "auxiliary_loss_mlp": 0.01024285,
      "balance_loss_clip": 1.04132485,
      "balance_loss_mlp": 1.01703107,
      "epoch": 0.8040642097036013,
      "flos": 23802382575360.0,
      "grad_norm": 2.040405946396933,
      "language_loss": 0.67413282,
      "learning_rate": 3.892309182505833e-07,
      "loss": 0.69563609,
      "num_input_tokens_seen": 144276110,
      "step": 6687,
      "time_per_iteration": 3.792588233947754
    },
    {
      "auxiliary_loss_clip": 0.0116473,
      "auxiliary_loss_mlp": 0.01025649,
      "balance_loss_clip": 1.04519033,
      "balance_loss_mlp": 1.01840734,
      "epoch": 0.8041844525942403,
      "flos": 25922046009600.0,
      "grad_norm": 3.360773166055262,
      "language_loss": 0.86347687,
      "learning_rate": 3.887693029057675e-07,
      "loss": 0.88538063,
      "num_input_tokens_seen": 144295620,
      "step": 6688,
      "time_per_iteration": 2.632033109664917
    },
    {
      "auxiliary_loss_clip": 0.01140102,
      "auxiliary_loss_mlp": 0.01021857,
      "balance_loss_clip": 1.04388797,
      "balance_loss_mlp": 1.0154047,
      "epoch": 0.8043046954848795,
      "flos": 25191120153600.0,
      "grad_norm": 1.7126691095787838,
      "language_loss": 0.81258261,
      "learning_rate": 3.8830793198336684e-07,
      "loss": 0.83420217,
      "num_input_tokens_seen": 144315210,
      "step": 6689,
      "time_per_iteration": 2.7543301582336426
    },
    {
      "auxiliary_loss_clip": 0.0115853,
      "auxiliary_loss_mlp": 0.01029591,
      "balance_loss_clip": 1.04578221,
      "balance_loss_mlp": 1.02201533,
      "epoch": 0.8044249383755185,
      "flos": 41719185123840.0,
      "grad_norm": 1.8027025949090252,
      "language_loss": 0.70240974,
      "learning_rate": 3.878468055533721e-07,
      "loss": 0.72429097,
      "num_input_tokens_seen": 144337750,
      "step": 6690,
      "time_per_iteration": 3.7838265895843506
    },
    {
      "auxiliary_loss_clip": 0.01132499,
      "auxiliary_loss_mlp": 0.01039048,
      "balance_loss_clip": 1.04430759,
      "balance_loss_mlp": 1.0308733,
      "epoch": 0.8045451812661576,
      "flos": 20631434860800.0,
      "grad_norm": 4.042632290894405,
      "language_loss": 0.84840524,
      "learning_rate": 3.8738592368573464e-07,
      "loss": 0.8701207,
      "num_input_tokens_seen": 144355305,
      "step": 6691,
      "time_per_iteration": 2.751680850982666
    },
    {
      "auxiliary_loss_clip": 0.01117986,
      "auxiliary_loss_mlp": 0.01025877,
      "balance_loss_clip": 1.04215741,
      "balance_loss_mlp": 1.01868248,
      "epoch": 0.8046654241567968,
      "flos": 29711806254720.0,
      "grad_norm": 2.096202333294587,
      "language_loss": 0.88072956,
      "learning_rate": 3.8692528645037137e-07,
      "loss": 0.90216815,
      "num_input_tokens_seen": 144374485,
      "step": 6692,
      "time_per_iteration": 3.6693036556243896
    },
    {
      "auxiliary_loss_clip": 0.01166649,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.04721117,
      "balance_loss_mlp": 1.02024925,
      "epoch": 0.8047856670474358,
      "flos": 17671389851520.0,
      "grad_norm": 2.14913359985059,
      "language_loss": 0.77669942,
      "learning_rate": 3.8646489391715907e-07,
      "loss": 0.79864144,
      "num_input_tokens_seen": 144388780,
      "step": 6693,
      "time_per_iteration": 2.5592777729034424
    },
    {
      "auxiliary_loss_clip": 0.01140112,
      "auxiliary_loss_mlp": 0.01027556,
      "balance_loss_clip": 1.04379141,
      "balance_loss_mlp": 1.01989043,
      "epoch": 0.8049059099380749,
      "flos": 17120699464320.0,
      "grad_norm": 2.333836135212007,
      "language_loss": 0.8795982,
      "learning_rate": 3.8600474615593903e-07,
      "loss": 0.90127486,
      "num_input_tokens_seen": 144403395,
      "step": 6694,
      "time_per_iteration": 2.6118319034576416
    },
    {
      "auxiliary_loss_clip": 0.0103498,
      "auxiliary_loss_mlp": 0.01000808,
      "balance_loss_clip": 1.01028836,
      "balance_loss_mlp": 0.99972898,
      "epoch": 0.805026152828714,
      "flos": 62212903240320.0,
      "grad_norm": 0.7858853051697917,
      "language_loss": 0.59614527,
      "learning_rate": 3.8554484323651605e-07,
      "loss": 0.61650318,
      "num_input_tokens_seen": 144465265,
      "step": 6695,
      "time_per_iteration": 3.3556320667266846
    },
    {
      "auxiliary_loss_clip": 0.01152528,
      "auxiliary_loss_mlp": 0.00761805,
      "balance_loss_clip": 1.04712987,
      "balance_loss_mlp": 1.00026059,
      "epoch": 0.8051463957193531,
      "flos": 21688608971520.0,
      "grad_norm": 2.0519758684973843,
      "language_loss": 0.79214931,
      "learning_rate": 3.85085185228657e-07,
      "loss": 0.81129253,
      "num_input_tokens_seen": 144484235,
      "step": 6696,
      "time_per_iteration": 2.6618385314941406
    },
    {
      "auxiliary_loss_clip": 0.01132417,
      "auxiliary_loss_mlp": 0.01028133,
      "balance_loss_clip": 1.04301572,
      "balance_loss_mlp": 1.02011645,
      "epoch": 0.8052666386099921,
      "flos": 32051458535040.0,
      "grad_norm": 2.030838826254038,
      "language_loss": 0.73146808,
      "learning_rate": 3.8462577220209114e-07,
      "loss": 0.75307357,
      "num_input_tokens_seen": 144504610,
      "step": 6697,
      "time_per_iteration": 2.8052375316619873
    },
    {
      "auxiliary_loss_clip": 0.01064558,
      "auxiliary_loss_mlp": 0.01002805,
      "balance_loss_clip": 1.0101881,
      "balance_loss_mlp": 1.00176167,
      "epoch": 0.8053868815006313,
      "flos": 67157875768320.0,
      "grad_norm": 0.7044477038079189,
      "language_loss": 0.58986294,
      "learning_rate": 3.8416660422651127e-07,
      "loss": 0.61053658,
      "num_input_tokens_seen": 144574260,
      "step": 6698,
      "time_per_iteration": 3.3441967964172363
    },
    {
      "auxiliary_loss_clip": 0.01127679,
      "auxiliary_loss_mlp": 0.01024281,
      "balance_loss_clip": 1.04054952,
      "balance_loss_mlp": 1.01704752,
      "epoch": 0.8055071243912704,
      "flos": 23837000307840.0,
      "grad_norm": 1.9639227705355988,
      "language_loss": 0.68217838,
      "learning_rate": 3.837076813715723e-07,
      "loss": 0.70369798,
      "num_input_tokens_seen": 144594145,
      "step": 6699,
      "time_per_iteration": 2.7055587768554688
    },
    {
      "auxiliary_loss_clip": 0.01122887,
      "auxiliary_loss_mlp": 0.01029587,
      "balance_loss_clip": 1.04019153,
      "balance_loss_mlp": 1.0210216,
      "epoch": 0.8056273672819094,
      "flos": 21324510760320.0,
      "grad_norm": 2.5556585898206916,
      "language_loss": 0.75173509,
      "learning_rate": 3.832490037068941e-07,
      "loss": 0.77325982,
      "num_input_tokens_seen": 144612935,
      "step": 6700,
      "time_per_iteration": 2.747830629348755
    },
    {
      "auxiliary_loss_clip": 0.01094081,
      "auxiliary_loss_mlp": 0.0102447,
      "balance_loss_clip": 1.03850842,
      "balance_loss_mlp": 1.01738954,
      "epoch": 0.8057476101725486,
      "flos": 25768383626880.0,
      "grad_norm": 1.8667292132826152,
      "language_loss": 0.75965178,
      "learning_rate": 3.827905713020554e-07,
      "loss": 0.7808373,
      "num_input_tokens_seen": 144630580,
      "step": 6701,
      "time_per_iteration": 2.8565828800201416
    },
    {
      "auxiliary_loss_clip": 0.01127833,
      "auxiliary_loss_mlp": 0.01025691,
      "balance_loss_clip": 1.03966975,
      "balance_loss_mlp": 1.01713753,
      "epoch": 0.8058678530631876,
      "flos": 24535283679360.0,
      "grad_norm": 2.007475077175213,
      "language_loss": 0.69141352,
      "learning_rate": 3.823323842266017e-07,
      "loss": 0.7129488,
      "num_input_tokens_seen": 144649975,
      "step": 6702,
      "time_per_iteration": 2.710167169570923
    },
    {
      "auxiliary_loss_clip": 0.0115359,
      "auxiliary_loss_mlp": 0.0101959,
      "balance_loss_clip": 1.04205835,
      "balance_loss_mlp": 1.01238382,
      "epoch": 0.8059880959538267,
      "flos": 24753728240640.0,
      "grad_norm": 2.424771603400656,
      "language_loss": 0.73059636,
      "learning_rate": 3.818744425500393e-07,
      "loss": 0.75232816,
      "num_input_tokens_seen": 144667990,
      "step": 6703,
      "time_per_iteration": 2.724198579788208
    },
    {
      "auxiliary_loss_clip": 0.01116741,
      "auxiliary_loss_mlp": 0.01025207,
      "balance_loss_clip": 1.03865385,
      "balance_loss_mlp": 1.0179472,
      "epoch": 0.8061083388444659,
      "flos": 22196349671040.0,
      "grad_norm": 1.8085836149025707,
      "language_loss": 0.8053444,
      "learning_rate": 3.8141674634183675e-07,
      "loss": 0.82676393,
      "num_input_tokens_seen": 144687020,
      "step": 6704,
      "time_per_iteration": 2.7691078186035156
    },
    {
      "auxiliary_loss_clip": 0.01109193,
      "auxiliary_loss_mlp": 0.01027808,
      "balance_loss_clip": 1.04166269,
      "balance_loss_mlp": 1.02137399,
      "epoch": 0.8062285817351049,
      "flos": 30044195735040.0,
      "grad_norm": 1.761189382591428,
      "language_loss": 0.66088951,
      "learning_rate": 3.809592956714278e-07,
      "loss": 0.6822595,
      "num_input_tokens_seen": 144710255,
      "step": 6705,
      "time_per_iteration": 2.795830011367798
    },
    {
      "auxiliary_loss_clip": 0.01158905,
      "auxiliary_loss_mlp": 0.01025814,
      "balance_loss_clip": 1.04744077,
      "balance_loss_mlp": 1.01832771,
      "epoch": 0.806348824625744,
      "flos": 22782591544320.0,
      "grad_norm": 8.558763051492198,
      "language_loss": 0.74781334,
      "learning_rate": 3.805020906082057e-07,
      "loss": 0.76966053,
      "num_input_tokens_seen": 144728830,
      "step": 6706,
      "time_per_iteration": 2.667769432067871
    },
    {
      "auxiliary_loss_clip": 0.01143106,
      "auxiliary_loss_mlp": 0.01030836,
      "balance_loss_clip": 1.04450202,
      "balance_loss_mlp": 1.02218723,
      "epoch": 0.8064690675163831,
      "flos": 23404600385280.0,
      "grad_norm": 3.6560542782517635,
      "language_loss": 0.80786294,
      "learning_rate": 3.8004513122152917e-07,
      "loss": 0.82960236,
      "num_input_tokens_seen": 144747140,
      "step": 6707,
      "time_per_iteration": 2.6309080123901367
    },
    {
      "auxiliary_loss_clip": 0.01129634,
      "auxiliary_loss_mlp": 0.01028291,
      "balance_loss_clip": 1.04481447,
      "balance_loss_mlp": 1.02124834,
      "epoch": 0.8065893104070222,
      "flos": 24060903736320.0,
      "grad_norm": 5.7532097711745385,
      "language_loss": 0.67128372,
      "learning_rate": 3.79588417580718e-07,
      "loss": 0.69286299,
      "num_input_tokens_seen": 144765250,
      "step": 6708,
      "time_per_iteration": 2.7516837120056152
    },
    {
      "auxiliary_loss_clip": 0.01157029,
      "auxiliary_loss_mlp": 0.01024596,
      "balance_loss_clip": 1.04798555,
      "balance_loss_mlp": 1.017169,
      "epoch": 0.8067095532976613,
      "flos": 22305410340480.0,
      "grad_norm": 1.9254754871975799,
      "language_loss": 0.76563334,
      "learning_rate": 3.791319497550558e-07,
      "loss": 0.7874496,
      "num_input_tokens_seen": 144783080,
      "step": 6709,
      "time_per_iteration": 2.616316318511963
    },
    {
      "auxiliary_loss_clip": 0.01132227,
      "auxiliary_loss_mlp": 0.00761978,
      "balance_loss_clip": 1.04366863,
      "balance_loss_mlp": 1.0002985,
      "epoch": 0.8068297961883004,
      "flos": 17129498296320.0,
      "grad_norm": 4.771874240411894,
      "language_loss": 0.7093364,
      "learning_rate": 3.78675727813788e-07,
      "loss": 0.7282784,
      "num_input_tokens_seen": 144800645,
      "step": 6710,
      "time_per_iteration": 2.6581382751464844
    },
    {
      "auxiliary_loss_clip": 0.01138666,
      "auxiliary_loss_mlp": 0.01025977,
      "balance_loss_clip": 1.04421604,
      "balance_loss_mlp": 1.01866639,
      "epoch": 0.8069500390789395,
      "flos": 22018843635840.0,
      "grad_norm": 1.6545849480560237,
      "language_loss": 0.73449981,
      "learning_rate": 3.782197518261225e-07,
      "loss": 0.75614619,
      "num_input_tokens_seen": 144820085,
      "step": 6711,
      "time_per_iteration": 2.6776797771453857
    },
    {
      "auxiliary_loss_clip": 0.01144127,
      "auxiliary_loss_mlp": 0.01025292,
      "balance_loss_clip": 1.04358244,
      "balance_loss_mlp": 1.0178802,
      "epoch": 0.8070702819695785,
      "flos": 19244241567360.0,
      "grad_norm": 2.4369533078048327,
      "language_loss": 0.95764345,
      "learning_rate": 3.777640218612319e-07,
      "loss": 0.97933763,
      "num_input_tokens_seen": 144838070,
      "step": 6712,
      "time_per_iteration": 3.5909345149993896
    },
    {
      "auxiliary_loss_clip": 0.01147924,
      "auxiliary_loss_mlp": 0.01022455,
      "balance_loss_clip": 1.04475319,
      "balance_loss_mlp": 1.01553154,
      "epoch": 0.8071905248602176,
      "flos": 21544320038400.0,
      "grad_norm": 2.4524030696783274,
      "language_loss": 0.71657729,
      "learning_rate": 3.773085379882488e-07,
      "loss": 0.73828101,
      "num_input_tokens_seen": 144857125,
      "step": 6713,
      "time_per_iteration": 2.6792399883270264
    },
    {
      "auxiliary_loss_clip": 0.01151621,
      "auxiliary_loss_mlp": 0.00762349,
      "balance_loss_clip": 1.04198134,
      "balance_loss_mlp": 1.00029445,
      "epoch": 0.8073107677508568,
      "flos": 37268309105280.0,
      "grad_norm": 1.9460315691344214,
      "language_loss": 0.76351786,
      "learning_rate": 3.768533002762715e-07,
      "loss": 0.78265762,
      "num_input_tokens_seen": 144880660,
      "step": 6714,
      "time_per_iteration": 3.8547730445861816
    },
    {
      "auxiliary_loss_clip": 0.01137919,
      "auxiliary_loss_mlp": 0.01028433,
      "balance_loss_clip": 1.04076064,
      "balance_loss_mlp": 1.02131629,
      "epoch": 0.8074310106414958,
      "flos": 28366269759360.0,
      "grad_norm": 1.650601631541225,
      "language_loss": 0.7687183,
      "learning_rate": 3.763983087943572e-07,
      "loss": 0.79038185,
      "num_input_tokens_seen": 144900050,
      "step": 6715,
      "time_per_iteration": 2.7230751514434814
    },
    {
      "auxiliary_loss_clip": 0.01145864,
      "auxiliary_loss_mlp": 0.0076205,
      "balance_loss_clip": 1.04383707,
      "balance_loss_mlp": 1.00029075,
      "epoch": 0.8075512535321349,
      "flos": 24281646768000.0,
      "grad_norm": 1.6440954139173636,
      "language_loss": 0.81088781,
      "learning_rate": 3.759435636115282e-07,
      "loss": 0.82996696,
      "num_input_tokens_seen": 144920835,
      "step": 6716,
      "time_per_iteration": 3.5402698516845703
    },
    {
      "auxiliary_loss_clip": 0.01090542,
      "auxiliary_loss_mlp": 0.00762236,
      "balance_loss_clip": 1.03979897,
      "balance_loss_mlp": 1.00030351,
      "epoch": 0.807671496422774,
      "flos": 26030855283840.0,
      "grad_norm": 2.0619252526920473,
      "language_loss": 0.73218077,
      "learning_rate": 3.7548906479676967e-07,
      "loss": 0.75070846,
      "num_input_tokens_seen": 144940430,
      "step": 6717,
      "time_per_iteration": 3.7854037284851074
    },
    {
      "auxiliary_loss_clip": 0.01156252,
      "auxiliary_loss_mlp": 0.01026961,
      "balance_loss_clip": 1.04444408,
      "balance_loss_mlp": 1.01875377,
      "epoch": 0.8077917393134131,
      "flos": 23730740899200.0,
      "grad_norm": 1.987188786642795,
      "language_loss": 0.7184993,
      "learning_rate": 3.7503481241902855e-07,
      "loss": 0.74033141,
      "num_input_tokens_seen": 144960405,
      "step": 6718,
      "time_per_iteration": 2.7212939262390137
    },
    {
      "auxiliary_loss_clip": 0.01138397,
      "auxiliary_loss_mlp": 0.00761992,
      "balance_loss_clip": 1.04276919,
      "balance_loss_mlp": 1.00030804,
      "epoch": 0.8079119822040521,
      "flos": 18402028398720.0,
      "grad_norm": 1.726930422083057,
      "language_loss": 0.80267918,
      "learning_rate": 3.745808065472145e-07,
      "loss": 0.82168305,
      "num_input_tokens_seen": 144977700,
      "step": 6719,
      "time_per_iteration": 2.704171895980835
    },
    {
      "auxiliary_loss_clip": 0.01149326,
      "auxiliary_loss_mlp": 0.01023754,
      "balance_loss_clip": 1.04778886,
      "balance_loss_mlp": 1.01647639,
      "epoch": 0.8080322250946913,
      "flos": 23621787970560.0,
      "grad_norm": 1.6201188853162733,
      "language_loss": 0.76063275,
      "learning_rate": 3.741270472501994e-07,
      "loss": 0.78236347,
      "num_input_tokens_seen": 144998340,
      "step": 6720,
      "time_per_iteration": 2.609226942062378
    },
    {
      "auxiliary_loss_clip": 0.011374,
      "auxiliary_loss_mlp": 0.01030645,
      "balance_loss_clip": 1.0452373,
      "balance_loss_mlp": 1.02391529,
      "epoch": 0.8081524679853304,
      "flos": 22820692896000.0,
      "grad_norm": 2.045703426244612,
      "language_loss": 0.72659254,
      "learning_rate": 3.736735345968183e-07,
      "loss": 0.74827302,
      "num_input_tokens_seen": 145017950,
      "step": 6721,
      "time_per_iteration": 2.631300210952759
    },
    {
      "auxiliary_loss_clip": 0.01154941,
      "auxiliary_loss_mlp": 0.0102374,
      "balance_loss_clip": 1.04708576,
      "balance_loss_mlp": 1.01673388,
      "epoch": 0.8082727108759694,
      "flos": 17640004343040.0,
      "grad_norm": 1.6119000616671963,
      "language_loss": 0.78926432,
      "learning_rate": 3.7322026865586986e-07,
      "loss": 0.81105113,
      "num_input_tokens_seen": 145036985,
      "step": 6722,
      "time_per_iteration": 2.6055054664611816
    },
    {
      "auxiliary_loss_clip": 0.0115989,
      "auxiliary_loss_mlp": 0.01027179,
      "balance_loss_clip": 1.04925394,
      "balance_loss_mlp": 1.01948977,
      "epoch": 0.8083929537666086,
      "flos": 25958172113280.0,
      "grad_norm": 2.0934348913813676,
      "language_loss": 0.73665059,
      "learning_rate": 3.7276724949611206e-07,
      "loss": 0.7585212,
      "num_input_tokens_seen": 145057095,
      "step": 6723,
      "time_per_iteration": 2.6585187911987305
    },
    {
      "auxiliary_loss_clip": 0.01144313,
      "auxiliary_loss_mlp": 0.01024472,
      "balance_loss_clip": 1.04570675,
      "balance_loss_mlp": 1.01662254,
      "epoch": 0.8085131966572476,
      "flos": 27089178629760.0,
      "grad_norm": 2.5045521792052488,
      "language_loss": 0.75154835,
      "learning_rate": 3.723144771862694e-07,
      "loss": 0.77323616,
      "num_input_tokens_seen": 145077735,
      "step": 6724,
      "time_per_iteration": 2.6524457931518555
    },
    {
      "auxiliary_loss_clip": 0.01128678,
      "auxiliary_loss_mlp": 0.01028739,
      "balance_loss_clip": 1.04312181,
      "balance_loss_mlp": 1.02108014,
      "epoch": 0.8086334395478867,
      "flos": 23988543788160.0,
      "grad_norm": 9.389125020229514,
      "language_loss": 0.76955557,
      "learning_rate": 3.718619517950263e-07,
      "loss": 0.79112971,
      "num_input_tokens_seen": 145098330,
      "step": 6725,
      "time_per_iteration": 2.7643635272979736
    },
    {
      "auxiliary_loss_clip": 0.01167739,
      "auxiliary_loss_mlp": 0.01026911,
      "balance_loss_clip": 1.04878473,
      "balance_loss_mlp": 1.0196507,
      "epoch": 0.8087536824385259,
      "flos": 20405879406720.0,
      "grad_norm": 2.070930702130001,
      "language_loss": 0.77393061,
      "learning_rate": 3.714096733910301e-07,
      "loss": 0.7958771,
      "num_input_tokens_seen": 145115855,
      "step": 6726,
      "time_per_iteration": 2.6047353744506836
    },
    {
      "auxiliary_loss_clip": 0.01163768,
      "auxiliary_loss_mlp": 0.01035056,
      "balance_loss_clip": 1.04878199,
      "balance_loss_mlp": 1.02675915,
      "epoch": 0.8088739253291649,
      "flos": 25919639798400.0,
      "grad_norm": 2.539625950193938,
      "language_loss": 0.69968915,
      "learning_rate": 3.709576420428926e-07,
      "loss": 0.72167736,
      "num_input_tokens_seen": 145136655,
      "step": 6727,
      "time_per_iteration": 2.7307746410369873
    },
    {
      "auxiliary_loss_clip": 0.01140689,
      "auxiliary_loss_mlp": 0.01023313,
      "balance_loss_clip": 1.04243803,
      "balance_loss_mlp": 1.01609826,
      "epoch": 0.808994168219804,
      "flos": 28402072640640.0,
      "grad_norm": 2.984272264057401,
      "language_loss": 0.73046154,
      "learning_rate": 3.7050585781918463e-07,
      "loss": 0.75210154,
      "num_input_tokens_seen": 145156955,
      "step": 6728,
      "time_per_iteration": 2.790555953979492
    },
    {
      "auxiliary_loss_clip": 0.01156501,
      "auxiliary_loss_mlp": 0.01026021,
      "balance_loss_clip": 1.04507196,
      "balance_loss_mlp": 1.01851082,
      "epoch": 0.8091144111104431,
      "flos": 17421056991360.0,
      "grad_norm": 2.272916350716052,
      "language_loss": 0.6869809,
      "learning_rate": 3.700543207884428e-07,
      "loss": 0.70880616,
      "num_input_tokens_seen": 145173865,
      "step": 6729,
      "time_per_iteration": 2.6495230197906494
    },
    {
      "auxiliary_loss_clip": 0.01150206,
      "auxiliary_loss_mlp": 0.01020785,
      "balance_loss_clip": 1.04368329,
      "balance_loss_mlp": 1.01371312,
      "epoch": 0.8092346540010822,
      "flos": 32153803361280.0,
      "grad_norm": 1.8686169143089237,
      "language_loss": 0.70947027,
      "learning_rate": 3.6960303101916466e-07,
      "loss": 0.73118019,
      "num_input_tokens_seen": 145193780,
      "step": 6730,
      "time_per_iteration": 2.744354486465454
    },
    {
      "auxiliary_loss_clip": 0.01064991,
      "auxiliary_loss_mlp": 0.00753638,
      "balance_loss_clip": 1.01078701,
      "balance_loss_mlp": 1.000247,
      "epoch": 0.8093548968917212,
      "flos": 58035093390720.0,
      "grad_norm": 0.7413906452416251,
      "language_loss": 0.55534536,
      "learning_rate": 3.6915198857981047e-07,
      "loss": 0.57353163,
      "num_input_tokens_seen": 145258980,
      "step": 6731,
      "time_per_iteration": 3.2873332500457764
    },
    {
      "auxiliary_loss_clip": 0.01120964,
      "auxiliary_loss_mlp": 0.0102323,
      "balance_loss_clip": 1.04180217,
      "balance_loss_mlp": 1.01587725,
      "epoch": 0.8094751397823604,
      "flos": 27381599251200.0,
      "grad_norm": 1.6238463886999253,
      "language_loss": 0.67773122,
      "learning_rate": 3.687011935388027e-07,
      "loss": 0.69917315,
      "num_input_tokens_seen": 145281875,
      "step": 6732,
      "time_per_iteration": 2.810391426086426
    },
    {
      "auxiliary_loss_clip": 0.01151766,
      "auxiliary_loss_mlp": 0.01020987,
      "balance_loss_clip": 1.04537225,
      "balance_loss_mlp": 1.0139327,
      "epoch": 0.8095953826729995,
      "flos": 24061083304320.0,
      "grad_norm": 1.9776199233314524,
      "language_loss": 0.72888982,
      "learning_rate": 3.6825064596452646e-07,
      "loss": 0.75061738,
      "num_input_tokens_seen": 145302220,
      "step": 6733,
      "time_per_iteration": 2.6122918128967285
    },
    {
      "auxiliary_loss_clip": 0.01151905,
      "auxiliary_loss_mlp": 0.01024893,
      "balance_loss_clip": 1.04301095,
      "balance_loss_mlp": 1.01825595,
      "epoch": 0.8097156255636385,
      "flos": 23951412103680.0,
      "grad_norm": 1.8562402281421826,
      "language_loss": 0.70700181,
      "learning_rate": 3.678003459253305e-07,
      "loss": 0.72876978,
      "num_input_tokens_seen": 145323070,
      "step": 6734,
      "time_per_iteration": 2.7278125286102295
    },
    {
      "auxiliary_loss_clip": 0.01124187,
      "auxiliary_loss_mlp": 0.01030044,
      "balance_loss_clip": 1.0424782,
      "balance_loss_mlp": 1.02271271,
      "epoch": 0.8098358684542777,
      "flos": 21799142098560.0,
      "grad_norm": 2.186608896382513,
      "language_loss": 0.74018025,
      "learning_rate": 3.673502934895236e-07,
      "loss": 0.76172256,
      "num_input_tokens_seen": 145342575,
      "step": 6735,
      "time_per_iteration": 2.6809206008911133
    },
    {
      "auxiliary_loss_clip": 0.01064295,
      "auxiliary_loss_mlp": 0.01000882,
      "balance_loss_clip": 1.0102421,
      "balance_loss_mlp": 0.99979132,
      "epoch": 0.8099561113449167,
      "flos": 68809515966720.0,
      "grad_norm": 0.6897237886489156,
      "language_loss": 0.57953131,
      "learning_rate": 3.669004887253802e-07,
      "loss": 0.60018313,
      "num_input_tokens_seen": 145408865,
      "step": 6736,
      "time_per_iteration": 3.3543224334716797
    },
    {
      "auxiliary_loss_clip": 0.01142986,
      "auxiliary_loss_mlp": 0.01023177,
      "balance_loss_clip": 1.04585862,
      "balance_loss_mlp": 1.01620924,
      "epoch": 0.8100763542355558,
      "flos": 23586056916480.0,
      "grad_norm": 1.5333678418541534,
      "language_loss": 0.78947115,
      "learning_rate": 3.664509317011335e-07,
      "loss": 0.81113279,
      "num_input_tokens_seen": 145429200,
      "step": 6737,
      "time_per_iteration": 2.699428081512451
    },
    {
      "auxiliary_loss_clip": 0.01155156,
      "auxiliary_loss_mlp": 0.0102535,
      "balance_loss_clip": 1.04902315,
      "balance_loss_mlp": 1.01784563,
      "epoch": 0.810196597126195,
      "flos": 31650408207360.0,
      "grad_norm": 1.8326739942157924,
      "language_loss": 0.73623377,
      "learning_rate": 3.6600162248498134e-07,
      "loss": 0.75803888,
      "num_input_tokens_seen": 145452830,
      "step": 6738,
      "time_per_iteration": 3.614037275314331
    },
    {
      "auxiliary_loss_clip": 0.01076915,
      "auxiliary_loss_mlp": 0.01025964,
      "balance_loss_clip": 1.03411484,
      "balance_loss_mlp": 1.01913023,
      "epoch": 0.810316840016834,
      "flos": 24900459298560.0,
      "grad_norm": 1.7875238942546223,
      "language_loss": 0.76167393,
      "learning_rate": 3.6555256114508426e-07,
      "loss": 0.78270268,
      "num_input_tokens_seen": 145472625,
      "step": 6739,
      "time_per_iteration": 2.786972761154175
    },
    {
      "auxiliary_loss_clip": 0.01138576,
      "auxiliary_loss_mlp": 0.01034726,
      "balance_loss_clip": 1.03992498,
      "balance_loss_mlp": 1.02711725,
      "epoch": 0.8104370829074731,
      "flos": 27965003950080.0,
      "grad_norm": 2.0184615393574994,
      "language_loss": 0.72946858,
      "learning_rate": 3.651037477495642e-07,
      "loss": 0.75120163,
      "num_input_tokens_seen": 145494075,
      "step": 6740,
      "time_per_iteration": 3.701734781265259
    },
    {
      "auxiliary_loss_clip": 0.01165754,
      "auxiliary_loss_mlp": 0.01030626,
      "balance_loss_clip": 1.04509497,
      "balance_loss_mlp": 1.02344394,
      "epoch": 0.8105573257981122,
      "flos": 24640752988800.0,
      "grad_norm": 2.2612466036966894,
      "language_loss": 0.67987198,
      "learning_rate": 3.6465518236650584e-07,
      "loss": 0.70183581,
      "num_input_tokens_seen": 145514220,
      "step": 6741,
      "time_per_iteration": 2.730714797973633
    },
    {
      "auxiliary_loss_clip": 0.01125416,
      "auxiliary_loss_mlp": 0.01021193,
      "balance_loss_clip": 1.04226601,
      "balance_loss_mlp": 1.01451421,
      "epoch": 0.8106775686887513,
      "flos": 26358935132160.0,
      "grad_norm": 1.9101157888147844,
      "language_loss": 0.78239787,
      "learning_rate": 3.642068650639558e-07,
      "loss": 0.80386394,
      "num_input_tokens_seen": 145533965,
      "step": 6742,
      "time_per_iteration": 3.628936767578125
    },
    {
      "auxiliary_loss_clip": 0.01131323,
      "auxiliary_loss_mlp": 0.01024899,
      "balance_loss_clip": 1.03879547,
      "balance_loss_mlp": 1.01782143,
      "epoch": 0.8107978115793903,
      "flos": 27271892136960.0,
      "grad_norm": 2.3403319295773026,
      "language_loss": 0.64612436,
      "learning_rate": 3.6375879590992334e-07,
      "loss": 0.66768658,
      "num_input_tokens_seen": 145554310,
      "step": 6743,
      "time_per_iteration": 3.655266284942627
    },
    {
      "auxiliary_loss_clip": 0.01134313,
      "auxiliary_loss_mlp": 0.01030752,
      "balance_loss_clip": 1.04222083,
      "balance_loss_mlp": 1.02359927,
      "epoch": 0.8109180544700295,
      "flos": 24934322845440.0,
      "grad_norm": 2.0379656965539934,
      "language_loss": 0.80922627,
      "learning_rate": 3.6331097497238173e-07,
      "loss": 0.83087695,
      "num_input_tokens_seen": 145573755,
      "step": 6744,
      "time_per_iteration": 2.702237129211426
    },
    {
      "auxiliary_loss_clip": 0.01123114,
      "auxiliary_loss_mlp": 0.01019525,
      "balance_loss_clip": 1.04154289,
      "balance_loss_mlp": 1.01274788,
      "epoch": 0.8110382973606686,
      "flos": 21105383840640.0,
      "grad_norm": 1.9017473837866932,
      "language_loss": 0.79699701,
      "learning_rate": 3.628634023192627e-07,
      "loss": 0.81842339,
      "num_input_tokens_seen": 145594000,
      "step": 6745,
      "time_per_iteration": 2.7435286045074463
    },
    {
      "auxiliary_loss_clip": 0.01153616,
      "auxiliary_loss_mlp": 0.01026671,
      "balance_loss_clip": 1.0446564,
      "balance_loss_mlp": 1.01913095,
      "epoch": 0.8111585402513076,
      "flos": 15414081500160.0,
      "grad_norm": 2.0425402172120264,
      "language_loss": 0.7547375,
      "learning_rate": 3.624160780184644e-07,
      "loss": 0.77654028,
      "num_input_tokens_seen": 145611215,
      "step": 6746,
      "time_per_iteration": 2.615886688232422
    },
    {
      "auxiliary_loss_clip": 0.01132685,
      "auxiliary_loss_mlp": 0.0102173,
      "balance_loss_clip": 1.04288495,
      "balance_loss_mlp": 1.0142436,
      "epoch": 0.8112787831419467,
      "flos": 24095736950400.0,
      "grad_norm": 2.0194326229268253,
      "language_loss": 0.74667335,
      "learning_rate": 3.6196900213784496e-07,
      "loss": 0.76821744,
      "num_input_tokens_seen": 145630530,
      "step": 6747,
      "time_per_iteration": 2.6693856716156006
    },
    {
      "auxiliary_loss_clip": 0.01154651,
      "auxiliary_loss_mlp": 0.01024483,
      "balance_loss_clip": 1.04503155,
      "balance_loss_mlp": 1.01764619,
      "epoch": 0.8113990260325858,
      "flos": 20483374999680.0,
      "grad_norm": 1.8730010996656736,
      "language_loss": 0.86623436,
      "learning_rate": 3.6152217474522527e-07,
      "loss": 0.88802576,
      "num_input_tokens_seen": 145647345,
      "step": 6748,
      "time_per_iteration": 2.6099674701690674
    },
    {
      "auxiliary_loss_clip": 0.01155692,
      "auxiliary_loss_mlp": 0.01026449,
      "balance_loss_clip": 1.04887307,
      "balance_loss_mlp": 1.01911736,
      "epoch": 0.8115192689232249,
      "flos": 24901141656960.0,
      "grad_norm": 1.8014404876793537,
      "language_loss": 0.73082066,
      "learning_rate": 3.6107559590838975e-07,
      "loss": 0.75264204,
      "num_input_tokens_seen": 145666330,
      "step": 6749,
      "time_per_iteration": 2.6364614963531494
    },
    {
      "auxiliary_loss_clip": 0.01092465,
      "auxiliary_loss_mlp": 0.01026997,
      "balance_loss_clip": 1.03848636,
      "balance_loss_mlp": 1.01960278,
      "epoch": 0.811639511813864,
      "flos": 24057204635520.0,
      "grad_norm": 2.6382949896127346,
      "language_loss": 0.66563833,
      "learning_rate": 3.606292656950822e-07,
      "loss": 0.6868329,
      "num_input_tokens_seen": 145684740,
      "step": 6750,
      "time_per_iteration": 2.7730424404144287
    },
    {
      "auxiliary_loss_clip": 0.01132285,
      "auxiliary_loss_mlp": 0.01030064,
      "balance_loss_clip": 1.04062319,
      "balance_loss_mlp": 1.02204108,
      "epoch": 0.8117597547045031,
      "flos": 23185150243200.0,
      "grad_norm": 2.092806495273232,
      "language_loss": 0.86498737,
      "learning_rate": 3.601831841730121e-07,
      "loss": 0.88661087,
      "num_input_tokens_seen": 145702660,
      "step": 6751,
      "time_per_iteration": 2.6612370014190674
    },
    {
      "auxiliary_loss_clip": 0.01150059,
      "auxiliary_loss_mlp": 0.01026838,
      "balance_loss_clip": 1.0446775,
      "balance_loss_mlp": 1.01977825,
      "epoch": 0.8118799975951422,
      "flos": 23040250778880.0,
      "grad_norm": 2.150153526122874,
      "language_loss": 0.72546262,
      "learning_rate": 3.5973735140984916e-07,
      "loss": 0.7472316,
      "num_input_tokens_seen": 145722830,
      "step": 6752,
      "time_per_iteration": 2.755699396133423
    },
    {
      "auxiliary_loss_clip": 0.01106616,
      "auxiliary_loss_mlp": 0.00761935,
      "balance_loss_clip": 1.03873062,
      "balance_loss_mlp": 1.00025082,
      "epoch": 0.8120002404857812,
      "flos": 24639962889600.0,
      "grad_norm": 2.1479293726628335,
      "language_loss": 0.7953099,
      "learning_rate": 3.5929176747322607e-07,
      "loss": 0.81399548,
      "num_input_tokens_seen": 145741935,
      "step": 6753,
      "time_per_iteration": 2.749873161315918
    },
    {
      "auxiliary_loss_clip": 0.01047047,
      "auxiliary_loss_mlp": 0.01000933,
      "balance_loss_clip": 1.01043868,
      "balance_loss_mlp": 0.99990767,
      "epoch": 0.8121204833764204,
      "flos": 57415742156160.0,
      "grad_norm": 0.8098266423300995,
      "language_loss": 0.56176656,
      "learning_rate": 3.588464324307372e-07,
      "loss": 0.58224636,
      "num_input_tokens_seen": 145805560,
      "step": 6754,
      "time_per_iteration": 3.342338800430298
    },
    {
      "auxiliary_loss_clip": 0.01154236,
      "auxiliary_loss_mlp": 0.01025152,
      "balance_loss_clip": 1.04342616,
      "balance_loss_mlp": 1.01819587,
      "epoch": 0.8122407262670595,
      "flos": 19464589549440.0,
      "grad_norm": 1.765450332735354,
      "language_loss": 0.75360906,
      "learning_rate": 3.584013463499391e-07,
      "loss": 0.7754029,
      "num_input_tokens_seen": 145824180,
      "step": 6755,
      "time_per_iteration": 2.6084117889404297
    },
    {
      "auxiliary_loss_clip": 0.01043866,
      "auxiliary_loss_mlp": 0.01001511,
      "balance_loss_clip": 1.00996268,
      "balance_loss_mlp": 1.00038493,
      "epoch": 0.8123609691576985,
      "flos": 56425325472000.0,
      "grad_norm": 0.7339927557820085,
      "language_loss": 0.64425051,
      "learning_rate": 3.579565092983521e-07,
      "loss": 0.66470432,
      "num_input_tokens_seen": 145885300,
      "step": 6756,
      "time_per_iteration": 3.1868698596954346
    },
    {
      "auxiliary_loss_clip": 0.011654,
      "auxiliary_loss_mlp": 0.01025218,
      "balance_loss_clip": 1.04529715,
      "balance_loss_mlp": 1.01784503,
      "epoch": 0.8124812120483377,
      "flos": 20631973564800.0,
      "grad_norm": 1.9543579265210251,
      "language_loss": 0.83786309,
      "learning_rate": 3.575119213434565e-07,
      "loss": 0.85976928,
      "num_input_tokens_seen": 145903815,
      "step": 6757,
      "time_per_iteration": 2.7193238735198975
    },
    {
      "auxiliary_loss_clip": 0.01150321,
      "auxiliary_loss_mlp": 0.0102234,
      "balance_loss_clip": 1.04619575,
      "balance_loss_mlp": 1.01547945,
      "epoch": 0.8126014549389767,
      "flos": 22492397566080.0,
      "grad_norm": 1.8004129726041063,
      "language_loss": 0.81609917,
      "learning_rate": 3.5706758255269765e-07,
      "loss": 0.83782566,
      "num_input_tokens_seen": 145922270,
      "step": 6758,
      "time_per_iteration": 2.916170358657837
    },
    {
      "auxiliary_loss_clip": 0.01142772,
      "auxiliary_loss_mlp": 0.0103049,
      "balance_loss_clip": 1.04452956,
      "balance_loss_mlp": 1.02332234,
      "epoch": 0.8127216978296158,
      "flos": 23287961946240.0,
      "grad_norm": 1.578065908580091,
      "language_loss": 0.69679046,
      "learning_rate": 3.566234929934795e-07,
      "loss": 0.71852303,
      "num_input_tokens_seen": 145941470,
      "step": 6759,
      "time_per_iteration": 2.6941497325897217
    },
    {
      "auxiliary_loss_clip": 0.0115065,
      "auxiliary_loss_mlp": 0.01023341,
      "balance_loss_clip": 1.04719126,
      "balance_loss_mlp": 1.01639974,
      "epoch": 0.812841940720255,
      "flos": 25154994049920.0,
      "grad_norm": 1.3984229417732994,
      "language_loss": 0.71819484,
      "learning_rate": 3.561796527331706e-07,
      "loss": 0.73993474,
      "num_input_tokens_seen": 145963145,
      "step": 6760,
      "time_per_iteration": 2.6374270915985107
    },
    {
      "auxiliary_loss_clip": 0.01127171,
      "auxiliary_loss_mlp": 0.01024087,
      "balance_loss_clip": 1.04285955,
      "balance_loss_mlp": 1.01705647,
      "epoch": 0.812962183610894,
      "flos": 26648446752000.0,
      "grad_norm": 1.8456232683533451,
      "language_loss": 0.77666843,
      "learning_rate": 3.5573606183910163e-07,
      "loss": 0.798181,
      "num_input_tokens_seen": 145983150,
      "step": 6761,
      "time_per_iteration": 2.765927314758301
    },
    {
      "auxiliary_loss_clip": 0.01158696,
      "auxiliary_loss_mlp": 0.01025123,
      "balance_loss_clip": 1.04464626,
      "balance_loss_mlp": 1.01762176,
      "epoch": 0.8130824265015331,
      "flos": 24966965329920.0,
      "grad_norm": 1.663396216963244,
      "language_loss": 0.78314865,
      "learning_rate": 3.5529272037856493e-07,
      "loss": 0.80498683,
      "num_input_tokens_seen": 146001365,
      "step": 6762,
      "time_per_iteration": 2.6404948234558105
    },
    {
      "auxiliary_loss_clip": 0.01019782,
      "auxiliary_loss_mlp": 0.01001993,
      "balance_loss_clip": 1.01004291,
      "balance_loss_mlp": 1.0009383,
      "epoch": 0.8132026693921722,
      "flos": 67622918175360.0,
      "grad_norm": 0.7267919999992576,
      "language_loss": 0.53819984,
      "learning_rate": 3.548496284188149e-07,
      "loss": 0.55841756,
      "num_input_tokens_seen": 146061570,
      "step": 6763,
      "time_per_iteration": 3.449946880340576
    },
    {
      "auxiliary_loss_clip": 0.01105027,
      "auxiliary_loss_mlp": 0.01023465,
      "balance_loss_clip": 1.04289222,
      "balance_loss_mlp": 1.01691461,
      "epoch": 0.8133229122828113,
      "flos": 19495149045120.0,
      "grad_norm": 1.7052395409267305,
      "language_loss": 0.7922774,
      "learning_rate": 3.544067860270681e-07,
      "loss": 0.81356233,
      "num_input_tokens_seen": 146079145,
      "step": 6764,
      "time_per_iteration": 4.241077899932861
    },
    {
      "auxiliary_loss_clip": 0.01128175,
      "auxiliary_loss_mlp": 0.01029039,
      "balance_loss_clip": 1.04152966,
      "balance_loss_mlp": 1.02127242,
      "epoch": 0.8134431551734503,
      "flos": 20668135582080.0,
      "grad_norm": 1.605901410645251,
      "language_loss": 0.70808089,
      "learning_rate": 3.539641932705029e-07,
      "loss": 0.72965306,
      "num_input_tokens_seen": 146097625,
      "step": 6765,
      "time_per_iteration": 2.6738879680633545
    },
    {
      "auxiliary_loss_clip": 0.011721,
      "auxiliary_loss_mlp": 0.01029411,
      "balance_loss_clip": 1.04835677,
      "balance_loss_mlp": 1.02160525,
      "epoch": 0.8135633980640895,
      "flos": 21507332008320.0,
      "grad_norm": 2.1668270650832313,
      "language_loss": 0.77378178,
      "learning_rate": 3.53521850216262e-07,
      "loss": 0.79579687,
      "num_input_tokens_seen": 146117195,
      "step": 6766,
      "time_per_iteration": 3.5958991050720215
    },
    {
      "auxiliary_loss_clip": 0.01168414,
      "auxiliary_loss_mlp": 0.01028152,
      "balance_loss_clip": 1.04705501,
      "balance_loss_mlp": 1.02080894,
      "epoch": 0.8136836409547286,
      "flos": 20554442058240.0,
      "grad_norm": 1.7462105110917778,
      "language_loss": 0.76822436,
      "learning_rate": 3.530797569314461e-07,
      "loss": 0.79018998,
      "num_input_tokens_seen": 146136220,
      "step": 6767,
      "time_per_iteration": 2.5776124000549316
    },
    {
      "auxiliary_loss_clip": 0.01168279,
      "auxiliary_loss_mlp": 0.01024433,
      "balance_loss_clip": 1.04812407,
      "balance_loss_mlp": 1.0173161,
      "epoch": 0.8138038838453676,
      "flos": 20299045380480.0,
      "grad_norm": 5.3353156622999105,
      "language_loss": 0.77876586,
      "learning_rate": 3.5263791348312235e-07,
      "loss": 0.80069292,
      "num_input_tokens_seen": 146155415,
      "step": 6768,
      "time_per_iteration": 2.6032814979553223
    },
    {
      "auxiliary_loss_clip": 0.0113586,
      "auxiliary_loss_mlp": 0.0102281,
      "balance_loss_clip": 1.04133892,
      "balance_loss_mlp": 1.01562428,
      "epoch": 0.8139241267360068,
      "flos": 29789840551680.0,
      "grad_norm": 1.8885355330848477,
      "language_loss": 0.70610976,
      "learning_rate": 3.521963199383171e-07,
      "loss": 0.72769654,
      "num_input_tokens_seen": 146178370,
      "step": 6769,
      "time_per_iteration": 3.6701419353485107
    },
    {
      "auxiliary_loss_clip": 0.01111533,
      "auxiliary_loss_mlp": 0.01025869,
      "balance_loss_clip": 1.04016232,
      "balance_loss_mlp": 1.01782537,
      "epoch": 0.8140443696266458,
      "flos": 19713270384000.0,
      "grad_norm": 1.9558474050020058,
      "language_loss": 0.76693845,
      "learning_rate": 3.517549763640197e-07,
      "loss": 0.78831244,
      "num_input_tokens_seen": 146196010,
      "step": 6770,
      "time_per_iteration": 2.755021810531616
    },
    {
      "auxiliary_loss_clip": 0.01151632,
      "auxiliary_loss_mlp": 0.00762366,
      "balance_loss_clip": 1.04766679,
      "balance_loss_mlp": 1.00030243,
      "epoch": 0.8141646125172849,
      "flos": 27160568910720.0,
      "grad_norm": 1.9021443885281621,
      "language_loss": 0.71119058,
      "learning_rate": 3.513138828271829e-07,
      "loss": 0.73033059,
      "num_input_tokens_seen": 146215880,
      "step": 6771,
      "time_per_iteration": 2.7321043014526367
    },
    {
      "auxiliary_loss_clip": 0.0111842,
      "auxiliary_loss_mlp": 0.01023915,
      "balance_loss_clip": 1.04097891,
      "balance_loss_mlp": 1.01740909,
      "epoch": 0.8142848554079241,
      "flos": 39673102700160.0,
      "grad_norm": 1.9270048577847785,
      "language_loss": 0.69917703,
      "learning_rate": 3.508730393947179e-07,
      "loss": 0.72060037,
      "num_input_tokens_seen": 146239135,
      "step": 6772,
      "time_per_iteration": 2.809476852416992
    },
    {
      "auxiliary_loss_clip": 0.01123513,
      "auxiliary_loss_mlp": 0.01021841,
      "balance_loss_clip": 1.04175425,
      "balance_loss_mlp": 1.01410413,
      "epoch": 0.8144050982985631,
      "flos": 22237288197120.0,
      "grad_norm": 1.5659830090760427,
      "language_loss": 0.72389346,
      "learning_rate": 3.504324461335024e-07,
      "loss": 0.74534702,
      "num_input_tokens_seen": 146259245,
      "step": 6773,
      "time_per_iteration": 2.7746129035949707
    },
    {
      "auxiliary_loss_clip": 0.0110208,
      "auxiliary_loss_mlp": 0.01026691,
      "balance_loss_clip": 1.03941131,
      "balance_loss_mlp": 1.01924086,
      "epoch": 0.8145253411892022,
      "flos": 23038239617280.0,
      "grad_norm": 1.6839746075585653,
      "language_loss": 0.88235205,
      "learning_rate": 3.499921031103732e-07,
      "loss": 0.90363979,
      "num_input_tokens_seen": 146280015,
      "step": 6774,
      "time_per_iteration": 2.800969123840332
    },
    {
      "auxiliary_loss_clip": 0.01133613,
      "auxiliary_loss_mlp": 0.01024395,
      "balance_loss_clip": 1.04057539,
      "balance_loss_mlp": 1.01735556,
      "epoch": 0.8146455840798413,
      "flos": 24827668387200.0,
      "grad_norm": 1.6582059915551792,
      "language_loss": 0.785864,
      "learning_rate": 3.4955201039212987e-07,
      "loss": 0.80744404,
      "num_input_tokens_seen": 146300935,
      "step": 6775,
      "time_per_iteration": 2.672597646713257
    },
    {
      "auxiliary_loss_clip": 0.01159293,
      "auxiliary_loss_mlp": 0.01032063,
      "balance_loss_clip": 1.04643619,
      "balance_loss_mlp": 1.02517867,
      "epoch": 0.8147658269704804,
      "flos": 19974520978560.0,
      "grad_norm": 2.095969376062296,
      "language_loss": 0.65290308,
      "learning_rate": 3.4911216804553465e-07,
      "loss": 0.67481661,
      "num_input_tokens_seen": 146319835,
      "step": 6776,
      "time_per_iteration": 2.650496244430542
    },
    {
      "auxiliary_loss_clip": 0.01137642,
      "auxiliary_loss_mlp": 0.01026416,
      "balance_loss_clip": 1.0433073,
      "balance_loss_mlp": 1.01853967,
      "epoch": 0.8148860698611194,
      "flos": 21178031097600.0,
      "grad_norm": 2.0547707339855794,
      "language_loss": 0.70269966,
      "learning_rate": 3.4867257613731017e-07,
      "loss": 0.72434026,
      "num_input_tokens_seen": 146339030,
      "step": 6777,
      "time_per_iteration": 2.6557295322418213
    },
    {
      "auxiliary_loss_clip": 0.01140502,
      "auxiliary_loss_mlp": 0.01026518,
      "balance_loss_clip": 1.04281688,
      "balance_loss_mlp": 1.01922822,
      "epoch": 0.8150063127517585,
      "flos": 19606903234560.0,
      "grad_norm": 1.7835040834888813,
      "language_loss": 0.855739,
      "learning_rate": 3.4823323473414343e-07,
      "loss": 0.87740922,
      "num_input_tokens_seen": 146358550,
      "step": 6778,
      "time_per_iteration": 2.688664197921753
    },
    {
      "auxiliary_loss_clip": 0.01134424,
      "auxiliary_loss_mlp": 0.01029502,
      "balance_loss_clip": 1.04386652,
      "balance_loss_mlp": 1.02121675,
      "epoch": 0.8151265556423977,
      "flos": 22638374438400.0,
      "grad_norm": 3.772169122838476,
      "language_loss": 0.75568068,
      "learning_rate": 3.477941439026812e-07,
      "loss": 0.77731991,
      "num_input_tokens_seen": 146376770,
      "step": 6779,
      "time_per_iteration": 2.7082858085632324
    },
    {
      "auxiliary_loss_clip": 0.01142026,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.04575777,
      "balance_loss_mlp": 1.0185802,
      "epoch": 0.8152467985330367,
      "flos": 17968048277760.0,
      "grad_norm": 1.677069985734151,
      "language_loss": 0.73124552,
      "learning_rate": 3.473553037095349e-07,
      "loss": 0.75292432,
      "num_input_tokens_seen": 146395795,
      "step": 6780,
      "time_per_iteration": 2.6599388122558594
    },
    {
      "auxiliary_loss_clip": 0.01131485,
      "auxiliary_loss_mlp": 0.01022629,
      "balance_loss_clip": 1.04079604,
      "balance_loss_mlp": 1.01486492,
      "epoch": 0.8153670414236758,
      "flos": 24969012405120.0,
      "grad_norm": 1.995856671081086,
      "language_loss": 0.83050299,
      "learning_rate": 3.469167142212743e-07,
      "loss": 0.85204411,
      "num_input_tokens_seen": 146417640,
      "step": 6781,
      "time_per_iteration": 2.7476744651794434
    },
    {
      "auxiliary_loss_clip": 0.01153447,
      "auxiliary_loss_mlp": 0.01031783,
      "balance_loss_clip": 1.04586673,
      "balance_loss_mlp": 1.0245738,
      "epoch": 0.8154872843143149,
      "flos": 31066069754880.0,
      "grad_norm": 2.6625612623950574,
      "language_loss": 0.62642157,
      "learning_rate": 3.4647837550443337e-07,
      "loss": 0.64827389,
      "num_input_tokens_seen": 146436205,
      "step": 6782,
      "time_per_iteration": 2.671539783477783
    },
    {
      "auxiliary_loss_clip": 0.01127045,
      "auxiliary_loss_mlp": 0.01023214,
      "balance_loss_clip": 1.0422107,
      "balance_loss_mlp": 1.01571035,
      "epoch": 0.815607527204954,
      "flos": 19391654983680.0,
      "grad_norm": 3.219560972561662,
      "language_loss": 0.74440861,
      "learning_rate": 3.460402876255086e-07,
      "loss": 0.76591122,
      "num_input_tokens_seen": 146453595,
      "step": 6783,
      "time_per_iteration": 2.6240627765655518
    },
    {
      "auxiliary_loss_clip": 0.01156864,
      "auxiliary_loss_mlp": 0.01026325,
      "balance_loss_clip": 1.04634035,
      "balance_loss_mlp": 1.01908886,
      "epoch": 0.815727770095593,
      "flos": 26140418743680.0,
      "grad_norm": 1.94070167111501,
      "language_loss": 0.71741283,
      "learning_rate": 3.456024506509574e-07,
      "loss": 0.73924476,
      "num_input_tokens_seen": 146474515,
      "step": 6784,
      "time_per_iteration": 2.6897995471954346
    },
    {
      "auxiliary_loss_clip": 0.01156674,
      "auxiliary_loss_mlp": 0.00762445,
      "balance_loss_clip": 1.04856813,
      "balance_loss_mlp": 1.00034916,
      "epoch": 0.8158480129862322,
      "flos": 25337527989120.0,
      "grad_norm": 1.7699741280867545,
      "language_loss": 0.73982215,
      "learning_rate": 3.4516486464719873e-07,
      "loss": 0.7590133,
      "num_input_tokens_seen": 146493905,
      "step": 6785,
      "time_per_iteration": 2.7292234897613525
    },
    {
      "auxiliary_loss_clip": 0.01105085,
      "auxiliary_loss_mlp": 0.01020222,
      "balance_loss_clip": 1.03913569,
      "balance_loss_mlp": 1.01296258,
      "epoch": 0.8159682558768713,
      "flos": 34423645559040.0,
      "grad_norm": 2.3015031322578423,
      "language_loss": 0.62340873,
      "learning_rate": 3.4472752968061445e-07,
      "loss": 0.64466178,
      "num_input_tokens_seen": 146518335,
      "step": 6786,
      "time_per_iteration": 2.831507682800293
    },
    {
      "auxiliary_loss_clip": 0.01151235,
      "auxiliary_loss_mlp": 0.0102215,
      "balance_loss_clip": 1.0431838,
      "balance_loss_mlp": 1.01523638,
      "epoch": 0.8160884987675103,
      "flos": 18653223185280.0,
      "grad_norm": 1.9242106479990093,
      "language_loss": 0.7405054,
      "learning_rate": 3.442904458175475e-07,
      "loss": 0.76223928,
      "num_input_tokens_seen": 146535655,
      "step": 6787,
      "time_per_iteration": 2.6088531017303467
    },
    {
      "auxiliary_loss_clip": 0.01149133,
      "auxiliary_loss_mlp": 0.01021235,
      "balance_loss_clip": 1.04317582,
      "balance_loss_mlp": 1.01396942,
      "epoch": 0.8162087416581495,
      "flos": 31430527102080.0,
      "grad_norm": 1.5957272577528205,
      "language_loss": 0.75919896,
      "learning_rate": 3.438536131243044e-07,
      "loss": 0.78090262,
      "num_input_tokens_seen": 146556815,
      "step": 6788,
      "time_per_iteration": 2.6711971759796143
    },
    {
      "auxiliary_loss_clip": 0.01140871,
      "auxiliary_loss_mlp": 0.01026308,
      "balance_loss_clip": 1.04209173,
      "balance_loss_mlp": 1.01776111,
      "epoch": 0.8163289845487885,
      "flos": 37593910915200.0,
      "grad_norm": 2.20547633117888,
      "language_loss": 0.62033814,
      "learning_rate": 3.434170316671503e-07,
      "loss": 0.64200997,
      "num_input_tokens_seen": 146581845,
      "step": 6789,
      "time_per_iteration": 3.727544069290161
    },
    {
      "auxiliary_loss_clip": 0.01117867,
      "auxiliary_loss_mlp": 0.01020254,
      "balance_loss_clip": 1.04487443,
      "balance_loss_mlp": 1.01314926,
      "epoch": 0.8164492274394276,
      "flos": 13953989554560.0,
      "grad_norm": 2.274752726647801,
      "language_loss": 0.89709854,
      "learning_rate": 3.4298070151231583e-07,
      "loss": 0.91847974,
      "num_input_tokens_seen": 146597245,
      "step": 6790,
      "time_per_iteration": 2.6696135997772217
    },
    {
      "auxiliary_loss_clip": 0.01142962,
      "auxiliary_loss_mlp": 0.01026564,
      "balance_loss_clip": 1.04367554,
      "balance_loss_mlp": 1.01980805,
      "epoch": 0.8165694703300668,
      "flos": 28986554747520.0,
      "grad_norm": 1.7878639961843614,
      "language_loss": 0.60046828,
      "learning_rate": 3.425446227259916e-07,
      "loss": 0.62216359,
      "num_input_tokens_seen": 146618210,
      "step": 6791,
      "time_per_iteration": 3.703967809677124
    },
    {
      "auxiliary_loss_clip": 0.0114091,
      "auxiliary_loss_mlp": 0.01024608,
      "balance_loss_clip": 1.04360437,
      "balance_loss_mlp": 1.01765847,
      "epoch": 0.8166897132207058,
      "flos": 25118365155840.0,
      "grad_norm": 2.1697990944121477,
      "language_loss": 0.82503033,
      "learning_rate": 3.421087953743296e-07,
      "loss": 0.84668547,
      "num_input_tokens_seen": 146637975,
      "step": 6792,
      "time_per_iteration": 2.7091760635375977
    },
    {
      "auxiliary_loss_clip": 0.01152963,
      "auxiliary_loss_mlp": 0.0102502,
      "balance_loss_clip": 1.04258275,
      "balance_loss_mlp": 1.01778126,
      "epoch": 0.8168099561113449,
      "flos": 23148593176320.0,
      "grad_norm": 1.929388794282692,
      "language_loss": 0.80264533,
      "learning_rate": 3.416732195234464e-07,
      "loss": 0.82442522,
      "num_input_tokens_seen": 146658030,
      "step": 6793,
      "time_per_iteration": 2.5943613052368164
    },
    {
      "auxiliary_loss_clip": 0.01153775,
      "auxiliary_loss_mlp": 0.01021909,
      "balance_loss_clip": 1.04283488,
      "balance_loss_mlp": 1.01501,
      "epoch": 0.816930199001984,
      "flos": 18407666833920.0,
      "grad_norm": 1.6858651072172797,
      "language_loss": 0.79413015,
      "learning_rate": 3.4123789523941613e-07,
      "loss": 0.81588697,
      "num_input_tokens_seen": 146677855,
      "step": 6794,
      "time_per_iteration": 3.5468177795410156
    },
    {
      "auxiliary_loss_clip": 0.01143687,
      "auxiliary_loss_mlp": 0.01027542,
      "balance_loss_clip": 1.04106283,
      "balance_loss_mlp": 1.02021086,
      "epoch": 0.8170504418926231,
      "flos": 21251324799360.0,
      "grad_norm": 1.5397155287058346,
      "language_loss": 0.63563758,
      "learning_rate": 3.4080282258827884e-07,
      "loss": 0.65734982,
      "num_input_tokens_seen": 146696230,
      "step": 6795,
      "time_per_iteration": 3.622075319290161
    },
    {
      "auxiliary_loss_clip": 0.01155362,
      "auxiliary_loss_mlp": 0.01024098,
      "balance_loss_clip": 1.046386,
      "balance_loss_mlp": 1.01682341,
      "epoch": 0.8171706847832622,
      "flos": 19099234362240.0,
      "grad_norm": 2.0425453612048052,
      "language_loss": 0.72574759,
      "learning_rate": 3.403680016360342e-07,
      "loss": 0.7475422,
      "num_input_tokens_seen": 146714835,
      "step": 6796,
      "time_per_iteration": 2.6547253131866455
    },
    {
      "auxiliary_loss_clip": 0.01148249,
      "auxiliary_loss_mlp": 0.01034553,
      "balance_loss_clip": 1.04631543,
      "balance_loss_mlp": 1.02678323,
      "epoch": 0.8172909276739013,
      "flos": 21470128496640.0,
      "grad_norm": 1.4700048684002351,
      "language_loss": 0.67752874,
      "learning_rate": 3.3993343244864403e-07,
      "loss": 0.69935679,
      "num_input_tokens_seen": 146734425,
      "step": 6797,
      "time_per_iteration": 2.813734292984009
    },
    {
      "auxiliary_loss_clip": 0.01151411,
      "auxiliary_loss_mlp": 0.01023312,
      "balance_loss_clip": 1.04508317,
      "balance_loss_mlp": 1.01665378,
      "epoch": 0.8174111705645404,
      "flos": 27599792417280.0,
      "grad_norm": 1.947826010283228,
      "language_loss": 0.72654766,
      "learning_rate": 3.394991150920323e-07,
      "loss": 0.74829489,
      "num_input_tokens_seen": 146757545,
      "step": 6798,
      "time_per_iteration": 2.6812334060668945
    },
    {
      "auxiliary_loss_clip": 0.01113109,
      "auxiliary_loss_mlp": 0.00762909,
      "balance_loss_clip": 1.04083431,
      "balance_loss_mlp": 1.00032032,
      "epoch": 0.8175314134551794,
      "flos": 14064594508800.0,
      "grad_norm": 2.037867654210426,
      "language_loss": 0.74162453,
      "learning_rate": 3.3906504963208396e-07,
      "loss": 0.76038468,
      "num_input_tokens_seen": 146774240,
      "step": 6799,
      "time_per_iteration": 2.7910609245300293
    },
    {
      "auxiliary_loss_clip": 0.01105151,
      "auxiliary_loss_mlp": 0.0103017,
      "balance_loss_clip": 1.04198909,
      "balance_loss_mlp": 1.02259409,
      "epoch": 0.8176516563458186,
      "flos": 22708076780160.0,
      "grad_norm": 1.7761928562735736,
      "language_loss": 0.66340107,
      "learning_rate": 3.3863123613464774e-07,
      "loss": 0.68475425,
      "num_input_tokens_seen": 146793140,
      "step": 6800,
      "time_per_iteration": 2.7528247833251953
    },
    {
      "auxiliary_loss_clip": 0.01139653,
      "auxiliary_loss_mlp": 0.01022469,
      "balance_loss_clip": 1.03985929,
      "balance_loss_mlp": 1.01553738,
      "epoch": 0.8177718992364577,
      "flos": 21945406279680.0,
      "grad_norm": 1.7618752587457034,
      "language_loss": 0.75051808,
      "learning_rate": 3.381976746655317e-07,
      "loss": 0.77213931,
      "num_input_tokens_seen": 146812895,
      "step": 6801,
      "time_per_iteration": 2.7613768577575684
    },
    {
      "auxiliary_loss_clip": 0.0110321,
      "auxiliary_loss_mlp": 0.01021971,
      "balance_loss_clip": 1.04286706,
      "balance_loss_mlp": 1.01461291,
      "epoch": 0.8178921421270967,
      "flos": 22017443005440.0,
      "grad_norm": 2.3711849956579765,
      "language_loss": 0.66908723,
      "learning_rate": 3.3776436529050756e-07,
      "loss": 0.69033897,
      "num_input_tokens_seen": 146832445,
      "step": 6802,
      "time_per_iteration": 2.7386040687561035
    },
    {
      "auxiliary_loss_clip": 0.01162426,
      "auxiliary_loss_mlp": 0.01021226,
      "balance_loss_clip": 1.04387558,
      "balance_loss_mlp": 1.01431751,
      "epoch": 0.8180123850177359,
      "flos": 33183111496320.0,
      "grad_norm": 1.7884706085642583,
      "language_loss": 0.72550929,
      "learning_rate": 3.373313080753073e-07,
      "loss": 0.74734586,
      "num_input_tokens_seen": 146856505,
      "step": 6803,
      "time_per_iteration": 2.73659610748291
    },
    {
      "auxiliary_loss_clip": 0.01146235,
      "auxiliary_loss_mlp": 0.01024023,
      "balance_loss_clip": 1.04173827,
      "balance_loss_mlp": 1.01662064,
      "epoch": 0.8181326279083749,
      "flos": 22091167670400.0,
      "grad_norm": 1.5053419981682647,
      "language_loss": 0.77437043,
      "learning_rate": 3.3689850308562527e-07,
      "loss": 0.79607302,
      "num_input_tokens_seen": 146876950,
      "step": 6804,
      "time_per_iteration": 2.6836154460906982
    },
    {
      "auxiliary_loss_clip": 0.01100999,
      "auxiliary_loss_mlp": 0.01027418,
      "balance_loss_clip": 1.04209661,
      "balance_loss_mlp": 1.02075982,
      "epoch": 0.818252870799014,
      "flos": 15705747936000.0,
      "grad_norm": 2.183016564911383,
      "language_loss": 0.77830648,
      "learning_rate": 3.364659503871183e-07,
      "loss": 0.79959059,
      "num_input_tokens_seen": 146894885,
      "step": 6805,
      "time_per_iteration": 2.680767774581909
    },
    {
      "auxiliary_loss_clip": 0.0111702,
      "auxiliary_loss_mlp": 0.01021652,
      "balance_loss_clip": 1.03714645,
      "balance_loss_mlp": 1.01509249,
      "epoch": 0.8183731136896532,
      "flos": 18770687637120.0,
      "grad_norm": 1.7656548376046912,
      "language_loss": 0.8377685,
      "learning_rate": 3.3603365004540417e-07,
      "loss": 0.85915518,
      "num_input_tokens_seen": 146913180,
      "step": 6806,
      "time_per_iteration": 2.7702226638793945
    },
    {
      "auxiliary_loss_clip": 0.01167225,
      "auxiliary_loss_mlp": 0.01028531,
      "balance_loss_clip": 1.04855084,
      "balance_loss_mlp": 1.02108908,
      "epoch": 0.8184933565802922,
      "flos": 26541792293760.0,
      "grad_norm": 2.137188309210202,
      "language_loss": 0.77125549,
      "learning_rate": 3.356016021260624e-07,
      "loss": 0.79321301,
      "num_input_tokens_seen": 146933510,
      "step": 6807,
      "time_per_iteration": 2.7126476764678955
    },
    {
      "auxiliary_loss_clip": 0.01153829,
      "auxiliary_loss_mlp": 0.01027971,
      "balance_loss_clip": 1.04523563,
      "balance_loss_mlp": 1.02080357,
      "epoch": 0.8186135994709313,
      "flos": 17530117660800.0,
      "grad_norm": 3.5725032667554237,
      "language_loss": 0.6569097,
      "learning_rate": 3.35169806694634e-07,
      "loss": 0.67872763,
      "num_input_tokens_seen": 146951760,
      "step": 6808,
      "time_per_iteration": 2.5987396240234375
    },
    {
      "auxiliary_loss_clip": 0.01036393,
      "auxiliary_loss_mlp": 0.01000374,
      "balance_loss_clip": 1.01467621,
      "balance_loss_mlp": 0.99934274,
      "epoch": 0.8187338423615703,
      "flos": 63480300675840.0,
      "grad_norm": 0.7158209206496644,
      "language_loss": 0.60612559,
      "learning_rate": 3.3473826381662186e-07,
      "loss": 0.62649333,
      "num_input_tokens_seen": 147022900,
      "step": 6809,
      "time_per_iteration": 3.3959803581237793
    },
    {
      "auxiliary_loss_clip": 0.01148081,
      "auxiliary_loss_mlp": 0.01023279,
      "balance_loss_clip": 1.04593217,
      "balance_loss_mlp": 1.0165168,
      "epoch": 0.8188540852522095,
      "flos": 17529974006400.0,
      "grad_norm": 1.8507835486934523,
      "language_loss": 0.81545275,
      "learning_rate": 3.3430697355749216e-07,
      "loss": 0.83716637,
      "num_input_tokens_seen": 147040590,
      "step": 6810,
      "time_per_iteration": 2.628107786178589
    },
    {
      "auxiliary_loss_clip": 0.01105418,
      "auxiliary_loss_mlp": 0.01023156,
      "balance_loss_clip": 1.03855443,
      "balance_loss_mlp": 1.01596451,
      "epoch": 0.8189743281428485,
      "flos": 14392530702720.0,
      "grad_norm": 2.129243998376342,
      "language_loss": 0.75437975,
      "learning_rate": 3.3387593598266907e-07,
      "loss": 0.7756654,
      "num_input_tokens_seen": 147057200,
      "step": 6811,
      "time_per_iteration": 2.6898815631866455
    },
    {
      "auxiliary_loss_clip": 0.01117077,
      "auxiliary_loss_mlp": 0.01025909,
      "balance_loss_clip": 1.03909302,
      "balance_loss_mlp": 1.0189023,
      "epoch": 0.8190945710334876,
      "flos": 25080479285760.0,
      "grad_norm": 1.8140498751227367,
      "language_loss": 0.78205192,
      "learning_rate": 3.3344515115754225e-07,
      "loss": 0.80348188,
      "num_input_tokens_seen": 147076180,
      "step": 6812,
      "time_per_iteration": 2.7593486309051514
    },
    {
      "auxiliary_loss_clip": 0.01127575,
      "auxiliary_loss_mlp": 0.0102507,
      "balance_loss_clip": 1.04146886,
      "balance_loss_mlp": 1.01745248,
      "epoch": 0.8192148139241268,
      "flos": 21507152440320.0,
      "grad_norm": 2.6296371305132604,
      "language_loss": 0.79686183,
      "learning_rate": 3.33014619147461e-07,
      "loss": 0.81838834,
      "num_input_tokens_seen": 147094205,
      "step": 6813,
      "time_per_iteration": 2.728896379470825
    },
    {
      "auxiliary_loss_clip": 0.01138478,
      "auxiliary_loss_mlp": 0.01027684,
      "balance_loss_clip": 1.04560375,
      "balance_loss_mlp": 1.02068019,
      "epoch": 0.8193350568147658,
      "flos": 23952166289280.0,
      "grad_norm": 2.333424950073784,
      "language_loss": 0.7172007,
      "learning_rate": 3.325843400177362e-07,
      "loss": 0.73886228,
      "num_input_tokens_seen": 147115545,
      "step": 6814,
      "time_per_iteration": 2.675199270248413
    },
    {
      "auxiliary_loss_clip": 0.01156394,
      "auxiliary_loss_mlp": 0.00762201,
      "balance_loss_clip": 1.04531801,
      "balance_loss_mlp": 1.00033975,
      "epoch": 0.8194552997054049,
      "flos": 20559469962240.0,
      "grad_norm": 1.891592898025834,
      "language_loss": 0.73589027,
      "learning_rate": 3.32154313833642e-07,
      "loss": 0.75507617,
      "num_input_tokens_seen": 147135700,
      "step": 6815,
      "time_per_iteration": 3.564971923828125
    },
    {
      "auxiliary_loss_clip": 0.01167365,
      "auxiliary_loss_mlp": 0.01025997,
      "balance_loss_clip": 1.04571378,
      "balance_loss_mlp": 1.01864219,
      "epoch": 0.819575542596044,
      "flos": 26031753123840.0,
      "grad_norm": 2.0546941626694246,
      "language_loss": 0.59619355,
      "learning_rate": 3.3172454066041164e-07,
      "loss": 0.61812717,
      "num_input_tokens_seen": 147155205,
      "step": 6816,
      "time_per_iteration": 2.576524257659912
    },
    {
      "auxiliary_loss_clip": 0.01093095,
      "auxiliary_loss_mlp": 0.00761777,
      "balance_loss_clip": 1.04022646,
      "balance_loss_mlp": 1.00022829,
      "epoch": 0.8196957854866831,
      "flos": 29096944220160.0,
      "grad_norm": 2.053170800429048,
      "language_loss": 0.76442891,
      "learning_rate": 3.3129502056324234e-07,
      "loss": 0.78297758,
      "num_input_tokens_seen": 147176570,
      "step": 6817,
      "time_per_iteration": 2.888990640640259
    },
    {
      "auxiliary_loss_clip": 0.01002923,
      "auxiliary_loss_mlp": 0.01001357,
      "balance_loss_clip": 1.00972974,
      "balance_loss_mlp": 1.0003016,
      "epoch": 0.8198160283773221,
      "flos": 69033631898880.0,
      "grad_norm": 0.7954721573869636,
      "language_loss": 0.59695387,
      "learning_rate": 3.3086575360729165e-07,
      "loss": 0.61699677,
      "num_input_tokens_seen": 147234105,
      "step": 6818,
      "time_per_iteration": 4.50211238861084
    },
    {
      "auxiliary_loss_clip": 0.01136302,
      "auxiliary_loss_mlp": 0.01025642,
      "balance_loss_clip": 1.04409206,
      "balance_loss_mlp": 1.01839101,
      "epoch": 0.8199362712679613,
      "flos": 16618058496000.0,
      "grad_norm": 1.926953100479827,
      "language_loss": 0.71491039,
      "learning_rate": 3.3043673985767906e-07,
      "loss": 0.73652983,
      "num_input_tokens_seen": 147253170,
      "step": 6819,
      "time_per_iteration": 3.029890298843384
    },
    {
      "auxiliary_loss_clip": 0.0110912,
      "auxiliary_loss_mlp": 0.01028331,
      "balance_loss_clip": 1.03571379,
      "balance_loss_mlp": 1.02119923,
      "epoch": 0.8200565141586004,
      "flos": 21757664868480.0,
      "grad_norm": 1.6440375861230732,
      "language_loss": 0.77597892,
      "learning_rate": 3.3000797937948564e-07,
      "loss": 0.79735345,
      "num_input_tokens_seen": 147271465,
      "step": 6820,
      "time_per_iteration": 3.6129424571990967
    },
    {
      "auxiliary_loss_clip": 0.01034633,
      "auxiliary_loss_mlp": 0.01000839,
      "balance_loss_clip": 1.0099647,
      "balance_loss_mlp": 0.99975377,
      "epoch": 0.8201767570492394,
      "flos": 69807112392960.0,
      "grad_norm": 0.9542628581485805,
      "language_loss": 0.64954996,
      "learning_rate": 3.295794722377534e-07,
      "loss": 0.66990465,
      "num_input_tokens_seen": 147335070,
      "step": 6821,
      "time_per_iteration": 4.214130878448486
    },
    {
      "auxiliary_loss_clip": 0.01162271,
      "auxiliary_loss_mlp": 0.01024254,
      "balance_loss_clip": 1.04453683,
      "balance_loss_mlp": 1.01754534,
      "epoch": 0.8202969999398786,
      "flos": 23111892455040.0,
      "grad_norm": 1.6409227124065595,
      "language_loss": 0.8032335,
      "learning_rate": 3.291512184974876e-07,
      "loss": 0.82509875,
      "num_input_tokens_seen": 147355460,
      "step": 6822,
      "time_per_iteration": 2.63578200340271
    },
    {
      "auxiliary_loss_clip": 0.01135538,
      "auxiliary_loss_mlp": 0.0102677,
      "balance_loss_clip": 1.04030132,
      "balance_loss_mlp": 1.0188899,
      "epoch": 0.8204172428305176,
      "flos": 28220616109440.0,
      "grad_norm": 1.7942194889419898,
      "language_loss": 0.66882432,
      "learning_rate": 3.2872321822365346e-07,
      "loss": 0.69044745,
      "num_input_tokens_seen": 147375675,
      "step": 6823,
      "time_per_iteration": 2.8146891593933105
    },
    {
      "auxiliary_loss_clip": 0.01149681,
      "auxiliary_loss_mlp": 0.01025926,
      "balance_loss_clip": 1.04458487,
      "balance_loss_mlp": 1.01904774,
      "epoch": 0.8205374857211567,
      "flos": 20887011106560.0,
      "grad_norm": 1.966202374375574,
      "language_loss": 0.73363161,
      "learning_rate": 3.282954714811783e-07,
      "loss": 0.75538766,
      "num_input_tokens_seen": 147394580,
      "step": 6824,
      "time_per_iteration": 2.7465476989746094
    },
    {
      "auxiliary_loss_clip": 0.01124685,
      "auxiliary_loss_mlp": 0.01028487,
      "balance_loss_clip": 1.03859639,
      "balance_loss_mlp": 1.02118492,
      "epoch": 0.8206577286117959,
      "flos": 13152140294400.0,
      "grad_norm": 2.3505178210132227,
      "language_loss": 0.7091049,
      "learning_rate": 3.2786797833495093e-07,
      "loss": 0.7306366,
      "num_input_tokens_seen": 147409935,
      "step": 6825,
      "time_per_iteration": 2.6675689220428467
    },
    {
      "auxiliary_loss_clip": 0.01164936,
      "auxiliary_loss_mlp": 0.01026295,
      "balance_loss_clip": 1.04634595,
      "balance_loss_mlp": 1.01978588,
      "epoch": 0.8207779715024349,
      "flos": 25265634917760.0,
      "grad_norm": 1.9069559303895203,
      "language_loss": 0.72501826,
      "learning_rate": 3.274407388498213e-07,
      "loss": 0.7469306,
      "num_input_tokens_seen": 147428065,
      "step": 6826,
      "time_per_iteration": 2.7370588779449463
    },
    {
      "auxiliary_loss_clip": 0.01120405,
      "auxiliary_loss_mlp": 0.01023993,
      "balance_loss_clip": 1.04068327,
      "balance_loss_mlp": 1.01701343,
      "epoch": 0.820898214393074,
      "flos": 19610243199360.0,
      "grad_norm": 5.5213273193389005,
      "language_loss": 0.74619353,
      "learning_rate": 3.270137530906021e-07,
      "loss": 0.76763749,
      "num_input_tokens_seen": 147447300,
      "step": 6827,
      "time_per_iteration": 2.757395029067993
    },
    {
      "auxiliary_loss_clip": 0.01099938,
      "auxiliary_loss_mlp": 0.01024548,
      "balance_loss_clip": 1.04060459,
      "balance_loss_mlp": 1.01785755,
      "epoch": 0.8210184572837131,
      "flos": 15596615439360.0,
      "grad_norm": 14.48063162950305,
      "language_loss": 0.83311737,
      "learning_rate": 3.265870211220665e-07,
      "loss": 0.85436225,
      "num_input_tokens_seen": 147465135,
      "step": 6828,
      "time_per_iteration": 2.7714548110961914
    },
    {
      "auxiliary_loss_clip": 0.0111787,
      "auxiliary_loss_mlp": 0.01031915,
      "balance_loss_clip": 1.0414896,
      "balance_loss_mlp": 1.02448225,
      "epoch": 0.8211387001743522,
      "flos": 20813932886400.0,
      "grad_norm": 1.8250351126514501,
      "language_loss": 0.81677771,
      "learning_rate": 3.2616054300894934e-07,
      "loss": 0.83827555,
      "num_input_tokens_seen": 147484585,
      "step": 6829,
      "time_per_iteration": 2.7387709617614746
    },
    {
      "auxiliary_loss_clip": 0.01125869,
      "auxiliary_loss_mlp": 0.01029995,
      "balance_loss_clip": 1.04103065,
      "balance_loss_mlp": 1.02214491,
      "epoch": 0.8212589430649913,
      "flos": 27704579368320.0,
      "grad_norm": 2.455088860433198,
      "language_loss": 0.84382081,
      "learning_rate": 3.2573431881594693e-07,
      "loss": 0.86537945,
      "num_input_tokens_seen": 147504130,
      "step": 6830,
      "time_per_iteration": 2.8171210289001465
    },
    {
      "auxiliary_loss_clip": 0.01092465,
      "auxiliary_loss_mlp": 0.01025556,
      "balance_loss_clip": 1.03477597,
      "balance_loss_mlp": 1.01816761,
      "epoch": 0.8213791859556304,
      "flos": 22455625017600.0,
      "grad_norm": 2.530409609510806,
      "language_loss": 0.65670943,
      "learning_rate": 3.2530834860771663e-07,
      "loss": 0.6778897,
      "num_input_tokens_seen": 147523510,
      "step": 6831,
      "time_per_iteration": 2.796443462371826
    },
    {
      "auxiliary_loss_clip": 0.01150943,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.04264617,
      "balance_loss_mlp": 1.02017069,
      "epoch": 0.8214994288462695,
      "flos": 16654471908480.0,
      "grad_norm": 2.0680328298978488,
      "language_loss": 0.74380672,
      "learning_rate": 3.248826324488794e-07,
      "loss": 0.76559019,
      "num_input_tokens_seen": 147540805,
      "step": 6832,
      "time_per_iteration": 2.648983955383301
    },
    {
      "auxiliary_loss_clip": 0.01166672,
      "auxiliary_loss_mlp": 0.01025501,
      "balance_loss_clip": 1.04863071,
      "balance_loss_mlp": 1.01893258,
      "epoch": 0.8216196717369085,
      "flos": 25221787390080.0,
      "grad_norm": 1.7599528645319884,
      "language_loss": 0.87514991,
      "learning_rate": 3.244571704040138e-07,
      "loss": 0.89707166,
      "num_input_tokens_seen": 147560965,
      "step": 6833,
      "time_per_iteration": 2.643709182739258
    },
    {
      "auxiliary_loss_clip": 0.01148139,
      "auxiliary_loss_mlp": 0.01024755,
      "balance_loss_clip": 1.04212093,
      "balance_loss_mlp": 1.01735854,
      "epoch": 0.8217399146275477,
      "flos": 25371930240000.0,
      "grad_norm": 1.899377404175653,
      "language_loss": 0.73382634,
      "learning_rate": 3.2403196253766374e-07,
      "loss": 0.75555527,
      "num_input_tokens_seen": 147580045,
      "step": 6834,
      "time_per_iteration": 2.700230836868286
    },
    {
      "auxiliary_loss_clip": 0.01148298,
      "auxiliary_loss_mlp": 0.01029285,
      "balance_loss_clip": 1.04324365,
      "balance_loss_mlp": 1.02066612,
      "epoch": 0.8218601575181868,
      "flos": 25629625388160.0,
      "grad_norm": 2.830981040625386,
      "language_loss": 0.79149902,
      "learning_rate": 3.2360700891433254e-07,
      "loss": 0.8132748,
      "num_input_tokens_seen": 147599070,
      "step": 6835,
      "time_per_iteration": 2.6842093467712402
    },
    {
      "auxiliary_loss_clip": 0.01021158,
      "auxiliary_loss_mlp": 0.01000819,
      "balance_loss_clip": 1.01012921,
      "balance_loss_mlp": 0.99976963,
      "epoch": 0.8219804004088258,
      "flos": 67660229427840.0,
      "grad_norm": 0.7930575116686561,
      "language_loss": 0.57276225,
      "learning_rate": 3.231823095984847e-07,
      "loss": 0.59298205,
      "num_input_tokens_seen": 147653710,
      "step": 6836,
      "time_per_iteration": 3.2144362926483154
    },
    {
      "auxiliary_loss_clip": 0.01136561,
      "auxiliary_loss_mlp": 0.01023428,
      "balance_loss_clip": 1.04313612,
      "balance_loss_mlp": 1.01697588,
      "epoch": 0.822100643299465,
      "flos": 19464266327040.0,
      "grad_norm": 3.933087390946373,
      "language_loss": 0.76283741,
      "learning_rate": 3.2275786465454814e-07,
      "loss": 0.7844373,
      "num_input_tokens_seen": 147670360,
      "step": 6837,
      "time_per_iteration": 2.6482715606689453
    },
    {
      "auxiliary_loss_clip": 0.01122733,
      "auxiliary_loss_mlp": 0.01022469,
      "balance_loss_clip": 1.04110289,
      "balance_loss_mlp": 1.01536751,
      "epoch": 0.822220886190104,
      "flos": 24681368292480.0,
      "grad_norm": 1.8144310590163144,
      "language_loss": 0.7562269,
      "learning_rate": 3.2233367414690917e-07,
      "loss": 0.77767891,
      "num_input_tokens_seen": 147692550,
      "step": 6838,
      "time_per_iteration": 2.768923282623291
    },
    {
      "auxiliary_loss_clip": 0.01120903,
      "auxiliary_loss_mlp": 0.01019917,
      "balance_loss_clip": 1.04031301,
      "balance_loss_mlp": 1.01327145,
      "epoch": 0.8223411290807431,
      "flos": 27819062991360.0,
      "grad_norm": 2.192067446991063,
      "language_loss": 0.84866214,
      "learning_rate": 3.219097381399183e-07,
      "loss": 0.87007034,
      "num_input_tokens_seen": 147709725,
      "step": 6839,
      "time_per_iteration": 2.909639358520508
    },
    {
      "auxiliary_loss_clip": 0.01144701,
      "auxiliary_loss_mlp": 0.01026866,
      "balance_loss_clip": 1.04399872,
      "balance_loss_mlp": 1.01936769,
      "epoch": 0.8224613719713821,
      "flos": 23218546913280.0,
      "grad_norm": 2.352031751023756,
      "language_loss": 0.81310326,
      "learning_rate": 3.2148605669788584e-07,
      "loss": 0.83481896,
      "num_input_tokens_seen": 147729615,
      "step": 6840,
      "time_per_iteration": 2.7260680198669434
    },
    {
      "auxiliary_loss_clip": 0.01140528,
      "auxiliary_loss_mlp": 0.01024479,
      "balance_loss_clip": 1.04476714,
      "balance_loss_mlp": 1.01808071,
      "epoch": 0.8225816148620213,
      "flos": 15706250726400.0,
      "grad_norm": 2.696695956319773,
      "language_loss": 0.77717996,
      "learning_rate": 3.2106262988508405e-07,
      "loss": 0.79883003,
      "num_input_tokens_seen": 147747665,
      "step": 6841,
      "time_per_iteration": 3.554579734802246
    },
    {
      "auxiliary_loss_clip": 0.01136592,
      "auxiliary_loss_mlp": 0.01026157,
      "balance_loss_clip": 1.04272604,
      "balance_loss_mlp": 1.01849782,
      "epoch": 0.8227018577526604,
      "flos": 18515111391360.0,
      "grad_norm": 1.824029120400195,
      "language_loss": 0.74170238,
      "learning_rate": 3.206394577657465e-07,
      "loss": 0.76332986,
      "num_input_tokens_seen": 147765445,
      "step": 6842,
      "time_per_iteration": 2.6430625915527344
    },
    {
      "auxiliary_loss_clip": 0.01156388,
      "auxiliary_loss_mlp": 0.01029685,
      "balance_loss_clip": 1.04479218,
      "balance_loss_mlp": 1.02215743,
      "epoch": 0.8228221006432994,
      "flos": 22236785406720.0,
      "grad_norm": 2.6196643057513516,
      "language_loss": 0.72625279,
      "learning_rate": 3.202165404040675e-07,
      "loss": 0.74811351,
      "num_input_tokens_seen": 147783365,
      "step": 6843,
      "time_per_iteration": 2.6649913787841797
    },
    {
      "auxiliary_loss_clip": 0.01092603,
      "auxiliary_loss_mlp": 0.01029962,
      "balance_loss_clip": 1.03896499,
      "balance_loss_mlp": 1.02204084,
      "epoch": 0.8229423435339386,
      "flos": 24097532630400.0,
      "grad_norm": 1.938151880741357,
      "language_loss": 0.74711716,
      "learning_rate": 3.1979387786420396e-07,
      "loss": 0.76834279,
      "num_input_tokens_seen": 147803605,
      "step": 6844,
      "time_per_iteration": 3.8545730113983154
    },
    {
      "auxiliary_loss_clip": 0.01138117,
      "auxiliary_loss_mlp": 0.01023619,
      "balance_loss_clip": 1.04012895,
      "balance_loss_mlp": 1.01667523,
      "epoch": 0.8230625864245776,
      "flos": 23878549365120.0,
      "grad_norm": 2.563518927888582,
      "language_loss": 0.81982243,
      "learning_rate": 3.1937147021027346e-07,
      "loss": 0.84143984,
      "num_input_tokens_seen": 147822060,
      "step": 6845,
      "time_per_iteration": 2.6494531631469727
    },
    {
      "auxiliary_loss_clip": 0.01150125,
      "auxiliary_loss_mlp": 0.01021499,
      "balance_loss_clip": 1.0456742,
      "balance_loss_mlp": 1.01466537,
      "epoch": 0.8231828293152167,
      "flos": 16581106379520.0,
      "grad_norm": 2.2357334054310853,
      "language_loss": 0.7664417,
      "learning_rate": 3.189493175063547e-07,
      "loss": 0.78815794,
      "num_input_tokens_seen": 147839295,
      "step": 6846,
      "time_per_iteration": 3.480252265930176
    },
    {
      "auxiliary_loss_clip": 0.01140905,
      "auxiliary_loss_mlp": 0.01026308,
      "balance_loss_clip": 1.0456351,
      "balance_loss_mlp": 1.01920295,
      "epoch": 0.8233030722058559,
      "flos": 18880071528960.0,
      "grad_norm": 1.985214438547369,
      "language_loss": 0.67680234,
      "learning_rate": 3.1852741981648776e-07,
      "loss": 0.69847453,
      "num_input_tokens_seen": 147857945,
      "step": 6847,
      "time_per_iteration": 3.59553599357605
    },
    {
      "auxiliary_loss_clip": 0.01112154,
      "auxiliary_loss_mlp": 0.01027301,
      "balance_loss_clip": 1.04046929,
      "balance_loss_mlp": 1.01948357,
      "epoch": 0.8234233150964949,
      "flos": 28439024757120.0,
      "grad_norm": 2.354619975278059,
      "language_loss": 0.69872355,
      "learning_rate": 3.1810577720467404e-07,
      "loss": 0.72011817,
      "num_input_tokens_seen": 147879675,
      "step": 6848,
      "time_per_iteration": 2.7346746921539307
    },
    {
      "auxiliary_loss_clip": 0.01142453,
      "auxiliary_loss_mlp": 0.01023929,
      "balance_loss_clip": 1.04334617,
      "balance_loss_mlp": 1.01637459,
      "epoch": 0.823543557987134,
      "flos": 33765941577600.0,
      "grad_norm": 1.5685924918047176,
      "language_loss": 0.56674284,
      "learning_rate": 3.176843897348769e-07,
      "loss": 0.58840668,
      "num_input_tokens_seen": 147902870,
      "step": 6849,
      "time_per_iteration": 2.810746431350708
    },
    {
      "auxiliary_loss_clip": 0.01130911,
      "auxiliary_loss_mlp": 0.01025135,
      "balance_loss_clip": 1.04127824,
      "balance_loss_mlp": 1.01794112,
      "epoch": 0.8236638008777731,
      "flos": 17092366611840.0,
      "grad_norm": 3.1687785230804284,
      "language_loss": 0.75636572,
      "learning_rate": 3.1726325747102034e-07,
      "loss": 0.77792615,
      "num_input_tokens_seen": 147921245,
      "step": 6850,
      "time_per_iteration": 2.749417543411255
    },
    {
      "auxiliary_loss_clip": 0.01102765,
      "auxiliary_loss_mlp": 0.01025391,
      "balance_loss_clip": 1.03689504,
      "balance_loss_mlp": 1.01829219,
      "epoch": 0.8237840437684122,
      "flos": 61639982334720.0,
      "grad_norm": 1.5031607549176016,
      "language_loss": 0.64180243,
      "learning_rate": 3.1684238047698974e-07,
      "loss": 0.66308403,
      "num_input_tokens_seen": 147949515,
      "step": 6851,
      "time_per_iteration": 3.1159825325012207
    },
    {
      "auxiliary_loss_clip": 0.01140625,
      "auxiliary_loss_mlp": 0.01026406,
      "balance_loss_clip": 1.04314041,
      "balance_loss_mlp": 1.01931298,
      "epoch": 0.8239042866590512,
      "flos": 27309023821440.0,
      "grad_norm": 2.0323512236743406,
      "language_loss": 0.53235209,
      "learning_rate": 3.1642175881663155e-07,
      "loss": 0.55402237,
      "num_input_tokens_seen": 147969245,
      "step": 6852,
      "time_per_iteration": 2.7231178283691406
    },
    {
      "auxiliary_loss_clip": 0.01164319,
      "auxiliary_loss_mlp": 0.01022013,
      "balance_loss_clip": 1.04529381,
      "balance_loss_mlp": 1.01539648,
      "epoch": 0.8240245295496904,
      "flos": 21726351187200.0,
      "grad_norm": 2.450953557615833,
      "language_loss": 0.83682191,
      "learning_rate": 3.160013925537537e-07,
      "loss": 0.85868526,
      "num_input_tokens_seen": 147990080,
      "step": 6853,
      "time_per_iteration": 2.653960704803467
    },
    {
      "auxiliary_loss_clip": 0.01127298,
      "auxiliary_loss_mlp": 0.01020404,
      "balance_loss_clip": 1.04135597,
      "balance_loss_mlp": 1.01324272,
      "epoch": 0.8241447724403295,
      "flos": 20009318279040.0,
      "grad_norm": 2.173573669098559,
      "language_loss": 0.75777602,
      "learning_rate": 3.155812817521266e-07,
      "loss": 0.77925301,
      "num_input_tokens_seen": 148010455,
      "step": 6854,
      "time_per_iteration": 2.707383632659912
    },
    {
      "auxiliary_loss_clip": 0.01142311,
      "auxiliary_loss_mlp": 0.01024594,
      "balance_loss_clip": 1.04642439,
      "balance_loss_mlp": 1.01746476,
      "epoch": 0.8242650153309685,
      "flos": 22272983337600.0,
      "grad_norm": 2.093092814069572,
      "language_loss": 0.78101254,
      "learning_rate": 3.151614264754787e-07,
      "loss": 0.80268157,
      "num_input_tokens_seen": 148028400,
      "step": 6855,
      "time_per_iteration": 2.6370902061462402
    },
    {
      "auxiliary_loss_clip": 0.0116525,
      "auxiliary_loss_mlp": 0.01022293,
      "balance_loss_clip": 1.04357171,
      "balance_loss_mlp": 1.01489592,
      "epoch": 0.8243852582216077,
      "flos": 22309971367680.0,
      "grad_norm": 2.951230582783915,
      "language_loss": 0.79173386,
      "learning_rate": 3.147418267875035e-07,
      "loss": 0.81360924,
      "num_input_tokens_seen": 148046530,
      "step": 6856,
      "time_per_iteration": 2.67295503616333
    },
    {
      "auxiliary_loss_clip": 0.01090905,
      "auxiliary_loss_mlp": 0.00762078,
      "balance_loss_clip": 1.03502238,
      "balance_loss_mlp": 1.00030792,
      "epoch": 0.8245055011122467,
      "flos": 24645421756800.0,
      "grad_norm": 2.852149866676685,
      "language_loss": 0.65997863,
      "learning_rate": 3.1432248275185315e-07,
      "loss": 0.67850852,
      "num_input_tokens_seen": 148067040,
      "step": 6857,
      "time_per_iteration": 2.7662220001220703
    },
    {
      "auxiliary_loss_clip": 0.01149363,
      "auxiliary_loss_mlp": 0.01026665,
      "balance_loss_clip": 1.04480612,
      "balance_loss_mlp": 1.01977432,
      "epoch": 0.8246257440028858,
      "flos": 17487275713920.0,
      "grad_norm": 2.022430315648165,
      "language_loss": 0.76989788,
      "learning_rate": 3.139033944321412e-07,
      "loss": 0.79165816,
      "num_input_tokens_seen": 148084400,
      "step": 6858,
      "time_per_iteration": 2.600883722305298
    },
    {
      "auxiliary_loss_clip": 0.01154287,
      "auxiliary_loss_mlp": 0.01026902,
      "balance_loss_clip": 1.04316688,
      "balance_loss_mlp": 1.01956475,
      "epoch": 0.824745986893525,
      "flos": 25010130499200.0,
      "grad_norm": 2.3256419937891617,
      "language_loss": 0.78577185,
      "learning_rate": 3.1348456189194507e-07,
      "loss": 0.80758375,
      "num_input_tokens_seen": 148104860,
      "step": 6859,
      "time_per_iteration": 2.6353957653045654
    },
    {
      "auxiliary_loss_clip": 0.01113992,
      "auxiliary_loss_mlp": 0.01026649,
      "balance_loss_clip": 1.03877759,
      "balance_loss_mlp": 1.01937735,
      "epoch": 0.824866229784164,
      "flos": 18772698798720.0,
      "grad_norm": 1.5531458437604966,
      "language_loss": 0.82940346,
      "learning_rate": 3.1306598519479876e-07,
      "loss": 0.85080987,
      "num_input_tokens_seen": 148124680,
      "step": 6860,
      "time_per_iteration": 2.679022789001465
    },
    {
      "auxiliary_loss_clip": 0.0113496,
      "auxiliary_loss_mlp": 0.0102252,
      "balance_loss_clip": 1.04322004,
      "balance_loss_mlp": 1.01583207,
      "epoch": 0.8249864726748031,
      "flos": 23842171866240.0,
      "grad_norm": 1.7889739653856476,
      "language_loss": 0.78536642,
      "learning_rate": 3.1264766440420177e-07,
      "loss": 0.80694127,
      "num_input_tokens_seen": 148147150,
      "step": 6861,
      "time_per_iteration": 2.7648589611053467
    },
    {
      "auxiliary_loss_clip": 0.01146401,
      "auxiliary_loss_mlp": 0.01028113,
      "balance_loss_clip": 1.04280543,
      "balance_loss_mlp": 1.02133858,
      "epoch": 0.8251067155654422,
      "flos": 20303103617280.0,
      "grad_norm": 1.949835237882574,
      "language_loss": 0.69111681,
      "learning_rate": 3.122295995836124e-07,
      "loss": 0.7128619,
      "num_input_tokens_seen": 148167020,
      "step": 6862,
      "time_per_iteration": 2.65055775642395
    },
    {
      "auxiliary_loss_clip": 0.01154074,
      "auxiliary_loss_mlp": 0.01026739,
      "balance_loss_clip": 1.04191899,
      "balance_loss_mlp": 1.01916885,
      "epoch": 0.8252269584560813,
      "flos": 25009699536000.0,
      "grad_norm": 2.199715816738437,
      "language_loss": 0.77306986,
      "learning_rate": 3.118117907964508e-07,
      "loss": 0.79487789,
      "num_input_tokens_seen": 148188965,
      "step": 6863,
      "time_per_iteration": 2.6835062503814697
    },
    {
      "auxiliary_loss_clip": 0.01130718,
      "auxiliary_loss_mlp": 0.01026144,
      "balance_loss_clip": 1.04249382,
      "balance_loss_mlp": 1.01942337,
      "epoch": 0.8253472013467203,
      "flos": 17128564542720.0,
      "grad_norm": 1.9240802559531334,
      "language_loss": 0.80223799,
      "learning_rate": 3.1139423810609856e-07,
      "loss": 0.82380658,
      "num_input_tokens_seen": 148205660,
      "step": 6864,
      "time_per_iteration": 2.684217691421509
    },
    {
      "auxiliary_loss_clip": 0.01163564,
      "auxiliary_loss_mlp": 0.01023364,
      "balance_loss_clip": 1.04194605,
      "balance_loss_mlp": 1.01572251,
      "epoch": 0.8254674442373595,
      "flos": 22414794232320.0,
      "grad_norm": 2.1033585067865097,
      "language_loss": 0.75494015,
      "learning_rate": 3.1097694157589714e-07,
      "loss": 0.77680945,
      "num_input_tokens_seen": 148225545,
      "step": 6865,
      "time_per_iteration": 2.668067455291748
    },
    {
      "auxiliary_loss_clip": 0.01150944,
      "auxiliary_loss_mlp": 0.0102819,
      "balance_loss_clip": 1.04736543,
      "balance_loss_mlp": 1.02096617,
      "epoch": 0.8255876871279986,
      "flos": 24786765774720.0,
      "grad_norm": 2.964766454436364,
      "language_loss": 0.7608614,
      "learning_rate": 3.105599012691511e-07,
      "loss": 0.78265274,
      "num_input_tokens_seen": 148243975,
      "step": 6866,
      "time_per_iteration": 2.691396951675415
    },
    {
      "auxiliary_loss_clip": 0.01148503,
      "auxiliary_loss_mlp": 0.01023933,
      "balance_loss_clip": 1.04434156,
      "balance_loss_mlp": 1.0172869,
      "epoch": 0.8257079300186376,
      "flos": 27455431656960.0,
      "grad_norm": 1.8177448676157129,
      "language_loss": 0.82310081,
      "learning_rate": 3.101431172491249e-07,
      "loss": 0.84482515,
      "num_input_tokens_seen": 148265520,
      "step": 6867,
      "time_per_iteration": 3.667149066925049
    },
    {
      "auxiliary_loss_clip": 0.01126367,
      "auxiliary_loss_mlp": 0.00762474,
      "balance_loss_clip": 1.03956985,
      "balance_loss_mlp": 1.00035715,
      "epoch": 0.8258281729092768,
      "flos": 16471866142080.0,
      "grad_norm": 2.231629388582422,
      "language_loss": 0.72044814,
      "learning_rate": 3.097265895790444e-07,
      "loss": 0.73933649,
      "num_input_tokens_seen": 148283730,
      "step": 6868,
      "time_per_iteration": 2.781785726547241
    },
    {
      "auxiliary_loss_clip": 0.01125108,
      "auxiliary_loss_mlp": 0.01027682,
      "balance_loss_clip": 1.04231453,
      "balance_loss_mlp": 1.02039266,
      "epoch": 0.8259484157999158,
      "flos": 21433822824960.0,
      "grad_norm": 2.115833492978135,
      "language_loss": 0.83084416,
      "learning_rate": 3.093103183220962e-07,
      "loss": 0.85237217,
      "num_input_tokens_seen": 148303775,
      "step": 6869,
      "time_per_iteration": 3.7169461250305176
    },
    {
      "auxiliary_loss_clip": 0.01054864,
      "auxiliary_loss_mlp": 0.01000525,
      "balance_loss_clip": 1.01064658,
      "balance_loss_mlp": 0.99936849,
      "epoch": 0.8260686586905549,
      "flos": 58322342453760.0,
      "grad_norm": 0.8225836507519573,
      "language_loss": 0.59376675,
      "learning_rate": 3.0889430354142796e-07,
      "loss": 0.61432064,
      "num_input_tokens_seen": 148365285,
      "step": 6870,
      "time_per_iteration": 3.225717067718506
    },
    {
      "auxiliary_loss_clip": 0.01123529,
      "auxiliary_loss_mlp": 0.01025522,
      "balance_loss_clip": 1.03915203,
      "balance_loss_mlp": 1.01804781,
      "epoch": 0.826188901581194,
      "flos": 27527288814720.0,
      "grad_norm": 1.9075894233445656,
      "language_loss": 0.6996727,
      "learning_rate": 3.084785453001497e-07,
      "loss": 0.72116321,
      "num_input_tokens_seen": 148386200,
      "step": 6871,
      "time_per_iteration": 2.730030059814453
    },
    {
      "auxiliary_loss_clip": 0.01141599,
      "auxiliary_loss_mlp": 0.00761658,
      "balance_loss_clip": 1.04596353,
      "balance_loss_mlp": 1.0003159,
      "epoch": 0.8263091444718331,
      "flos": 23696051339520.0,
      "grad_norm": 2.7753753906915457,
      "language_loss": 0.8199389,
      "learning_rate": 3.080630436613314e-07,
      "loss": 0.8389715,
      "num_input_tokens_seen": 148403970,
      "step": 6872,
      "time_per_iteration": 4.576741695404053
    },
    {
      "auxiliary_loss_clip": 0.01142384,
      "auxiliary_loss_mlp": 0.0102135,
      "balance_loss_clip": 1.04177213,
      "balance_loss_mlp": 1.01445079,
      "epoch": 0.8264293873624722,
      "flos": 17165157523200.0,
      "grad_norm": 2.1358032959107187,
      "language_loss": 0.86287141,
      "learning_rate": 3.076477986880039e-07,
      "loss": 0.88450873,
      "num_input_tokens_seen": 148421765,
      "step": 6873,
      "time_per_iteration": 2.6175484657287598
    },
    {
      "auxiliary_loss_clip": 0.01136373,
      "auxiliary_loss_mlp": 0.01024142,
      "balance_loss_clip": 1.0443691,
      "balance_loss_mlp": 1.01696587,
      "epoch": 0.8265496302531112,
      "flos": 24098645952000.0,
      "grad_norm": 2.253268624515649,
      "language_loss": 0.69369936,
      "learning_rate": 3.0723281044315986e-07,
      "loss": 0.71530455,
      "num_input_tokens_seen": 148443720,
      "step": 6874,
      "time_per_iteration": 2.655113697052002
    },
    {
      "auxiliary_loss_clip": 0.01162482,
      "auxiliary_loss_mlp": 0.01025041,
      "balance_loss_clip": 1.04431462,
      "balance_loss_mlp": 1.01839757,
      "epoch": 0.8266698731437504,
      "flos": 14099894599680.0,
      "grad_norm": 2.072491775220985,
      "language_loss": 0.76558852,
      "learning_rate": 3.068180789897521e-07,
      "loss": 0.78746367,
      "num_input_tokens_seen": 148462130,
      "step": 6875,
      "time_per_iteration": 2.5548360347747803
    },
    {
      "auxiliary_loss_clip": 0.01153416,
      "auxiliary_loss_mlp": 0.01028534,
      "balance_loss_clip": 1.04295528,
      "balance_loss_mlp": 1.02091026,
      "epoch": 0.8267901160343895,
      "flos": 30777563715840.0,
      "grad_norm": 1.458009570379395,
      "language_loss": 0.81511939,
      "learning_rate": 3.064036043906966e-07,
      "loss": 0.83693892,
      "num_input_tokens_seen": 148485570,
      "step": 6876,
      "time_per_iteration": 2.671701431274414
    },
    {
      "auxiliary_loss_clip": 0.01133442,
      "auxiliary_loss_mlp": 0.01023876,
      "balance_loss_clip": 1.04322791,
      "balance_loss_mlp": 1.01637173,
      "epoch": 0.8269103589250285,
      "flos": 40624915242240.0,
      "grad_norm": 1.9827016228190217,
      "language_loss": 0.67826402,
      "learning_rate": 3.059893867088668e-07,
      "loss": 0.69983721,
      "num_input_tokens_seen": 148509715,
      "step": 6877,
      "time_per_iteration": 2.8650858402252197
    },
    {
      "auxiliary_loss_clip": 0.0115195,
      "auxiliary_loss_mlp": 0.01025434,
      "balance_loss_clip": 1.04621649,
      "balance_loss_mlp": 1.01837957,
      "epoch": 0.8270306018156677,
      "flos": 30263645877120.0,
      "grad_norm": 1.998923847080477,
      "language_loss": 0.67202789,
      "learning_rate": 3.055754260071004e-07,
      "loss": 0.6938017,
      "num_input_tokens_seen": 148532010,
      "step": 6878,
      "time_per_iteration": 2.6947569847106934
    },
    {
      "auxiliary_loss_clip": 0.01152336,
      "auxiliary_loss_mlp": 0.01023093,
      "balance_loss_clip": 1.04327726,
      "balance_loss_mlp": 1.01660514,
      "epoch": 0.8271508447063067,
      "flos": 25226599812480.0,
      "grad_norm": 4.023478785188813,
      "language_loss": 0.73942006,
      "learning_rate": 3.051617223481948e-07,
      "loss": 0.76117432,
      "num_input_tokens_seen": 148553330,
      "step": 6879,
      "time_per_iteration": 2.6372623443603516
    },
    {
      "auxiliary_loss_clip": 0.0113727,
      "auxiliary_loss_mlp": 0.01027581,
      "balance_loss_clip": 1.04294944,
      "balance_loss_mlp": 1.01989818,
      "epoch": 0.8272710875969458,
      "flos": 17566602900480.0,
      "grad_norm": 1.938147120761069,
      "language_loss": 0.75182259,
      "learning_rate": 3.047482757949078e-07,
      "loss": 0.77347112,
      "num_input_tokens_seen": 148570960,
      "step": 6880,
      "time_per_iteration": 2.675318956375122
    },
    {
      "auxiliary_loss_clip": 0.01120053,
      "auxiliary_loss_mlp": 0.00762017,
      "balance_loss_clip": 1.03933787,
      "balance_loss_mlp": 1.00028944,
      "epoch": 0.827391330487585,
      "flos": 19755465886080.0,
      "grad_norm": 2.021647288884602,
      "language_loss": 0.85845852,
      "learning_rate": 3.043350864099605e-07,
      "loss": 0.87727928,
      "num_input_tokens_seen": 148589520,
      "step": 6881,
      "time_per_iteration": 2.6630618572235107
    },
    {
      "auxiliary_loss_clip": 0.01154736,
      "auxiliary_loss_mlp": 0.01023861,
      "balance_loss_clip": 1.0430789,
      "balance_loss_mlp": 1.01655674,
      "epoch": 0.827511573378224,
      "flos": 16835174254080.0,
      "grad_norm": 2.243049787858883,
      "language_loss": 0.80511105,
      "learning_rate": 3.039221542560315e-07,
      "loss": 0.82689703,
      "num_input_tokens_seen": 148606085,
      "step": 6882,
      "time_per_iteration": 2.5902864933013916
    },
    {
      "auxiliary_loss_clip": 0.01153419,
      "auxiliary_loss_mlp": 0.01020627,
      "balance_loss_clip": 1.04627717,
      "balance_loss_mlp": 1.01373923,
      "epoch": 0.8276318162688631,
      "flos": 18369242259840.0,
      "grad_norm": 1.8147681616399893,
      "language_loss": 0.73521298,
      "learning_rate": 3.0350947939576356e-07,
      "loss": 0.75695348,
      "num_input_tokens_seen": 148625240,
      "step": 6883,
      "time_per_iteration": 2.6904993057250977
    },
    {
      "auxiliary_loss_clip": 0.01158903,
      "auxiliary_loss_mlp": 0.01026369,
      "balance_loss_clip": 1.04635811,
      "balance_loss_mlp": 1.01894212,
      "epoch": 0.8277520591595022,
      "flos": 19352691705600.0,
      "grad_norm": 1.649741636592778,
      "language_loss": 0.72427964,
      "learning_rate": 3.0309706189175876e-07,
      "loss": 0.74613237,
      "num_input_tokens_seen": 148645075,
      "step": 6884,
      "time_per_iteration": 2.7054338455200195
    },
    {
      "auxiliary_loss_clip": 0.0104523,
      "auxiliary_loss_mlp": 0.01000808,
      "balance_loss_clip": 1.01041222,
      "balance_loss_mlp": 0.99968702,
      "epoch": 0.8278723020501413,
      "flos": 67918858329600.0,
      "grad_norm": 0.7696649978691301,
      "language_loss": 0.57408834,
      "learning_rate": 3.0268490180658045e-07,
      "loss": 0.5945487,
      "num_input_tokens_seen": 148707855,
      "step": 6885,
      "time_per_iteration": 3.2412540912628174
    },
    {
      "auxiliary_loss_clip": 0.01171355,
      "auxiliary_loss_mlp": 0.01026037,
      "balance_loss_clip": 1.04918337,
      "balance_loss_mlp": 1.01908994,
      "epoch": 0.8279925449407803,
      "flos": 18185738653440.0,
      "grad_norm": 2.8077949817409174,
      "language_loss": 0.79323328,
      "learning_rate": 3.0227299920275305e-07,
      "loss": 0.81520724,
      "num_input_tokens_seen": 148724170,
      "step": 6886,
      "time_per_iteration": 2.5954487323760986
    },
    {
      "auxiliary_loss_clip": 0.01128624,
      "auxiliary_loss_mlp": 0.01026779,
      "balance_loss_clip": 1.04439616,
      "balance_loss_mlp": 1.01874399,
      "epoch": 0.8281127878314195,
      "flos": 20631434860800.0,
      "grad_norm": 2.1696416595507713,
      "language_loss": 0.85906786,
      "learning_rate": 3.018613541427613e-07,
      "loss": 0.88062191,
      "num_input_tokens_seen": 148743690,
      "step": 6887,
      "time_per_iteration": 2.7092034816741943
    },
    {
      "auxiliary_loss_clip": 0.01162889,
      "auxiliary_loss_mlp": 0.01026806,
      "balance_loss_clip": 1.04309034,
      "balance_loss_mlp": 1.01963246,
      "epoch": 0.8282330307220586,
      "flos": 18004282122240.0,
      "grad_norm": 1.7644289077917026,
      "language_loss": 0.73549163,
      "learning_rate": 3.0144996668905243e-07,
      "loss": 0.75738859,
      "num_input_tokens_seen": 148761070,
      "step": 6888,
      "time_per_iteration": 2.56958270072937
    },
    {
      "auxiliary_loss_clip": 0.01098022,
      "auxiliary_loss_mlp": 0.00761963,
      "balance_loss_clip": 1.03678775,
      "balance_loss_mlp": 1.00029647,
      "epoch": 0.8283532736126976,
      "flos": 20084120352000.0,
      "grad_norm": 1.9885599824076756,
      "language_loss": 0.82546413,
      "learning_rate": 3.010388369040331e-07,
      "loss": 0.844064,
      "num_input_tokens_seen": 148779730,
      "step": 6889,
      "time_per_iteration": 2.8491427898406982
    },
    {
      "auxiliary_loss_clip": 0.01148727,
      "auxiliary_loss_mlp": 0.01027118,
      "balance_loss_clip": 1.04416275,
      "balance_loss_mlp": 1.0195725,
      "epoch": 0.8284735165033368,
      "flos": 31868421805440.0,
      "grad_norm": 2.1587240715254765,
      "language_loss": 0.8301208,
      "learning_rate": 3.0062796485007156e-07,
      "loss": 0.85187924,
      "num_input_tokens_seen": 148800670,
      "step": 6890,
      "time_per_iteration": 2.711315631866455
    },
    {
      "auxiliary_loss_clip": 0.01164732,
      "auxiliary_loss_mlp": 0.0076239,
      "balance_loss_clip": 1.04417527,
      "balance_loss_mlp": 1.00032163,
      "epoch": 0.8285937593939758,
      "flos": 26651319840000.0,
      "grad_norm": 3.064607766289369,
      "language_loss": 0.65458012,
      "learning_rate": 3.002173505894965e-07,
      "loss": 0.67385131,
      "num_input_tokens_seen": 148819820,
      "step": 6891,
      "time_per_iteration": 2.621171236038208
    },
    {
      "auxiliary_loss_clip": 0.0115783,
      "auxiliary_loss_mlp": 0.01028173,
      "balance_loss_clip": 1.04501128,
      "balance_loss_mlp": 1.02051091,
      "epoch": 0.8287140022846149,
      "flos": 20193683811840.0,
      "grad_norm": 4.732032073761448,
      "language_loss": 0.63101268,
      "learning_rate": 2.998069941845973e-07,
      "loss": 0.65287268,
      "num_input_tokens_seen": 148838890,
      "step": 6892,
      "time_per_iteration": 2.5424044132232666
    },
    {
      "auxiliary_loss_clip": 0.010641,
      "auxiliary_loss_mlp": 0.01001911,
      "balance_loss_clip": 1.0102818,
      "balance_loss_mlp": 1.00083256,
      "epoch": 0.8288342451752541,
      "flos": 70755980019840.0,
      "grad_norm": 0.7639893419810463,
      "language_loss": 0.5742805,
      "learning_rate": 2.993968956976258e-07,
      "loss": 0.59494054,
      "num_input_tokens_seen": 148906635,
      "step": 6893,
      "time_per_iteration": 4.207201719284058
    },
    {
      "auxiliary_loss_clip": 0.01173939,
      "auxiliary_loss_mlp": 0.01026638,
      "balance_loss_clip": 1.04838729,
      "balance_loss_mlp": 1.01946783,
      "epoch": 0.8289544880658931,
      "flos": 24572235795840.0,
      "grad_norm": 2.629436096199152,
      "language_loss": 0.70249015,
      "learning_rate": 2.9898705519079313e-07,
      "loss": 0.72449589,
      "num_input_tokens_seen": 148925740,
      "step": 6894,
      "time_per_iteration": 2.6489667892456055
    },
    {
      "auxiliary_loss_clip": 0.01131731,
      "auxiliary_loss_mlp": 0.01020901,
      "balance_loss_clip": 1.04176688,
      "balance_loss_mlp": 1.01403785,
      "epoch": 0.8290747309565322,
      "flos": 22273378387200.0,
      "grad_norm": 1.7304504399810638,
      "language_loss": 0.74345016,
      "learning_rate": 2.985774727262715e-07,
      "loss": 0.76497638,
      "num_input_tokens_seen": 148944585,
      "step": 6895,
      "time_per_iteration": 2.8332114219665527
    },
    {
      "auxiliary_loss_clip": 0.01163331,
      "auxiliary_loss_mlp": 0.01023846,
      "balance_loss_clip": 1.04386532,
      "balance_loss_mlp": 1.01690769,
      "epoch": 0.8291949738471713,
      "flos": 23255570856960.0,
      "grad_norm": 1.9033257590703767,
      "language_loss": 0.81624132,
      "learning_rate": 2.981681483661949e-07,
      "loss": 0.83811307,
      "num_input_tokens_seen": 148964170,
      "step": 6896,
      "time_per_iteration": 3.7558043003082275
    },
    {
      "auxiliary_loss_clip": 0.01153868,
      "auxiliary_loss_mlp": 0.0102583,
      "balance_loss_clip": 1.04669404,
      "balance_loss_mlp": 1.01862717,
      "epoch": 0.8293152167378104,
      "flos": 52555768185600.0,
      "grad_norm": 1.8195632971407565,
      "language_loss": 0.7141301,
      "learning_rate": 2.9775908217265633e-07,
      "loss": 0.7359271,
      "num_input_tokens_seen": 148989405,
      "step": 6897,
      "time_per_iteration": 2.959160804748535
    },
    {
      "auxiliary_loss_clip": 0.01011354,
      "auxiliary_loss_mlp": 0.01000949,
      "balance_loss_clip": 1.01029181,
      "balance_loss_mlp": 0.99979228,
      "epoch": 0.8294354596284494,
      "flos": 63356156294400.0,
      "grad_norm": 0.8293398263021645,
      "language_loss": 0.5034461,
      "learning_rate": 2.9735027420771253e-07,
      "loss": 0.52356911,
      "num_input_tokens_seen": 149049740,
      "step": 6898,
      "time_per_iteration": 4.367146015167236
    },
    {
      "auxiliary_loss_clip": 0.01132803,
      "auxiliary_loss_mlp": 0.01023917,
      "balance_loss_clip": 1.04462063,
      "balance_loss_mlp": 1.01734877,
      "epoch": 0.8295557025190886,
      "flos": 24827021942400.0,
      "grad_norm": 1.6132516790129687,
      "language_loss": 0.71373212,
      "learning_rate": 2.969417245333774e-07,
      "loss": 0.73529935,
      "num_input_tokens_seen": 149069120,
      "step": 6899,
      "time_per_iteration": 4.0404932498931885
    },
    {
      "auxiliary_loss_clip": 0.0111941,
      "auxiliary_loss_mlp": 0.01020754,
      "balance_loss_clip": 1.04331577,
      "balance_loss_mlp": 1.01393509,
      "epoch": 0.8296759454097277,
      "flos": 25118580637440.0,
      "grad_norm": 2.296449887913235,
      "language_loss": 0.77969223,
      "learning_rate": 2.9653343321162915e-07,
      "loss": 0.80109388,
      "num_input_tokens_seen": 149088630,
      "step": 6900,
      "time_per_iteration": 2.9579837322235107
    },
    {
      "auxiliary_loss_clip": 0.01125565,
      "auxiliary_loss_mlp": 0.01030646,
      "balance_loss_clip": 1.04472113,
      "balance_loss_mlp": 1.02315116,
      "epoch": 0.8297961883003667,
      "flos": 24132581326080.0,
      "grad_norm": 2.1239738408946214,
      "language_loss": 0.65194774,
      "learning_rate": 2.9612540030440446e-07,
      "loss": 0.67350984,
      "num_input_tokens_seen": 149109175,
      "step": 6901,
      "time_per_iteration": 2.7515783309936523
    },
    {
      "auxiliary_loss_clip": 0.01044333,
      "auxiliary_loss_mlp": 0.01001402,
      "balance_loss_clip": 1.01100278,
      "balance_loss_mlp": 1.00024021,
      "epoch": 0.8299164311910058,
      "flos": 67446561375360.0,
      "grad_norm": 0.8484105544830768,
      "language_loss": 0.64074349,
      "learning_rate": 2.9571762587360206e-07,
      "loss": 0.66120088,
      "num_input_tokens_seen": 149165560,
      "step": 6902,
      "time_per_iteration": 3.1909103393554688
    },
    {
      "auxiliary_loss_clip": 0.01105254,
      "auxiliary_loss_mlp": 0.01029135,
      "balance_loss_clip": 1.03401589,
      "balance_loss_mlp": 1.02185166,
      "epoch": 0.8300366740816449,
      "flos": 25228682801280.0,
      "grad_norm": 1.8234935162403185,
      "language_loss": 0.74050653,
      "learning_rate": 2.953101099810806e-07,
      "loss": 0.76185042,
      "num_input_tokens_seen": 149185165,
      "step": 6903,
      "time_per_iteration": 2.736765146255493
    },
    {
      "auxiliary_loss_clip": 0.01148841,
      "auxiliary_loss_mlp": 0.01026552,
      "balance_loss_clip": 1.04657149,
      "balance_loss_mlp": 1.01981354,
      "epoch": 0.830156916972284,
      "flos": 18041018757120.0,
      "grad_norm": 2.201152945331171,
      "language_loss": 0.8320533,
      "learning_rate": 2.9490285268865965e-07,
      "loss": 0.85380727,
      "num_input_tokens_seen": 149202655,
      "step": 6904,
      "time_per_iteration": 2.5887012481689453
    },
    {
      "auxiliary_loss_clip": 0.01158876,
      "auxiliary_loss_mlp": 0.01027848,
      "balance_loss_clip": 1.04755235,
      "balance_loss_mlp": 1.02074337,
      "epoch": 0.830277159862923,
      "flos": 26322485806080.0,
      "grad_norm": 2.049425247208737,
      "language_loss": 0.79858911,
      "learning_rate": 2.9449585405812085e-07,
      "loss": 0.82045633,
      "num_input_tokens_seen": 149220035,
      "step": 6905,
      "time_per_iteration": 2.7241194248199463
    },
    {
      "auxiliary_loss_clip": 0.01127116,
      "auxiliary_loss_mlp": 0.01025406,
      "balance_loss_clip": 1.042521,
      "balance_loss_mlp": 1.01827765,
      "epoch": 0.8303974027535622,
      "flos": 19938861751680.0,
      "grad_norm": 1.8403439441451048,
      "language_loss": 0.73768473,
      "learning_rate": 2.940891141512043e-07,
      "loss": 0.75920999,
      "num_input_tokens_seen": 149238055,
      "step": 6906,
      "time_per_iteration": 2.7087695598602295
    },
    {
      "auxiliary_loss_clip": 0.01137095,
      "auxiliary_loss_mlp": 0.01030369,
      "balance_loss_clip": 1.04198551,
      "balance_loss_mlp": 1.02312124,
      "epoch": 0.8305176456442013,
      "flos": 17165552572800.0,
      "grad_norm": 2.198733560498124,
      "language_loss": 0.72272718,
      "learning_rate": 2.9368263302961385e-07,
      "loss": 0.74440181,
      "num_input_tokens_seen": 149256755,
      "step": 6907,
      "time_per_iteration": 2.6908416748046875
    },
    {
      "auxiliary_loss_clip": 0.010942,
      "auxiliary_loss_mlp": 0.01025847,
      "balance_loss_clip": 1.03566265,
      "balance_loss_mlp": 1.01814342,
      "epoch": 0.8306378885348403,
      "flos": 25627614226560.0,
      "grad_norm": 1.9231458316807981,
      "language_loss": 0.79334396,
      "learning_rate": 2.9327641075501075e-07,
      "loss": 0.81454444,
      "num_input_tokens_seen": 149275745,
      "step": 6908,
      "time_per_iteration": 2.8732964992523193
    },
    {
      "auxiliary_loss_clip": 0.01130173,
      "auxiliary_loss_mlp": 0.01027956,
      "balance_loss_clip": 1.03926826,
      "balance_loss_mlp": 1.02036285,
      "epoch": 0.8307581314254795,
      "flos": 33947864985600.0,
      "grad_norm": 2.8896379394197784,
      "language_loss": 0.66836548,
      "learning_rate": 2.9287044738901866e-07,
      "loss": 0.68994671,
      "num_input_tokens_seen": 149293730,
      "step": 6909,
      "time_per_iteration": 2.906161069869995
    },
    {
      "auxiliary_loss_clip": 0.01152362,
      "auxiliary_loss_mlp": 0.00761856,
      "balance_loss_clip": 1.043347,
      "balance_loss_mlp": 1.00036192,
      "epoch": 0.8308783743161186,
      "flos": 17562724231680.0,
      "grad_norm": 2.2784822058040493,
      "language_loss": 0.90547252,
      "learning_rate": 2.9246474299322274e-07,
      "loss": 0.92461467,
      "num_input_tokens_seen": 149309290,
      "step": 6910,
      "time_per_iteration": 2.729814052581787
    },
    {
      "auxiliary_loss_clip": 0.01030367,
      "auxiliary_loss_mlp": 0.01002008,
      "balance_loss_clip": 1.00917411,
      "balance_loss_mlp": 1.00093496,
      "epoch": 0.8309986172067576,
      "flos": 69412885649280.0,
      "grad_norm": 0.8876647815973384,
      "language_loss": 0.63141698,
      "learning_rate": 2.920592976291678e-07,
      "loss": 0.65174073,
      "num_input_tokens_seen": 149366620,
      "step": 6911,
      "time_per_iteration": 3.222320795059204
    },
    {
      "auxiliary_loss_clip": 0.01150811,
      "auxiliary_loss_mlp": 0.01024274,
      "balance_loss_clip": 1.04295039,
      "balance_loss_mlp": 1.01654959,
      "epoch": 0.8311188600973968,
      "flos": 22309755886080.0,
      "grad_norm": 1.965992516192072,
      "language_loss": 0.80837864,
      "learning_rate": 2.916541113583595e-07,
      "loss": 0.8301295,
      "num_input_tokens_seen": 149385120,
      "step": 6912,
      "time_per_iteration": 2.684483766555786
    },
    {
      "auxiliary_loss_clip": 0.01130691,
      "auxiliary_loss_mlp": 0.01026734,
      "balance_loss_clip": 1.04552412,
      "balance_loss_mlp": 1.0190568,
      "epoch": 0.8312391029880358,
      "flos": 18770077105920.0,
      "grad_norm": 2.1070939777772084,
      "language_loss": 0.66598034,
      "learning_rate": 2.912491842422642e-07,
      "loss": 0.6875546,
      "num_input_tokens_seen": 149402825,
      "step": 6913,
      "time_per_iteration": 2.6930036544799805
    },
    {
      "auxiliary_loss_clip": 0.0115166,
      "auxiliary_loss_mlp": 0.01027434,
      "balance_loss_clip": 1.04362988,
      "balance_loss_mlp": 1.02048993,
      "epoch": 0.8313593458786749,
      "flos": 20376648714240.0,
      "grad_norm": 1.6819199638623243,
      "language_loss": 0.71018171,
      "learning_rate": 2.9084451634230857e-07,
      "loss": 0.73197263,
      "num_input_tokens_seen": 149422125,
      "step": 6914,
      "time_per_iteration": 2.6302640438079834
    },
    {
      "auxiliary_loss_clip": 0.01123428,
      "auxiliary_loss_mlp": 0.01028328,
      "balance_loss_clip": 1.04107094,
      "balance_loss_mlp": 1.02105951,
      "epoch": 0.831479588769314,
      "flos": 32124069878400.0,
      "grad_norm": 2.3521413154079207,
      "language_loss": 0.71246755,
      "learning_rate": 2.9044010771988125e-07,
      "loss": 0.73398513,
      "num_input_tokens_seen": 149441940,
      "step": 6915,
      "time_per_iteration": 2.8047239780426025
    },
    {
      "auxiliary_loss_clip": 0.0113191,
      "auxiliary_loss_mlp": 0.01030803,
      "balance_loss_clip": 1.0428772,
      "balance_loss_mlp": 1.02365637,
      "epoch": 0.8315998316599531,
      "flos": 45185929338240.0,
      "grad_norm": 1.8169549431571592,
      "language_loss": 0.71967554,
      "learning_rate": 2.900359584363303e-07,
      "loss": 0.74130267,
      "num_input_tokens_seen": 149465045,
      "step": 6916,
      "time_per_iteration": 2.9225351810455322
    },
    {
      "auxiliary_loss_clip": 0.01107633,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.04263592,
      "balance_loss_mlp": 1.01683879,
      "epoch": 0.8317200745505922,
      "flos": 18363747479040.0,
      "grad_norm": 2.292316795099922,
      "language_loss": 0.84706467,
      "learning_rate": 2.8963206855296494e-07,
      "loss": 0.86838871,
      "num_input_tokens_seen": 149481285,
      "step": 6917,
      "time_per_iteration": 2.7519233226776123
    },
    {
      "auxiliary_loss_clip": 0.01154833,
      "auxiliary_loss_mlp": 0.01026699,
      "balance_loss_clip": 1.04508829,
      "balance_loss_mlp": 1.01963043,
      "epoch": 0.8318403174412313,
      "flos": 24206557386240.0,
      "grad_norm": 1.8113582486503708,
      "language_loss": 0.77350533,
      "learning_rate": 2.892284381310548e-07,
      "loss": 0.79532069,
      "num_input_tokens_seen": 149502700,
      "step": 6918,
      "time_per_iteration": 2.731346607208252
    },
    {
      "auxiliary_loss_clip": 0.01134044,
      "auxiliary_loss_mlp": 0.01026609,
      "balance_loss_clip": 1.04204237,
      "balance_loss_mlp": 1.01911092,
      "epoch": 0.8319605603318704,
      "flos": 22418780641920.0,
      "grad_norm": 4.695828968899663,
      "language_loss": 0.72206217,
      "learning_rate": 2.888250672318302e-07,
      "loss": 0.74366874,
      "num_input_tokens_seen": 149520100,
      "step": 6919,
      "time_per_iteration": 3.507978677749634
    },
    {
      "auxiliary_loss_clip": 0.01169787,
      "auxiliary_loss_mlp": 0.01025642,
      "balance_loss_clip": 1.04887509,
      "balance_loss_mlp": 1.01849198,
      "epoch": 0.8320808032225094,
      "flos": 37414501459200.0,
      "grad_norm": 7.529636998992757,
      "language_loss": 0.68715096,
      "learning_rate": 2.884219559164831e-07,
      "loss": 0.70910525,
      "num_input_tokens_seen": 149543245,
      "step": 6920,
      "time_per_iteration": 2.78218936920166
    },
    {
      "auxiliary_loss_clip": 0.01152442,
      "auxiliary_loss_mlp": 0.01022702,
      "balance_loss_clip": 1.04595542,
      "balance_loss_mlp": 1.01522791,
      "epoch": 0.8322010461131486,
      "flos": 12787395638400.0,
      "grad_norm": 2.499125854533929,
      "language_loss": 0.81368315,
      "learning_rate": 2.880191042461635e-07,
      "loss": 0.83543456,
      "num_input_tokens_seen": 149559185,
      "step": 6921,
      "time_per_iteration": 3.8115532398223877
    },
    {
      "auxiliary_loss_clip": 0.0111557,
      "auxiliary_loss_mlp": 0.01022635,
      "balance_loss_clip": 1.03995848,
      "balance_loss_mlp": 1.01557469,
      "epoch": 0.8323212890037877,
      "flos": 15815455050240.0,
      "grad_norm": 2.392163793002885,
      "language_loss": 0.80242693,
      "learning_rate": 2.876165122819849e-07,
      "loss": 0.82380903,
      "num_input_tokens_seen": 149577165,
      "step": 6922,
      "time_per_iteration": 2.747318744659424
    },
    {
      "auxiliary_loss_clip": 0.01163222,
      "auxiliary_loss_mlp": 0.0102518,
      "balance_loss_clip": 1.04446042,
      "balance_loss_mlp": 1.01795316,
      "epoch": 0.8324415318944267,
      "flos": 21719276208000.0,
      "grad_norm": 1.5913941281476578,
      "language_loss": 0.79572314,
      "learning_rate": 2.872141800850201e-07,
      "loss": 0.81760716,
      "num_input_tokens_seen": 149594340,
      "step": 6923,
      "time_per_iteration": 2.5436248779296875
    },
    {
      "auxiliary_loss_clip": 0.01165914,
      "auxiliary_loss_mlp": 0.01023858,
      "balance_loss_clip": 1.04652572,
      "balance_loss_mlp": 1.01684248,
      "epoch": 0.8325617747850659,
      "flos": 34198700636160.0,
      "grad_norm": 1.820792424788784,
      "language_loss": 0.73310322,
      "learning_rate": 2.868121077163024e-07,
      "loss": 0.75500095,
      "num_input_tokens_seen": 149613895,
      "step": 6924,
      "time_per_iteration": 2.6740427017211914
    },
    {
      "auxiliary_loss_clip": 0.01153397,
      "auxiliary_loss_mlp": 0.01022502,
      "balance_loss_clip": 1.04200625,
      "balance_loss_mlp": 1.01558518,
      "epoch": 0.8326820176757049,
      "flos": 18369457741440.0,
      "grad_norm": 1.8680298109177198,
      "language_loss": 0.72317708,
      "learning_rate": 2.864102952368257e-07,
      "loss": 0.74493611,
      "num_input_tokens_seen": 149631820,
      "step": 6925,
      "time_per_iteration": 4.370633125305176
    },
    {
      "auxiliary_loss_clip": 0.01096188,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.03367221,
      "balance_loss_mlp": 1.01985359,
      "epoch": 0.832802260566344,
      "flos": 35991325716480.0,
      "grad_norm": 1.552991814931484,
      "language_loss": 0.59365404,
      "learning_rate": 2.860087427075444e-07,
      "loss": 0.6148836,
      "num_input_tokens_seen": 149656070,
      "step": 6926,
      "time_per_iteration": 2.9004361629486084
    },
    {
      "auxiliary_loss_clip": 0.01131762,
      "auxiliary_loss_mlp": 0.01026724,
      "balance_loss_clip": 1.04137349,
      "balance_loss_mlp": 1.02005768,
      "epoch": 0.8329225034569832,
      "flos": 14244434928000.0,
      "grad_norm": 2.4404291254182287,
      "language_loss": 0.85973775,
      "learning_rate": 2.856074501893744e-07,
      "loss": 0.88132262,
      "num_input_tokens_seen": 149671270,
      "step": 6927,
      "time_per_iteration": 2.6743197441101074
    },
    {
      "auxiliary_loss_clip": 0.01155476,
      "auxiliary_loss_mlp": 0.01026991,
      "balance_loss_clip": 1.04830277,
      "balance_loss_mlp": 1.01969838,
      "epoch": 0.8330427463476222,
      "flos": 18077468083200.0,
      "grad_norm": 1.8192466664158928,
      "language_loss": 0.81531829,
      "learning_rate": 2.8520641774319054e-07,
      "loss": 0.83714294,
      "num_input_tokens_seen": 149689360,
      "step": 6928,
      "time_per_iteration": 2.6387438774108887
    },
    {
      "auxiliary_loss_clip": 0.01136606,
      "auxiliary_loss_mlp": 0.01025544,
      "balance_loss_clip": 1.03857946,
      "balance_loss_mlp": 1.01830459,
      "epoch": 0.8331629892382613,
      "flos": 18040839189120.0,
      "grad_norm": 1.8777610140858128,
      "language_loss": 0.75467509,
      "learning_rate": 2.848056454298309e-07,
      "loss": 0.77629662,
      "num_input_tokens_seen": 149706685,
      "step": 6929,
      "time_per_iteration": 2.6104047298431396
    },
    {
      "auxiliary_loss_clip": 0.01137764,
      "auxiliary_loss_mlp": 0.01026921,
      "balance_loss_clip": 1.04410315,
      "balance_loss_mlp": 1.01916659,
      "epoch": 0.8332832321289004,
      "flos": 17457398576640.0,
      "grad_norm": 2.282239066253034,
      "language_loss": 0.65346092,
      "learning_rate": 2.844051333100905e-07,
      "loss": 0.67510778,
      "num_input_tokens_seen": 149724230,
      "step": 6930,
      "time_per_iteration": 2.6615798473358154
    },
    {
      "auxiliary_loss_clip": 0.01138217,
      "auxiliary_loss_mlp": 0.01023013,
      "balance_loss_clip": 1.04612589,
      "balance_loss_mlp": 1.01610756,
      "epoch": 0.8334034750195395,
      "flos": 15084852416640.0,
      "grad_norm": 1.9005960920133897,
      "language_loss": 0.83848256,
      "learning_rate": 2.840048814447269e-07,
      "loss": 0.86009485,
      "num_input_tokens_seen": 149742395,
      "step": 6931,
      "time_per_iteration": 2.655243158340454
    },
    {
      "auxiliary_loss_clip": 0.01129273,
      "auxiliary_loss_mlp": 0.01028132,
      "balance_loss_clip": 1.04056931,
      "balance_loss_mlp": 1.02051115,
      "epoch": 0.8335237179101785,
      "flos": 19427170556160.0,
      "grad_norm": 2.445454972383859,
      "language_loss": 0.74034965,
      "learning_rate": 2.836048898944587e-07,
      "loss": 0.76192373,
      "num_input_tokens_seen": 149760820,
      "step": 6932,
      "time_per_iteration": 2.645317554473877
    },
    {
      "auxiliary_loss_clip": 0.01137909,
      "auxiliary_loss_mlp": 0.01024226,
      "balance_loss_clip": 1.04434729,
      "balance_loss_mlp": 1.01773834,
      "epoch": 0.8336439608008177,
      "flos": 21762046327680.0,
      "grad_norm": 2.617799933526437,
      "language_loss": 0.72997952,
      "learning_rate": 2.832051587199642e-07,
      "loss": 0.75160092,
      "num_input_tokens_seen": 149778075,
      "step": 6933,
      "time_per_iteration": 2.659097671508789
    },
    {
      "auxiliary_loss_clip": 0.01055053,
      "auxiliary_loss_mlp": 0.01001661,
      "balance_loss_clip": 1.0094887,
      "balance_loss_mlp": 1.00057602,
      "epoch": 0.8337642036914568,
      "flos": 59702783990400.0,
      "grad_norm": 0.8106477396634857,
      "language_loss": 0.57702243,
      "learning_rate": 2.828056879818821e-07,
      "loss": 0.59758961,
      "num_input_tokens_seen": 149837150,
      "step": 6934,
      "time_per_iteration": 3.156630277633667
    },
    {
      "auxiliary_loss_clip": 0.01121564,
      "auxiliary_loss_mlp": 0.01024203,
      "balance_loss_clip": 1.03780556,
      "balance_loss_mlp": 1.01773608,
      "epoch": 0.8338844465820958,
      "flos": 27162185022720.0,
      "grad_norm": 1.9561354136624716,
      "language_loss": 0.83049148,
      "learning_rate": 2.824064777408117e-07,
      "loss": 0.85194921,
      "num_input_tokens_seen": 149856940,
      "step": 6935,
      "time_per_iteration": 2.742537260055542
    },
    {
      "auxiliary_loss_clip": 0.0115133,
      "auxiliary_loss_mlp": 0.01026984,
      "balance_loss_clip": 1.04536891,
      "balance_loss_mlp": 1.01985502,
      "epoch": 0.8340046894727349,
      "flos": 30481264425600.0,
      "grad_norm": 2.30564830791528,
      "language_loss": 0.75979996,
      "learning_rate": 2.8200752805731263e-07,
      "loss": 0.78158313,
      "num_input_tokens_seen": 149879930,
      "step": 6936,
      "time_per_iteration": 2.7043325901031494
    },
    {
      "auxiliary_loss_clip": 0.01151034,
      "auxiliary_loss_mlp": 0.01026543,
      "balance_loss_clip": 1.04521227,
      "balance_loss_mlp": 1.01992404,
      "epoch": 0.834124932363374,
      "flos": 27126166659840.0,
      "grad_norm": 1.4320006619683023,
      "language_loss": 0.81088042,
      "learning_rate": 2.8160883899190625e-07,
      "loss": 0.83265615,
      "num_input_tokens_seen": 149903200,
      "step": 6937,
      "time_per_iteration": 2.8850889205932617
    },
    {
      "auxiliary_loss_clip": 0.01115015,
      "auxiliary_loss_mlp": 0.01022758,
      "balance_loss_clip": 1.04166412,
      "balance_loss_mlp": 1.01537263,
      "epoch": 0.8342451752540131,
      "flos": 24569865498240.0,
      "grad_norm": 2.0764549710500733,
      "language_loss": 0.73394012,
      "learning_rate": 2.8121041060507234e-07,
      "loss": 0.75531781,
      "num_input_tokens_seen": 149922230,
      "step": 6938,
      "time_per_iteration": 2.752775192260742
    },
    {
      "auxiliary_loss_clip": 0.01155761,
      "auxiliary_loss_mlp": 0.01024121,
      "balance_loss_clip": 1.04410243,
      "balance_loss_mlp": 1.01669478,
      "epoch": 0.8343654181446521,
      "flos": 26615085995520.0,
      "grad_norm": 1.9083064553441433,
      "language_loss": 0.71075499,
      "learning_rate": 2.808122429572528e-07,
      "loss": 0.73255378,
      "num_input_tokens_seen": 149942435,
      "step": 6939,
      "time_per_iteration": 2.7183587551116943
    },
    {
      "auxiliary_loss_clip": 0.01130593,
      "auxiliary_loss_mlp": 0.01022488,
      "balance_loss_clip": 1.04147267,
      "balance_loss_mlp": 1.01545501,
      "epoch": 0.8344856610352913,
      "flos": 20777268078720.0,
      "grad_norm": 3.195150156041863,
      "language_loss": 0.76000774,
      "learning_rate": 2.804143361088489e-07,
      "loss": 0.78153849,
      "num_input_tokens_seen": 149961615,
      "step": 6940,
      "time_per_iteration": 2.7594761848449707
    },
    {
      "auxiliary_loss_clip": 0.01131744,
      "auxiliary_loss_mlp": 0.01024999,
      "balance_loss_clip": 1.04161334,
      "balance_loss_mlp": 1.01825225,
      "epoch": 0.8346059039259304,
      "flos": 26095960684800.0,
      "grad_norm": 2.289741030076065,
      "language_loss": 0.78059584,
      "learning_rate": 2.8001669012022277e-07,
      "loss": 0.80216324,
      "num_input_tokens_seen": 149979585,
      "step": 6941,
      "time_per_iteration": 2.683696746826172
    },
    {
      "auxiliary_loss_clip": 0.01153803,
      "auxiliary_loss_mlp": 0.01023888,
      "balance_loss_clip": 1.04694963,
      "balance_loss_mlp": 1.01647317,
      "epoch": 0.8347261468165694,
      "flos": 29027708755200.0,
      "grad_norm": 1.603878145062125,
      "language_loss": 0.69060361,
      "learning_rate": 2.7961930505169795e-07,
      "loss": 0.71238053,
      "num_input_tokens_seen": 150003830,
      "step": 6942,
      "time_per_iteration": 2.7634029388427734
    },
    {
      "auxiliary_loss_clip": 0.01156569,
      "auxiliary_loss_mlp": 0.00762421,
      "balance_loss_clip": 1.04647732,
      "balance_loss_mlp": 1.0003109,
      "epoch": 0.8348463897072086,
      "flos": 26396461866240.0,
      "grad_norm": 1.9879392064093055,
      "language_loss": 0.76688904,
      "learning_rate": 2.792221809635558e-07,
      "loss": 0.78607893,
      "num_input_tokens_seen": 150024460,
      "step": 6943,
      "time_per_iteration": 2.667833089828491
    },
    {
      "auxiliary_loss_clip": 0.01084255,
      "auxiliary_loss_mlp": 0.01027849,
      "balance_loss_clip": 1.03934729,
      "balance_loss_mlp": 1.02080345,
      "epoch": 0.8349666325978476,
      "flos": 23367720096000.0,
      "grad_norm": 2.1795786053919226,
      "language_loss": 0.75003266,
      "learning_rate": 2.788253179160411e-07,
      "loss": 0.77115369,
      "num_input_tokens_seen": 150045620,
      "step": 6944,
      "time_per_iteration": 2.8700051307678223
    },
    {
      "auxiliary_loss_clip": 0.01137276,
      "auxiliary_loss_mlp": 0.01021681,
      "balance_loss_clip": 1.04302907,
      "balance_loss_mlp": 1.0149132,
      "epoch": 0.8350868754884867,
      "flos": 12896528135040.0,
      "grad_norm": 2.1806518001903665,
      "language_loss": 0.64790952,
      "learning_rate": 2.7842871596935725e-07,
      "loss": 0.66949916,
      "num_input_tokens_seen": 150064135,
      "step": 6945,
      "time_per_iteration": 3.842698574066162
    },
    {
      "auxiliary_loss_clip": 0.01156784,
      "auxiliary_loss_mlp": 0.01025912,
      "balance_loss_clip": 1.04413915,
      "balance_loss_mlp": 1.01897717,
      "epoch": 0.8352071183791259,
      "flos": 26505522535680.0,
      "grad_norm": 1.6410961565916997,
      "language_loss": 0.69138777,
      "learning_rate": 2.780323751836682e-07,
      "loss": 0.71321476,
      "num_input_tokens_seen": 150085350,
      "step": 6946,
      "time_per_iteration": 2.6883273124694824
    },
    {
      "auxiliary_loss_clip": 0.01136112,
      "auxiliary_loss_mlp": 0.00762105,
      "balance_loss_clip": 1.0402813,
      "balance_loss_mlp": 1.00033724,
      "epoch": 0.8353273612697649,
      "flos": 20668063754880.0,
      "grad_norm": 1.6535325408017343,
      "language_loss": 0.78626525,
      "learning_rate": 2.7763629561909876e-07,
      "loss": 0.80524743,
      "num_input_tokens_seen": 150106180,
      "step": 6947,
      "time_per_iteration": 3.6024527549743652
    },
    {
      "auxiliary_loss_clip": 0.01162621,
      "auxiliary_loss_mlp": 0.01022826,
      "balance_loss_clip": 1.04405856,
      "balance_loss_mlp": 1.0161767,
      "epoch": 0.835447604160404,
      "flos": 19754137082880.0,
      "grad_norm": 2.729547521258697,
      "language_loss": 0.76896966,
      "learning_rate": 2.772404773357335e-07,
      "loss": 0.79082417,
      "num_input_tokens_seen": 150125585,
      "step": 6948,
      "time_per_iteration": 2.680018186569214
    },
    {
      "auxiliary_loss_clip": 0.0111533,
      "auxiliary_loss_mlp": 0.01024085,
      "balance_loss_clip": 1.04051232,
      "balance_loss_mlp": 1.01718545,
      "epoch": 0.8355678470510431,
      "flos": 23435842239360.0,
      "grad_norm": 2.8562464878327902,
      "language_loss": 0.78414255,
      "learning_rate": 2.7684492039361853e-07,
      "loss": 0.80553675,
      "num_input_tokens_seen": 150144810,
      "step": 6949,
      "time_per_iteration": 2.6965789794921875
    },
    {
      "auxiliary_loss_clip": 0.01168538,
      "auxiliary_loss_mlp": 0.01024925,
      "balance_loss_clip": 1.04911804,
      "balance_loss_mlp": 1.01766479,
      "epoch": 0.8356880899416822,
      "flos": 21214588164480.0,
      "grad_norm": 1.6610893400646594,
      "language_loss": 0.83647728,
      "learning_rate": 2.764496248527586e-07,
      "loss": 0.85841191,
      "num_input_tokens_seen": 150163785,
      "step": 6950,
      "time_per_iteration": 4.2908408641815186
    },
    {
      "auxiliary_loss_clip": 0.01132121,
      "auxiliary_loss_mlp": 0.01022721,
      "balance_loss_clip": 1.04104352,
      "balance_loss_mlp": 1.01598287,
      "epoch": 0.8358083328323213,
      "flos": 28037543466240.0,
      "grad_norm": 2.5824855894385106,
      "language_loss": 0.78644133,
      "learning_rate": 2.760545907731211e-07,
      "loss": 0.80798978,
      "num_input_tokens_seen": 150184360,
      "step": 6951,
      "time_per_iteration": 2.7243332862854004
    },
    {
      "auxiliary_loss_clip": 0.01151549,
      "auxiliary_loss_mlp": 0.01028331,
      "balance_loss_clip": 1.04236186,
      "balance_loss_mlp": 1.0214082,
      "epoch": 0.8359285757229604,
      "flos": 27783655159680.0,
      "grad_norm": 1.6997004968865645,
      "language_loss": 0.67868143,
      "learning_rate": 2.75659818214631e-07,
      "loss": 0.70048022,
      "num_input_tokens_seen": 150205465,
      "step": 6952,
      "time_per_iteration": 2.682063341140747
    },
    {
      "auxiliary_loss_clip": 0.01142542,
      "auxiliary_loss_mlp": 0.0102313,
      "balance_loss_clip": 1.04373074,
      "balance_loss_mlp": 1.01573312,
      "epoch": 0.8360488186135995,
      "flos": 21435115714560.0,
      "grad_norm": 1.7247182491966533,
      "language_loss": 0.78227949,
      "learning_rate": 2.752653072371749e-07,
      "loss": 0.80393624,
      "num_input_tokens_seen": 150224900,
      "step": 6953,
      "time_per_iteration": 2.656730890274048
    },
    {
      "auxiliary_loss_clip": 0.01117231,
      "auxiliary_loss_mlp": 0.01029805,
      "balance_loss_clip": 1.0417397,
      "balance_loss_mlp": 1.02322459,
      "epoch": 0.8361690615042385,
      "flos": 27632327160960.0,
      "grad_norm": 1.7212566432344993,
      "language_loss": 0.74942958,
      "learning_rate": 2.7487105790060105e-07,
      "loss": 0.77089989,
      "num_input_tokens_seen": 150244310,
      "step": 6954,
      "time_per_iteration": 2.7086880207061768
    },
    {
      "auxiliary_loss_clip": 0.01153625,
      "auxiliary_loss_mlp": 0.0102235,
      "balance_loss_clip": 1.04428899,
      "balance_loss_mlp": 1.01546609,
      "epoch": 0.8362893043948777,
      "flos": 39202529598720.0,
      "grad_norm": 1.8320703755418757,
      "language_loss": 0.69461942,
      "learning_rate": 2.7447707026471587e-07,
      "loss": 0.71637917,
      "num_input_tokens_seen": 150267285,
      "step": 6955,
      "time_per_iteration": 2.7473676204681396
    },
    {
      "auxiliary_loss_clip": 0.01122806,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.04002833,
      "balance_loss_mlp": 1.02145696,
      "epoch": 0.8364095472855168,
      "flos": 24785329230720.0,
      "grad_norm": 2.5912299531918768,
      "language_loss": 0.7976712,
      "learning_rate": 2.740833443892874e-07,
      "loss": 0.81918204,
      "num_input_tokens_seen": 150285455,
      "step": 6956,
      "time_per_iteration": 2.709413528442383
    },
    {
      "auxiliary_loss_clip": 0.01136542,
      "auxiliary_loss_mlp": 0.01024544,
      "balance_loss_clip": 1.04131198,
      "balance_loss_mlp": 1.01793671,
      "epoch": 0.8365297901761558,
      "flos": 22743412784640.0,
      "grad_norm": 1.7398432603468486,
      "language_loss": 0.79643595,
      "learning_rate": 2.7368988033404327e-07,
      "loss": 0.81804681,
      "num_input_tokens_seen": 150302970,
      "step": 6957,
      "time_per_iteration": 2.691215753555298
    },
    {
      "auxiliary_loss_clip": 0.0112513,
      "auxiliary_loss_mlp": 0.0102357,
      "balance_loss_clip": 1.04152334,
      "balance_loss_mlp": 1.01717424,
      "epoch": 0.836650033066795,
      "flos": 28396003242240.0,
      "grad_norm": 1.5622276238943975,
      "language_loss": 0.8457827,
      "learning_rate": 2.732966781586712e-07,
      "loss": 0.86726964,
      "num_input_tokens_seen": 150322715,
      "step": 6958,
      "time_per_iteration": 2.773498058319092
    },
    {
      "auxiliary_loss_clip": 0.01144043,
      "auxiliary_loss_mlp": 0.01025315,
      "balance_loss_clip": 1.04143167,
      "balance_loss_mlp": 1.01840949,
      "epoch": 0.836770275957434,
      "flos": 22236857233920.0,
      "grad_norm": 1.6616435272934111,
      "language_loss": 0.66622663,
      "learning_rate": 2.729037379228205e-07,
      "loss": 0.68792021,
      "num_input_tokens_seen": 150342900,
      "step": 6959,
      "time_per_iteration": 2.6601452827453613
    },
    {
      "auxiliary_loss_clip": 0.01140661,
      "auxiliary_loss_mlp": 0.01029079,
      "balance_loss_clip": 1.04701686,
      "balance_loss_mlp": 1.02225089,
      "epoch": 0.8368905188480731,
      "flos": 22491930689280.0,
      "grad_norm": 1.5298461478555523,
      "language_loss": 0.80395037,
      "learning_rate": 2.725110596860998e-07,
      "loss": 0.82564777,
      "num_input_tokens_seen": 150363580,
      "step": 6960,
      "time_per_iteration": 2.61474609375
    },
    {
      "auxiliary_loss_clip": 0.01107096,
      "auxiliary_loss_mlp": 0.01023896,
      "balance_loss_clip": 1.04175925,
      "balance_loss_mlp": 1.01687777,
      "epoch": 0.8370107617387123,
      "flos": 13370405287680.0,
      "grad_norm": 2.022454069160928,
      "language_loss": 0.70238638,
      "learning_rate": 2.7211864350807776e-07,
      "loss": 0.72369635,
      "num_input_tokens_seen": 150381780,
      "step": 6961,
      "time_per_iteration": 2.759600877761841
    },
    {
      "auxiliary_loss_clip": 0.01167546,
      "auxiliary_loss_mlp": 0.0102807,
      "balance_loss_clip": 1.04706633,
      "balance_loss_mlp": 1.02117944,
      "epoch": 0.8371310046293513,
      "flos": 25261289372160.0,
      "grad_norm": 1.6360609389194645,
      "language_loss": 0.7353406,
      "learning_rate": 2.717264894482836e-07,
      "loss": 0.7572968,
      "num_input_tokens_seen": 150402120,
      "step": 6962,
      "time_per_iteration": 2.682410717010498
    },
    {
      "auxiliary_loss_clip": 0.01154549,
      "auxiliary_loss_mlp": 0.0102506,
      "balance_loss_clip": 1.04635978,
      "balance_loss_mlp": 1.0176152,
      "epoch": 0.8372512475199904,
      "flos": 19792705311360.0,
      "grad_norm": 2.1787163153914637,
      "language_loss": 0.80855393,
      "learning_rate": 2.7133459756620646e-07,
      "loss": 0.83035004,
      "num_input_tokens_seen": 150419315,
      "step": 6963,
      "time_per_iteration": 2.649040460586548
    },
    {
      "auxiliary_loss_clip": 0.0114648,
      "auxiliary_loss_mlp": 0.01018782,
      "balance_loss_clip": 1.04381657,
      "balance_loss_mlp": 1.01194799,
      "epoch": 0.8373714904106295,
      "flos": 19391224020480.0,
      "grad_norm": 1.6240082562631224,
      "language_loss": 0.73497939,
      "learning_rate": 2.7094296792129733e-07,
      "loss": 0.75663197,
      "num_input_tokens_seen": 150438915,
      "step": 6964,
      "time_per_iteration": 2.6730337142944336
    },
    {
      "auxiliary_loss_clip": 0.01150866,
      "auxiliary_loss_mlp": 0.01022725,
      "balance_loss_clip": 1.043414,
      "balance_loss_mlp": 1.01542282,
      "epoch": 0.8374917333012686,
      "flos": 14975935401600.0,
      "grad_norm": 1.7661703526574635,
      "language_loss": 0.75496614,
      "learning_rate": 2.7055160057296424e-07,
      "loss": 0.77670205,
      "num_input_tokens_seen": 150456155,
      "step": 6965,
      "time_per_iteration": 2.6447768211364746
    },
    {
      "auxiliary_loss_clip": 0.01123953,
      "auxiliary_loss_mlp": 0.01026921,
      "balance_loss_clip": 1.04136562,
      "balance_loss_mlp": 1.01976275,
      "epoch": 0.8376119761919076,
      "flos": 30331839847680.0,
      "grad_norm": 1.6031496584173668,
      "language_loss": 0.72237158,
      "learning_rate": 2.7016049558057896e-07,
      "loss": 0.74388039,
      "num_input_tokens_seen": 150478115,
      "step": 6966,
      "time_per_iteration": 2.805903434753418
    },
    {
      "auxiliary_loss_clip": 0.01153846,
      "auxiliary_loss_mlp": 0.01023246,
      "balance_loss_clip": 1.04736912,
      "balance_loss_mlp": 1.01602149,
      "epoch": 0.8377322190825467,
      "flos": 29423336129280.0,
      "grad_norm": 1.6583781180368389,
      "language_loss": 0.70793593,
      "learning_rate": 2.6976965300347074e-07,
      "loss": 0.72970682,
      "num_input_tokens_seen": 150500725,
      "step": 6967,
      "time_per_iteration": 2.699808359146118
    },
    {
      "auxiliary_loss_clip": 0.01132811,
      "auxiliary_loss_mlp": 0.01026507,
      "balance_loss_clip": 1.04148948,
      "balance_loss_mlp": 1.01961422,
      "epoch": 0.8378524619731859,
      "flos": 26687086807680.0,
      "grad_norm": 2.9504190943416244,
      "language_loss": 0.6922363,
      "learning_rate": 2.693790729009309e-07,
      "loss": 0.71382952,
      "num_input_tokens_seen": 150522335,
      "step": 6968,
      "time_per_iteration": 2.7191340923309326
    },
    {
      "auxiliary_loss_clip": 0.01137376,
      "auxiliary_loss_mlp": 0.01021762,
      "balance_loss_clip": 1.04343688,
      "balance_loss_mlp": 1.01468658,
      "epoch": 0.8379727048638249,
      "flos": 20703866636160.0,
      "grad_norm": 1.7418338712959534,
      "language_loss": 0.88372296,
      "learning_rate": 2.6898875533220946e-07,
      "loss": 0.90531433,
      "num_input_tokens_seen": 150541640,
      "step": 6969,
      "time_per_iteration": 2.7307398319244385
    },
    {
      "auxiliary_loss_clip": 0.01161439,
      "auxiliary_loss_mlp": 0.01019904,
      "balance_loss_clip": 1.04563558,
      "balance_loss_mlp": 1.01363683,
      "epoch": 0.838092947754464,
      "flos": 20084084438400.0,
      "grad_norm": 2.0010800646827835,
      "language_loss": 0.81853807,
      "learning_rate": 2.685987003565171e-07,
      "loss": 0.84035152,
      "num_input_tokens_seen": 150559680,
      "step": 6970,
      "time_per_iteration": 2.6180191040039062
    },
    {
      "auxiliary_loss_clip": 0.01113633,
      "auxiliary_loss_mlp": 0.01022113,
      "balance_loss_clip": 1.04127896,
      "balance_loss_mlp": 1.01500487,
      "epoch": 0.8382131906451031,
      "flos": 18113270964480.0,
      "grad_norm": 2.6728955739784306,
      "language_loss": 0.75138009,
      "learning_rate": 2.6820890803302566e-07,
      "loss": 0.77273756,
      "num_input_tokens_seen": 150575205,
      "step": 6971,
      "time_per_iteration": 3.5802676677703857
    },
    {
      "auxiliary_loss_clip": 0.01136773,
      "auxiliary_loss_mlp": 0.01025168,
      "balance_loss_clip": 1.04504228,
      "balance_loss_mlp": 1.01831353,
      "epoch": 0.8383334335357422,
      "flos": 17092653920640.0,
      "grad_norm": 2.0663337904546912,
      "language_loss": 0.81727618,
      "learning_rate": 2.6781937842086557e-07,
      "loss": 0.83889556,
      "num_input_tokens_seen": 150593995,
      "step": 6972,
      "time_per_iteration": 2.688053607940674
    },
    {
      "auxiliary_loss_clip": 0.01153575,
      "auxiliary_loss_mlp": 0.01023938,
      "balance_loss_clip": 1.04478717,
      "balance_loss_mlp": 1.01690483,
      "epoch": 0.8384536764263812,
      "flos": 20704728562560.0,
      "grad_norm": 2.071723705513684,
      "language_loss": 0.67204165,
      "learning_rate": 2.6743011157912933e-07,
      "loss": 0.69381678,
      "num_input_tokens_seen": 150613715,
      "step": 6973,
      "time_per_iteration": 3.601346015930176
    },
    {
      "auxiliary_loss_clip": 0.0110494,
      "auxiliary_loss_mlp": 0.01021795,
      "balance_loss_clip": 1.0343436,
      "balance_loss_mlp": 1.01432633,
      "epoch": 0.8385739193170204,
      "flos": 28986842056320.0,
      "grad_norm": 2.2157827594547,
      "language_loss": 0.65132111,
      "learning_rate": 2.6704110756686725e-07,
      "loss": 0.67258847,
      "num_input_tokens_seen": 150634540,
      "step": 6974,
      "time_per_iteration": 2.7911980152130127
    },
    {
      "auxiliary_loss_clip": 0.01134653,
      "auxiliary_loss_mlp": 0.00762102,
      "balance_loss_clip": 1.04156375,
      "balance_loss_mlp": 1.00028193,
      "epoch": 0.8386941622076595,
      "flos": 23438068882560.0,
      "grad_norm": 1.84554909251744,
      "language_loss": 0.83899009,
      "learning_rate": 2.6665236644309085e-07,
      "loss": 0.85795766,
      "num_input_tokens_seen": 150654850,
      "step": 6975,
      "time_per_iteration": 2.7316012382507324
    },
    {
      "auxiliary_loss_clip": 0.01152296,
      "auxiliary_loss_mlp": 0.01025471,
      "balance_loss_clip": 1.04457295,
      "balance_loss_mlp": 1.01907289,
      "epoch": 0.8388144050982985,
      "flos": 23002724044800.0,
      "grad_norm": 1.838204913583312,
      "language_loss": 0.79792708,
      "learning_rate": 2.662638882667727e-07,
      "loss": 0.81970477,
      "num_input_tokens_seen": 150673790,
      "step": 6976,
      "time_per_iteration": 4.411527395248413
    },
    {
      "auxiliary_loss_clip": 0.01167,
      "auxiliary_loss_mlp": 0.0102742,
      "balance_loss_clip": 1.04473424,
      "balance_loss_mlp": 1.02035403,
      "epoch": 0.8389346479889377,
      "flos": 24280353878400.0,
      "grad_norm": 2.0631165611587643,
      "language_loss": 0.72908771,
      "learning_rate": 2.658756730968443e-07,
      "loss": 0.75103199,
      "num_input_tokens_seen": 150692255,
      "step": 6977,
      "time_per_iteration": 2.6788253784179688
    },
    {
      "auxiliary_loss_clip": 0.01141242,
      "auxiliary_loss_mlp": 0.01022056,
      "balance_loss_clip": 1.04516315,
      "balance_loss_mlp": 1.01516569,
      "epoch": 0.8390548908795767,
      "flos": 21215019127680.0,
      "grad_norm": 1.945889137960177,
      "language_loss": 0.88383639,
      "learning_rate": 2.654877209921975e-07,
      "loss": 0.90546942,
      "num_input_tokens_seen": 150709790,
      "step": 6978,
      "time_per_iteration": 2.7129976749420166
    },
    {
      "auxiliary_loss_clip": 0.01115461,
      "auxiliary_loss_mlp": 0.01030892,
      "balance_loss_clip": 1.03854418,
      "balance_loss_mlp": 1.02288079,
      "epoch": 0.8391751337702158,
      "flos": 35627299332480.0,
      "grad_norm": 2.293842820281841,
      "language_loss": 0.62911761,
      "learning_rate": 2.651000320116843e-07,
      "loss": 0.65058112,
      "num_input_tokens_seen": 150730675,
      "step": 6979,
      "time_per_iteration": 2.831265926361084
    },
    {
      "auxiliary_loss_clip": 0.0111999,
      "auxiliary_loss_mlp": 0.00763164,
      "balance_loss_clip": 1.04119182,
      "balance_loss_mlp": 1.00031948,
      "epoch": 0.839295376660855,
      "flos": 21325229032320.0,
      "grad_norm": 2.211234443006644,
      "language_loss": 0.76118541,
      "learning_rate": 2.647126062141163e-07,
      "loss": 0.78001696,
      "num_input_tokens_seen": 150749750,
      "step": 6980,
      "time_per_iteration": 2.7487103939056396
    },
    {
      "auxiliary_loss_clip": 0.01140194,
      "auxiliary_loss_mlp": 0.01024791,
      "balance_loss_clip": 1.04015064,
      "balance_loss_mlp": 1.0178858,
      "epoch": 0.839415619551494,
      "flos": 18442535961600.0,
      "grad_norm": 1.9620609120349832,
      "language_loss": 0.83808976,
      "learning_rate": 2.643254436582669e-07,
      "loss": 0.85973966,
      "num_input_tokens_seen": 150769240,
      "step": 6981,
      "time_per_iteration": 2.7452313899993896
    },
    {
      "auxiliary_loss_clip": 0.01113608,
      "auxiliary_loss_mlp": 0.0102325,
      "balance_loss_clip": 1.04101145,
      "balance_loss_mlp": 1.01576352,
      "epoch": 0.8395358624421331,
      "flos": 23221958705280.0,
      "grad_norm": 1.738223522358968,
      "language_loss": 0.82014567,
      "learning_rate": 2.6393854440286743e-07,
      "loss": 0.84151423,
      "num_input_tokens_seen": 150788410,
      "step": 6982,
      "time_per_iteration": 2.749258279800415
    },
    {
      "auxiliary_loss_clip": 0.01167569,
      "auxiliary_loss_mlp": 0.0103245,
      "balance_loss_clip": 1.04897523,
      "balance_loss_mlp": 1.02578616,
      "epoch": 0.8396561053327722,
      "flos": 24381657210240.0,
      "grad_norm": 2.0963858468313434,
      "language_loss": 0.70762157,
      "learning_rate": 2.6355190850661045e-07,
      "loss": 0.72962177,
      "num_input_tokens_seen": 150805245,
      "step": 6983,
      "time_per_iteration": 2.5884456634521484
    },
    {
      "auxiliary_loss_clip": 0.01138932,
      "auxiliary_loss_mlp": 0.01022086,
      "balance_loss_clip": 1.04584336,
      "balance_loss_mlp": 1.01494551,
      "epoch": 0.8397763482234113,
      "flos": 22237755073920.0,
      "grad_norm": 1.7944106453538728,
      "language_loss": 0.86556828,
      "learning_rate": 2.631655360281486e-07,
      "loss": 0.88717842,
      "num_input_tokens_seen": 150824920,
      "step": 6984,
      "time_per_iteration": 2.659560203552246
    },
    {
      "auxiliary_loss_clip": 0.0115766,
      "auxiliary_loss_mlp": 0.00762414,
      "balance_loss_clip": 1.04434347,
      "balance_loss_mlp": 1.00032902,
      "epoch": 0.8398965911140504,
      "flos": 22163743100160.0,
      "grad_norm": 2.04143474504253,
      "language_loss": 0.65331852,
      "learning_rate": 2.6277942702609323e-07,
      "loss": 0.67251927,
      "num_input_tokens_seen": 150844400,
      "step": 6985,
      "time_per_iteration": 2.5993106365203857
    },
    {
      "auxiliary_loss_clip": 0.01125262,
      "auxiliary_loss_mlp": 0.0102337,
      "balance_loss_clip": 1.04267311,
      "balance_loss_mlp": 1.01626539,
      "epoch": 0.8400168340046895,
      "flos": 21542775753600.0,
      "grad_norm": 2.2294006909275677,
      "language_loss": 0.87365502,
      "learning_rate": 2.623935815590186e-07,
      "loss": 0.8951413,
      "num_input_tokens_seen": 150862780,
      "step": 6986,
      "time_per_iteration": 2.715299606323242
    },
    {
      "auxiliary_loss_clip": 0.0113987,
      "auxiliary_loss_mlp": 0.010237,
      "balance_loss_clip": 1.04602766,
      "balance_loss_mlp": 1.01657724,
      "epoch": 0.8401370768953286,
      "flos": 22491966602880.0,
      "grad_norm": 5.49628294358022,
      "language_loss": 0.80774987,
      "learning_rate": 2.6200799968545516e-07,
      "loss": 0.82938552,
      "num_input_tokens_seen": 150883075,
      "step": 6987,
      "time_per_iteration": 2.657796859741211
    },
    {
      "auxiliary_loss_clip": 0.0104398,
      "auxiliary_loss_mlp": 0.01001793,
      "balance_loss_clip": 1.01385546,
      "balance_loss_mlp": 1.00069666,
      "epoch": 0.8402573197859676,
      "flos": 59238890818560.0,
      "grad_norm": 1.2369501984306077,
      "language_loss": 0.56422448,
      "learning_rate": 2.616226814638969e-07,
      "loss": 0.58468223,
      "num_input_tokens_seen": 150948180,
      "step": 6988,
      "time_per_iteration": 3.2862749099731445
    },
    {
      "auxiliary_loss_clip": 0.01141679,
      "auxiliary_loss_mlp": 0.01021775,
      "balance_loss_clip": 1.04574263,
      "balance_loss_mlp": 1.01483691,
      "epoch": 0.8403775626766068,
      "flos": 22674608282880.0,
      "grad_norm": 1.944297281431451,
      "language_loss": 0.77621794,
      "learning_rate": 2.612376269527954e-07,
      "loss": 0.79785246,
      "num_input_tokens_seen": 150967885,
      "step": 6989,
      "time_per_iteration": 2.6740596294403076
    },
    {
      "auxiliary_loss_clip": 0.01136484,
      "auxiliary_loss_mlp": 0.01029104,
      "balance_loss_clip": 1.04451239,
      "balance_loss_mlp": 1.02229404,
      "epoch": 0.8404978055672458,
      "flos": 19609704495360.0,
      "grad_norm": 1.8160304535090925,
      "language_loss": 0.67219532,
      "learning_rate": 2.608528362105635e-07,
      "loss": 0.69385117,
      "num_input_tokens_seen": 150987255,
      "step": 6990,
      "time_per_iteration": 2.645740270614624
    },
    {
      "auxiliary_loss_clip": 0.01125828,
      "auxiliary_loss_mlp": 0.010257,
      "balance_loss_clip": 1.04073668,
      "balance_loss_mlp": 1.0183568,
      "epoch": 0.8406180484578849,
      "flos": 27526929678720.0,
      "grad_norm": 1.8435575235186863,
      "language_loss": 0.73242462,
      "learning_rate": 2.6046830929557374e-07,
      "loss": 0.75393993,
      "num_input_tokens_seen": 151006905,
      "step": 6991,
      "time_per_iteration": 2.8330512046813965
    },
    {
      "auxiliary_loss_clip": 0.01119346,
      "auxiliary_loss_mlp": 0.0103008,
      "balance_loss_clip": 1.04183793,
      "balance_loss_mlp": 1.0229187,
      "epoch": 0.8407382913485241,
      "flos": 22127473342080.0,
      "grad_norm": 1.7899578276103818,
      "language_loss": 0.84913933,
      "learning_rate": 2.6008404626615776e-07,
      "loss": 0.8706336,
      "num_input_tokens_seen": 151025405,
      "step": 6992,
      "time_per_iteration": 2.797581911087036
    },
    {
      "auxiliary_loss_clip": 0.01156149,
      "auxiliary_loss_mlp": 0.01028373,
      "balance_loss_clip": 1.04540968,
      "balance_loss_mlp": 1.02114296,
      "epoch": 0.8408585342391631,
      "flos": 13918473982080.0,
      "grad_norm": 6.064781602813566,
      "language_loss": 0.7370891,
      "learning_rate": 2.597000471806092e-07,
      "loss": 0.75893432,
      "num_input_tokens_seen": 151041970,
      "step": 6993,
      "time_per_iteration": 2.633920431137085
    },
    {
      "auxiliary_loss_clip": 0.01135768,
      "auxiliary_loss_mlp": 0.01028484,
      "balance_loss_clip": 1.0458045,
      "balance_loss_mlp": 1.02072895,
      "epoch": 0.8409787771298022,
      "flos": 20187865808640.0,
      "grad_norm": 1.9524396115806804,
      "language_loss": 0.73114771,
      "learning_rate": 2.593163120971793e-07,
      "loss": 0.75279021,
      "num_input_tokens_seen": 151060835,
      "step": 6994,
      "time_per_iteration": 2.6739163398742676
    },
    {
      "auxiliary_loss_clip": 0.01100822,
      "auxiliary_loss_mlp": 0.01023529,
      "balance_loss_clip": 1.03693962,
      "balance_loss_mlp": 1.01669538,
      "epoch": 0.8410990200204413,
      "flos": 23142523777920.0,
      "grad_norm": 1.8553947982649048,
      "language_loss": 0.69391561,
      "learning_rate": 2.5893284107408165e-07,
      "loss": 0.71515912,
      "num_input_tokens_seen": 151078205,
      "step": 6995,
      "time_per_iteration": 2.762742519378662
    },
    {
      "auxiliary_loss_clip": 0.01108347,
      "auxiliary_loss_mlp": 0.01024856,
      "balance_loss_clip": 1.04152846,
      "balance_loss_mlp": 1.01787341,
      "epoch": 0.8412192629110804,
      "flos": 24027219757440.0,
      "grad_norm": 1.7060053288870858,
      "language_loss": 0.77827573,
      "learning_rate": 2.5854963416948726e-07,
      "loss": 0.79960769,
      "num_input_tokens_seen": 151100470,
      "step": 6996,
      "time_per_iteration": 2.802072763442993
    },
    {
      "auxiliary_loss_clip": 0.01108827,
      "auxiliary_loss_mlp": 0.01025264,
      "balance_loss_clip": 1.03562391,
      "balance_loss_mlp": 1.01812911,
      "epoch": 0.8413395058017195,
      "flos": 25591703604480.0,
      "grad_norm": 1.6374101168180368,
      "language_loss": 0.69195354,
      "learning_rate": 2.5816669144152816e-07,
      "loss": 0.71329439,
      "num_input_tokens_seen": 151121650,
      "step": 6997,
      "time_per_iteration": 3.707207441329956
    },
    {
      "auxiliary_loss_clip": 0.0106359,
      "auxiliary_loss_mlp": 0.0100144,
      "balance_loss_clip": 1.00962865,
      "balance_loss_mlp": 1.00037336,
      "epoch": 0.8414597486923585,
      "flos": 63635396624640.0,
      "grad_norm": 0.8472904490178057,
      "language_loss": 0.6626665,
      "learning_rate": 2.5778401294829777e-07,
      "loss": 0.68331683,
      "num_input_tokens_seen": 151180390,
      "step": 6998,
      "time_per_iteration": 3.2839386463165283
    },
    {
      "auxiliary_loss_clip": 0.0114934,
      "auxiliary_loss_mlp": 0.00762232,
      "balance_loss_clip": 1.04424119,
      "balance_loss_mlp": 1.00030518,
      "epoch": 0.8415799915829977,
      "flos": 19098731571840.0,
      "grad_norm": 1.739850638902741,
      "language_loss": 0.65066063,
      "learning_rate": 2.574015987478473e-07,
      "loss": 0.66977632,
      "num_input_tokens_seen": 151198520,
      "step": 6999,
      "time_per_iteration": 3.7726216316223145
    },
    {
      "auxiliary_loss_clip": 0.01144831,
      "auxiliary_loss_mlp": 0.01025158,
      "balance_loss_clip": 1.0449903,
      "balance_loss_mlp": 1.01810944,
      "epoch": 0.8417002344736367,
      "flos": 19821612781440.0,
      "grad_norm": 2.208586185950456,
      "language_loss": 0.87014008,
      "learning_rate": 2.570194488981887e-07,
      "loss": 0.89183998,
      "num_input_tokens_seen": 151215065,
      "step": 7000,
      "time_per_iteration": 2.639275312423706
    },
    {
      "auxiliary_loss_clip": 0.01063595,
      "auxiliary_loss_mlp": 0.01000695,
      "balance_loss_clip": 1.00962543,
      "balance_loss_mlp": 0.99964601,
      "epoch": 0.8418204773642758,
      "flos": 62161516834560.0,
      "grad_norm": 0.8367606008631114,
      "language_loss": 0.60348427,
      "learning_rate": 2.566375634572939e-07,
      "loss": 0.62412721,
      "num_input_tokens_seen": 151275705,
      "step": 7001,
      "time_per_iteration": 3.1223394870758057
    },
    {
      "auxiliary_loss_clip": 0.0113068,
      "auxiliary_loss_mlp": 0.01021491,
      "balance_loss_clip": 1.04214501,
      "balance_loss_mlp": 1.01444578,
      "epoch": 0.841940720254915,
      "flos": 17092905315840.0,
      "grad_norm": 2.159089342480216,
      "language_loss": 0.76371706,
      "learning_rate": 2.562559424830943e-07,
      "loss": 0.78523874,
      "num_input_tokens_seen": 151293665,
      "step": 7002,
      "time_per_iteration": 4.6003031730651855
    },
    {
      "auxiliary_loss_clip": 0.01134261,
      "auxiliary_loss_mlp": 0.01026676,
      "balance_loss_clip": 1.04182374,
      "balance_loss_mlp": 1.01929712,
      "epoch": 0.842060963145554,
      "flos": 16283586026880.0,
      "grad_norm": 2.4084769928116248,
      "language_loss": 0.7027722,
      "learning_rate": 2.5587458603348256e-07,
      "loss": 0.72438157,
      "num_input_tokens_seen": 151310955,
      "step": 7003,
      "time_per_iteration": 2.723414421081543
    },
    {
      "auxiliary_loss_clip": 0.01118105,
      "auxiliary_loss_mlp": 0.01025449,
      "balance_loss_clip": 1.03960752,
      "balance_loss_mlp": 1.01812959,
      "epoch": 0.8421812060361931,
      "flos": 21908238681600.0,
      "grad_norm": 1.8636974819042524,
      "language_loss": 0.84031522,
      "learning_rate": 2.554934941663085e-07,
      "loss": 0.86175072,
      "num_input_tokens_seen": 151328490,
      "step": 7004,
      "time_per_iteration": 2.7512331008911133
    },
    {
      "auxiliary_loss_clip": 0.01123125,
      "auxiliary_loss_mlp": 0.0102224,
      "balance_loss_clip": 1.0409168,
      "balance_loss_mlp": 1.01462245,
      "epoch": 0.8423014489268322,
      "flos": 27777693502080.0,
      "grad_norm": 2.0728739590679375,
      "language_loss": 0.73428416,
      "learning_rate": 2.5511266693938484e-07,
      "loss": 0.75573778,
      "num_input_tokens_seen": 151346950,
      "step": 7005,
      "time_per_iteration": 2.7689740657806396
    },
    {
      "auxiliary_loss_clip": 0.01137224,
      "auxiliary_loss_mlp": 0.0102759,
      "balance_loss_clip": 1.0451864,
      "balance_loss_mlp": 1.01981497,
      "epoch": 0.8424216918174713,
      "flos": 25117610970240.0,
      "grad_norm": 1.5590548070233865,
      "language_loss": 0.77958858,
      "learning_rate": 2.547321044104822e-07,
      "loss": 0.80123675,
      "num_input_tokens_seen": 151368445,
      "step": 7006,
      "time_per_iteration": 2.7984139919281006
    },
    {
      "auxiliary_loss_clip": 0.01172034,
      "auxiliary_loss_mlp": 0.0102309,
      "balance_loss_clip": 1.04991269,
      "balance_loss_mlp": 1.01585948,
      "epoch": 0.8425419347081103,
      "flos": 24748448941440.0,
      "grad_norm": 1.8094242763878217,
      "language_loss": 0.7691561,
      "learning_rate": 2.5435180663733113e-07,
      "loss": 0.7911073,
      "num_input_tokens_seen": 151388745,
      "step": 7007,
      "time_per_iteration": 2.710113763809204
    },
    {
      "auxiliary_loss_clip": 0.01120089,
      "auxiliary_loss_mlp": 0.01023057,
      "balance_loss_clip": 1.04015744,
      "balance_loss_mlp": 1.0158751,
      "epoch": 0.8426621775987495,
      "flos": 24820916630400.0,
      "grad_norm": 2.336410961902473,
      "language_loss": 0.71521246,
      "learning_rate": 2.539717736776241e-07,
      "loss": 0.73664391,
      "num_input_tokens_seen": 151404970,
      "step": 7008,
      "time_per_iteration": 2.7850589752197266
    },
    {
      "auxiliary_loss_clip": 0.01148829,
      "auxiliary_loss_mlp": 0.01020958,
      "balance_loss_clip": 1.04475522,
      "balance_loss_mlp": 1.01420176,
      "epoch": 0.8427824204893886,
      "flos": 23550074467200.0,
      "grad_norm": 1.529002743032788,
      "language_loss": 0.76523453,
      "learning_rate": 2.535920055890097e-07,
      "loss": 0.78693247,
      "num_input_tokens_seen": 151426265,
      "step": 7009,
      "time_per_iteration": 2.6552512645721436
    },
    {
      "auxiliary_loss_clip": 0.01105516,
      "auxiliary_loss_mlp": 0.01027151,
      "balance_loss_clip": 1.03949189,
      "balance_loss_mlp": 1.01965904,
      "epoch": 0.8429026633800276,
      "flos": 16143858120960.0,
      "grad_norm": 8.999618408915026,
      "language_loss": 0.64795172,
      "learning_rate": 2.5321250242910006e-07,
      "loss": 0.66927838,
      "num_input_tokens_seen": 151444180,
      "step": 7010,
      "time_per_iteration": 2.7009873390197754
    },
    {
      "auxiliary_loss_clip": 0.01166204,
      "auxiliary_loss_mlp": 0.01024535,
      "balance_loss_clip": 1.04684448,
      "balance_loss_mlp": 1.01788008,
      "epoch": 0.8430229062706668,
      "flos": 22198540400640.0,
      "grad_norm": 1.74431619907716,
      "language_loss": 0.86437142,
      "learning_rate": 2.5283326425546493e-07,
      "loss": 0.88627881,
      "num_input_tokens_seen": 151463290,
      "step": 7011,
      "time_per_iteration": 2.6431782245635986
    },
    {
      "auxiliary_loss_clip": 0.01116967,
      "auxiliary_loss_mlp": 0.01019903,
      "balance_loss_clip": 1.04379439,
      "balance_loss_mlp": 1.01368964,
      "epoch": 0.8431431491613058,
      "flos": 35330317683840.0,
      "grad_norm": 1.9529512025556373,
      "language_loss": 0.69271219,
      "learning_rate": 2.5245429112563443e-07,
      "loss": 0.71408093,
      "num_input_tokens_seen": 151483965,
      "step": 7012,
      "time_per_iteration": 2.8054752349853516
    },
    {
      "auxiliary_loss_clip": 0.01151978,
      "auxiliary_loss_mlp": 0.01026271,
      "balance_loss_clip": 1.04676676,
      "balance_loss_mlp": 1.01891005,
      "epoch": 0.8432633920519449,
      "flos": 25812374808960.0,
      "grad_norm": 2.0137130052381216,
      "language_loss": 0.82254303,
      "learning_rate": 2.5207558309709865e-07,
      "loss": 0.84432554,
      "num_input_tokens_seen": 151503700,
      "step": 7013,
      "time_per_iteration": 2.693586587905884
    },
    {
      "auxiliary_loss_clip": 0.01037977,
      "auxiliary_loss_mlp": 0.00753656,
      "balance_loss_clip": 1.00973487,
      "balance_loss_mlp": 1.0001626,
      "epoch": 0.8433836349425841,
      "flos": 64959531592320.0,
      "grad_norm": 0.6570822625597135,
      "language_loss": 0.56242132,
      "learning_rate": 2.516971402273065e-07,
      "loss": 0.58033764,
      "num_input_tokens_seen": 151569765,
      "step": 7014,
      "time_per_iteration": 3.2729172706604004
    },
    {
      "auxiliary_loss_clip": 0.01134999,
      "auxiliary_loss_mlp": 0.01027392,
      "balance_loss_clip": 1.04065633,
      "balance_loss_mlp": 1.02075469,
      "epoch": 0.8435038778332231,
      "flos": 20229989483520.0,
      "grad_norm": 3.146310841218998,
      "language_loss": 0.67463106,
      "learning_rate": 2.513189625736687e-07,
      "loss": 0.69625497,
      "num_input_tokens_seen": 151586660,
      "step": 7015,
      "time_per_iteration": 2.687856435775757
    },
    {
      "auxiliary_loss_clip": 0.01129493,
      "auxiliary_loss_mlp": 0.0102521,
      "balance_loss_clip": 1.04158998,
      "balance_loss_mlp": 1.01777434,
      "epoch": 0.8436241207238622,
      "flos": 20992229020800.0,
      "grad_norm": 2.2554639897278412,
      "language_loss": 0.71552044,
      "learning_rate": 2.509410501935534e-07,
      "loss": 0.7370674,
      "num_input_tokens_seen": 151602295,
      "step": 7016,
      "time_per_iteration": 2.6681063175201416
    },
    {
      "auxiliary_loss_clip": 0.01141635,
      "auxiliary_loss_mlp": 0.01025483,
      "balance_loss_clip": 1.04518795,
      "balance_loss_mlp": 1.0181458,
      "epoch": 0.8437443636145013,
      "flos": 14682257804160.0,
      "grad_norm": 2.448269051211799,
      "language_loss": 0.75735784,
      "learning_rate": 2.5056340314429116e-07,
      "loss": 0.77902901,
      "num_input_tokens_seen": 151619760,
      "step": 7017,
      "time_per_iteration": 2.631019115447998
    },
    {
      "auxiliary_loss_clip": 0.01110437,
      "auxiliary_loss_mlp": 0.01022451,
      "balance_loss_clip": 1.03904843,
      "balance_loss_mlp": 1.01502419,
      "epoch": 0.8438646065051404,
      "flos": 21608814908160.0,
      "grad_norm": 2.1716708436909102,
      "language_loss": 0.80142826,
      "learning_rate": 2.5018602148316904e-07,
      "loss": 0.82275712,
      "num_input_tokens_seen": 151635795,
      "step": 7018,
      "time_per_iteration": 2.677196741104126
    },
    {
      "auxiliary_loss_clip": 0.01165058,
      "auxiliary_loss_mlp": 0.01019599,
      "balance_loss_clip": 1.04740286,
      "balance_loss_mlp": 1.01312041,
      "epoch": 0.8439848493957794,
      "flos": 23289937194240.0,
      "grad_norm": 1.7484482050868222,
      "language_loss": 0.80210567,
      "learning_rate": 2.498089052674359e-07,
      "loss": 0.82395226,
      "num_input_tokens_seen": 151653770,
      "step": 7019,
      "time_per_iteration": 2.6859614849090576
    },
    {
      "auxiliary_loss_clip": 0.01153381,
      "auxiliary_loss_mlp": 0.01028089,
      "balance_loss_clip": 1.04529476,
      "balance_loss_mlp": 1.02100241,
      "epoch": 0.8441050922864186,
      "flos": 19719339782400.0,
      "grad_norm": 1.9056904206983771,
      "language_loss": 0.75338936,
      "learning_rate": 2.494320545543007e-07,
      "loss": 0.77520406,
      "num_input_tokens_seen": 151673340,
      "step": 7020,
      "time_per_iteration": 2.673088312149048
    },
    {
      "auxiliary_loss_clip": 0.011689,
      "auxiliary_loss_mlp": 0.0102658,
      "balance_loss_clip": 1.04656053,
      "balance_loss_mlp": 1.01940954,
      "epoch": 0.8442253351770577,
      "flos": 21835268202240.0,
      "grad_norm": 1.7180170385952607,
      "language_loss": 0.66982859,
      "learning_rate": 2.490554694009308e-07,
      "loss": 0.69178343,
      "num_input_tokens_seen": 151694205,
      "step": 7021,
      "time_per_iteration": 2.724266529083252
    },
    {
      "auxiliary_loss_clip": 0.01154934,
      "auxiliary_loss_mlp": 0.01020967,
      "balance_loss_clip": 1.04336262,
      "balance_loss_mlp": 1.014431,
      "epoch": 0.8443455780676967,
      "flos": 34346365447680.0,
      "grad_norm": 1.5585749575031145,
      "language_loss": 0.78261769,
      "learning_rate": 2.4867914986445426e-07,
      "loss": 0.8043766,
      "num_input_tokens_seen": 151716595,
      "step": 7022,
      "time_per_iteration": 3.7058355808258057
    },
    {
      "auxiliary_loss_clip": 0.01141334,
      "auxiliary_loss_mlp": 0.01024842,
      "balance_loss_clip": 1.04170442,
      "balance_loss_mlp": 1.01817775,
      "epoch": 0.8444658209583359,
      "flos": 48214599281280.0,
      "grad_norm": 2.4291379961990964,
      "language_loss": 0.71314448,
      "learning_rate": 2.483030960019581e-07,
      "loss": 0.73480624,
      "num_input_tokens_seen": 151740525,
      "step": 7023,
      "time_per_iteration": 2.966651439666748
    },
    {
      "auxiliary_loss_clip": 0.01019181,
      "auxiliary_loss_mlp": 0.01000878,
      "balance_loss_clip": 1.00808215,
      "balance_loss_mlp": 0.99979371,
      "epoch": 0.8445860638489749,
      "flos": 68484773105280.0,
      "grad_norm": 0.7338019974225316,
      "language_loss": 0.55466616,
      "learning_rate": 2.479273078704891e-07,
      "loss": 0.57486671,
      "num_input_tokens_seen": 151793890,
      "step": 7024,
      "time_per_iteration": 3.2191364765167236
    },
    {
      "auxiliary_loss_clip": 0.01018302,
      "auxiliary_loss_mlp": 0.01000342,
      "balance_loss_clip": 1.01334238,
      "balance_loss_mlp": 0.99920958,
      "epoch": 0.844706306739614,
      "flos": 62833331882880.0,
      "grad_norm": 0.7741569441432007,
      "language_loss": 0.64670718,
      "learning_rate": 2.475517855270552e-07,
      "loss": 0.66689372,
      "num_input_tokens_seen": 151853970,
      "step": 7025,
      "time_per_iteration": 4.198790073394775
    },
    {
      "auxiliary_loss_clip": 0.01163936,
      "auxiliary_loss_mlp": 0.01028206,
      "balance_loss_clip": 1.04598677,
      "balance_loss_mlp": 1.02153897,
      "epoch": 0.8448265496302532,
      "flos": 14976114969600.0,
      "grad_norm": 2.0759049913267007,
      "language_loss": 0.7254976,
      "learning_rate": 2.4717652902862143e-07,
      "loss": 0.747419,
      "num_input_tokens_seen": 151872945,
      "step": 7026,
      "time_per_iteration": 2.5440597534179688
    },
    {
      "auxiliary_loss_clip": 0.01140994,
      "auxiliary_loss_mlp": 0.01027947,
      "balance_loss_clip": 1.0416491,
      "balance_loss_mlp": 1.02110124,
      "epoch": 0.8449467925208922,
      "flos": 23441265192960.0,
      "grad_norm": 1.7079991438787616,
      "language_loss": 0.81183183,
      "learning_rate": 2.4680153843211495e-07,
      "loss": 0.83352125,
      "num_input_tokens_seen": 151892875,
      "step": 7027,
      "time_per_iteration": 2.6581881046295166
    },
    {
      "auxiliary_loss_clip": 0.01137541,
      "auxiliary_loss_mlp": 0.0102712,
      "balance_loss_clip": 1.04580116,
      "balance_loss_mlp": 1.0192821,
      "epoch": 0.8450670354115313,
      "flos": 22748045639040.0,
      "grad_norm": 2.096913556250651,
      "language_loss": 0.72426206,
      "learning_rate": 2.464268137944212e-07,
      "loss": 0.74590874,
      "num_input_tokens_seen": 151914170,
      "step": 7028,
      "time_per_iteration": 3.611950635910034
    },
    {
      "auxiliary_loss_clip": 0.01097353,
      "auxiliary_loss_mlp": 0.01026593,
      "balance_loss_clip": 1.03791738,
      "balance_loss_mlp": 1.01949668,
      "epoch": 0.8451872783021703,
      "flos": 29825571605760.0,
      "grad_norm": 1.790828883094282,
      "language_loss": 0.78204376,
      "learning_rate": 2.46052355172385e-07,
      "loss": 0.80328321,
      "num_input_tokens_seen": 151932210,
      "step": 7029,
      "time_per_iteration": 2.7777621746063232
    },
    {
      "auxiliary_loss_clip": 0.01167353,
      "auxiliary_loss_mlp": 0.0102024,
      "balance_loss_clip": 1.04588389,
      "balance_loss_mlp": 1.01309979,
      "epoch": 0.8453075211928095,
      "flos": 21870029589120.0,
      "grad_norm": 2.058876764666817,
      "language_loss": 0.74269986,
      "learning_rate": 2.456781626228128e-07,
      "loss": 0.76457572,
      "num_input_tokens_seen": 151951715,
      "step": 7030,
      "time_per_iteration": 2.6207592487335205
    },
    {
      "auxiliary_loss_clip": 0.01022978,
      "auxiliary_loss_mlp": 0.00753964,
      "balance_loss_clip": 1.0097456,
      "balance_loss_mlp": 1.00023866,
      "epoch": 0.8454277640834486,
      "flos": 58751869288320.0,
      "grad_norm": 0.9294692087443818,
      "language_loss": 0.66250712,
      "learning_rate": 2.453042362024675e-07,
      "loss": 0.68027657,
      "num_input_tokens_seen": 152004960,
      "step": 7031,
      "time_per_iteration": 3.3698291778564453
    },
    {
      "auxiliary_loss_clip": 0.01163406,
      "auxiliary_loss_mlp": 0.01025278,
      "balance_loss_clip": 1.04462099,
      "balance_loss_mlp": 1.01822042,
      "epoch": 0.8455480069740876,
      "flos": 27090076469760.0,
      "grad_norm": 1.650540415263846,
      "language_loss": 0.73158592,
      "learning_rate": 2.449305759680751e-07,
      "loss": 0.75347275,
      "num_input_tokens_seen": 152026285,
      "step": 7032,
      "time_per_iteration": 2.8203601837158203
    },
    {
      "auxiliary_loss_clip": 0.01123231,
      "auxiliary_loss_mlp": 0.01025167,
      "balance_loss_clip": 1.04423881,
      "balance_loss_mlp": 1.01782322,
      "epoch": 0.8456682498647268,
      "flos": 27198670262400.0,
      "grad_norm": 1.5311474882454255,
      "language_loss": 0.75364614,
      "learning_rate": 2.445571819763188e-07,
      "loss": 0.77513009,
      "num_input_tokens_seen": 152048585,
      "step": 7033,
      "time_per_iteration": 2.74275541305542
    },
    {
      "auxiliary_loss_clip": 0.01165106,
      "auxiliary_loss_mlp": 0.01024014,
      "balance_loss_clip": 1.04581904,
      "balance_loss_mlp": 1.01724303,
      "epoch": 0.8457884927553658,
      "flos": 20631901737600.0,
      "grad_norm": 1.744691832268476,
      "language_loss": 0.58519399,
      "learning_rate": 2.4418405428384227e-07,
      "loss": 0.60708523,
      "num_input_tokens_seen": 152068795,
      "step": 7034,
      "time_per_iteration": 2.6598153114318848
    },
    {
      "auxiliary_loss_clip": 0.01164135,
      "auxiliary_loss_mlp": 0.0076206,
      "balance_loss_clip": 1.04432368,
      "balance_loss_mlp": 1.0003264,
      "epoch": 0.8459087356460049,
      "flos": 15299023259520.0,
      "grad_norm": 1.7901131306148133,
      "language_loss": 0.71786171,
      "learning_rate": 2.4381119294724864e-07,
      "loss": 0.73712373,
      "num_input_tokens_seen": 152086240,
      "step": 7035,
      "time_per_iteration": 2.750171184539795
    },
    {
      "auxiliary_loss_clip": 0.01165368,
      "auxiliary_loss_mlp": 0.01025723,
      "balance_loss_clip": 1.04599094,
      "balance_loss_mlp": 1.01909554,
      "epoch": 0.846028978536644,
      "flos": 18843155326080.0,
      "grad_norm": 6.305948687171912,
      "language_loss": 0.53793579,
      "learning_rate": 2.434385980231004e-07,
      "loss": 0.5598467,
      "num_input_tokens_seen": 152105080,
      "step": 7036,
      "time_per_iteration": 2.637657403945923
    },
    {
      "auxiliary_loss_clip": 0.0115339,
      "auxiliary_loss_mlp": 0.01025476,
      "balance_loss_clip": 1.04558074,
      "balance_loss_mlp": 1.01874328,
      "epoch": 0.8461492214272831,
      "flos": 52661740285440.0,
      "grad_norm": 1.7613863885543728,
      "language_loss": 0.65282613,
      "learning_rate": 2.4306626956792043e-07,
      "loss": 0.67461479,
      "num_input_tokens_seen": 152130025,
      "step": 7037,
      "time_per_iteration": 2.9432764053344727
    },
    {
      "auxiliary_loss_clip": 0.01150433,
      "auxiliary_loss_mlp": 0.01028093,
      "balance_loss_clip": 1.04275322,
      "balance_loss_mlp": 1.02119339,
      "epoch": 0.8462694643179222,
      "flos": 18588405093120.0,
      "grad_norm": 3.6945433498684905,
      "language_loss": 0.75846136,
      "learning_rate": 2.4269420763819017e-07,
      "loss": 0.78024662,
      "num_input_tokens_seen": 152148070,
      "step": 7038,
      "time_per_iteration": 2.576801061630249
    },
    {
      "auxiliary_loss_clip": 0.01147919,
      "auxiliary_loss_mlp": 0.01027504,
      "balance_loss_clip": 1.0434804,
      "balance_loss_mlp": 1.02045321,
      "epoch": 0.8463897072085613,
      "flos": 24387080163840.0,
      "grad_norm": 5.254667853083792,
      "language_loss": 0.83874106,
      "learning_rate": 2.4232241229035223e-07,
      "loss": 0.86049527,
      "num_input_tokens_seen": 152165825,
      "step": 7039,
      "time_per_iteration": 2.617311954498291
    },
    {
      "auxiliary_loss_clip": 0.01055637,
      "auxiliary_loss_mlp": 0.01000542,
      "balance_loss_clip": 1.01017976,
      "balance_loss_mlp": 0.99946958,
      "epoch": 0.8465099500992004,
      "flos": 68702140258560.0,
      "grad_norm": 0.7992574556891276,
      "language_loss": 0.56714725,
      "learning_rate": 2.419508835808064e-07,
      "loss": 0.58770907,
      "num_input_tokens_seen": 152222380,
      "step": 7040,
      "time_per_iteration": 3.148341655731201
    },
    {
      "auxiliary_loss_clip": 0.01138112,
      "auxiliary_loss_mlp": 0.01023773,
      "balance_loss_clip": 1.04408526,
      "balance_loss_mlp": 1.0161109,
      "epoch": 0.8466301929898394,
      "flos": 13735724561280.0,
      "grad_norm": 2.2954023563087227,
      "language_loss": 0.63160312,
      "learning_rate": 2.415796215659134e-07,
      "loss": 0.65322196,
      "num_input_tokens_seen": 152239085,
      "step": 7041,
      "time_per_iteration": 2.711677312850952
    },
    {
      "auxiliary_loss_clip": 0.01125935,
      "auxiliary_loss_mlp": 0.01027803,
      "balance_loss_clip": 1.03857005,
      "balance_loss_mlp": 1.02041793,
      "epoch": 0.8467504358804786,
      "flos": 19241260738560.0,
      "grad_norm": 1.9916184455094037,
      "language_loss": 0.7761305,
      "learning_rate": 2.412086263019939e-07,
      "loss": 0.79766786,
      "num_input_tokens_seen": 152257110,
      "step": 7042,
      "time_per_iteration": 2.719024896621704
    },
    {
      "auxiliary_loss_clip": 0.01161164,
      "auxiliary_loss_mlp": 0.01026071,
      "balance_loss_clip": 1.04586339,
      "balance_loss_mlp": 1.01968145,
      "epoch": 0.8468706787711177,
      "flos": 21324115710720.0,
      "grad_norm": 2.160140082011248,
      "language_loss": 0.79931897,
      "learning_rate": 2.408378978453276e-07,
      "loss": 0.82119131,
      "num_input_tokens_seen": 152277230,
      "step": 7043,
      "time_per_iteration": 2.6342647075653076
    },
    {
      "auxiliary_loss_clip": 0.01055115,
      "auxiliary_loss_mlp": 0.01002697,
      "balance_loss_clip": 1.00955546,
      "balance_loss_mlp": 1.00158262,
      "epoch": 0.8469909216617567,
      "flos": 64877439058560.0,
      "grad_norm": 0.8161229485733912,
      "language_loss": 0.63929856,
      "learning_rate": 2.404674362521533e-07,
      "loss": 0.6598767,
      "num_input_tokens_seen": 152335725,
      "step": 7044,
      "time_per_iteration": 3.118412971496582
    },
    {
      "auxiliary_loss_clip": 0.01151388,
      "auxiliary_loss_mlp": 0.01028961,
      "balance_loss_clip": 1.045223,
      "balance_loss_mlp": 1.02175784,
      "epoch": 0.8471111645523959,
      "flos": 19280583152640.0,
      "grad_norm": 2.2791399808911894,
      "language_loss": 0.74284536,
      "learning_rate": 2.4009724157866997e-07,
      "loss": 0.7646488,
      "num_input_tokens_seen": 152352785,
      "step": 7045,
      "time_per_iteration": 2.553032398223877
    },
    {
      "auxiliary_loss_clip": 0.01162876,
      "auxiliary_loss_mlp": 0.01022107,
      "balance_loss_clip": 1.04493117,
      "balance_loss_mlp": 1.01523507,
      "epoch": 0.8472314074430349,
      "flos": 22015826893440.0,
      "grad_norm": 2.0631344794248725,
      "language_loss": 0.76700121,
      "learning_rate": 2.3972731388103564e-07,
      "loss": 0.78885102,
      "num_input_tokens_seen": 152371265,
      "step": 7046,
      "time_per_iteration": 2.6717097759246826
    },
    {
      "auxiliary_loss_clip": 0.01005423,
      "auxiliary_loss_mlp": 0.01003665,
      "balance_loss_clip": 1.01020527,
      "balance_loss_mlp": 1.00258625,
      "epoch": 0.847351650333674,
      "flos": 57882580243200.0,
      "grad_norm": 0.8119867572220022,
      "language_loss": 0.62355244,
      "learning_rate": 2.393576532153687e-07,
      "loss": 0.64364332,
      "num_input_tokens_seen": 152435050,
      "step": 7047,
      "time_per_iteration": 3.4974520206451416
    },
    {
      "auxiliary_loss_clip": 0.01052422,
      "auxiliary_loss_mlp": 0.01001826,
      "balance_loss_clip": 1.01044035,
      "balance_loss_mlp": 1.00075328,
      "epoch": 0.8474718932243132,
      "flos": 41284238313600.0,
      "grad_norm": 0.9268370997294185,
      "language_loss": 0.57778144,
      "learning_rate": 2.389882596377453e-07,
      "loss": 0.59832394,
      "num_input_tokens_seen": 152489315,
      "step": 7048,
      "time_per_iteration": 4.663876056671143
    },
    {
      "auxiliary_loss_clip": 0.0116231,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.04331541,
      "balance_loss_mlp": 1.01999116,
      "epoch": 0.8475921361149522,
      "flos": 38180906974080.0,
      "grad_norm": 1.8488675169436,
      "language_loss": 0.76770693,
      "learning_rate": 2.386191332042031e-07,
      "loss": 0.7896018,
      "num_input_tokens_seen": 152511210,
      "step": 7049,
      "time_per_iteration": 2.7592718601226807
    },
    {
      "auxiliary_loss_clip": 0.01168773,
      "auxiliary_loss_mlp": 0.01027821,
      "balance_loss_clip": 1.04669166,
      "balance_loss_mlp": 1.0205102,
      "epoch": 0.8477123790055913,
      "flos": 25375054723200.0,
      "grad_norm": 1.6532979096179117,
      "language_loss": 0.72595763,
      "learning_rate": 2.3825027397073794e-07,
      "loss": 0.74792361,
      "num_input_tokens_seen": 152531685,
      "step": 7050,
      "time_per_iteration": 3.645735502243042
    },
    {
      "auxiliary_loss_clip": 0.01150227,
      "auxiliary_loss_mlp": 0.01023875,
      "balance_loss_clip": 1.04850125,
      "balance_loss_mlp": 1.01698756,
      "epoch": 0.8478326218962304,
      "flos": 30225185389440.0,
      "grad_norm": 2.6993129632934343,
      "language_loss": 0.66546714,
      "learning_rate": 2.3788168199330515e-07,
      "loss": 0.68720824,
      "num_input_tokens_seen": 152553245,
      "step": 7051,
      "time_per_iteration": 2.683839797973633
    },
    {
      "auxiliary_loss_clip": 0.01121504,
      "auxiliary_loss_mlp": 0.01024045,
      "balance_loss_clip": 1.0369426,
      "balance_loss_mlp": 1.01723528,
      "epoch": 0.8479528647868695,
      "flos": 38213800853760.0,
      "grad_norm": 1.652982133840298,
      "language_loss": 0.7270661,
      "learning_rate": 2.3751335732782074e-07,
      "loss": 0.74852157,
      "num_input_tokens_seen": 152574505,
      "step": 7052,
      "time_per_iteration": 2.813793897628784
    },
    {
      "auxiliary_loss_clip": 0.01151843,
      "auxiliary_loss_mlp": 0.01021552,
      "balance_loss_clip": 1.04649878,
      "balance_loss_mlp": 1.01481366,
      "epoch": 0.8480731076775085,
      "flos": 20957790856320.0,
      "grad_norm": 2.0146958948031295,
      "language_loss": 0.79529226,
      "learning_rate": 2.371453000301582e-07,
      "loss": 0.81702626,
      "num_input_tokens_seen": 152593190,
      "step": 7053,
      "time_per_iteration": 2.6675665378570557
    },
    {
      "auxiliary_loss_clip": 0.01119206,
      "auxiliary_loss_mlp": 0.01027432,
      "balance_loss_clip": 1.04158258,
      "balance_loss_mlp": 1.02029765,
      "epoch": 0.8481933505681477,
      "flos": 32596510487040.0,
      "grad_norm": 1.9168889550430275,
      "language_loss": 0.74312669,
      "learning_rate": 2.3677751015615222e-07,
      "loss": 0.76459306,
      "num_input_tokens_seen": 152615265,
      "step": 7054,
      "time_per_iteration": 4.62365984916687
    },
    {
      "auxiliary_loss_clip": 0.01126451,
      "auxiliary_loss_mlp": 0.01025454,
      "balance_loss_clip": 1.03987098,
      "balance_loss_mlp": 1.0184207,
      "epoch": 0.8483135934587868,
      "flos": 20741177888640.0,
      "grad_norm": 2.191017697170778,
      "language_loss": 0.8551929,
      "learning_rate": 2.3640998776159593e-07,
      "loss": 0.87671196,
      "num_input_tokens_seen": 152632770,
      "step": 7055,
      "time_per_iteration": 2.774322748184204
    },
    {
      "auxiliary_loss_clip": 0.01136604,
      "auxiliary_loss_mlp": 0.01023614,
      "balance_loss_clip": 1.04237628,
      "balance_loss_mlp": 1.0173521,
      "epoch": 0.8484338363494258,
      "flos": 21653057485440.0,
      "grad_norm": 1.827601702448024,
      "language_loss": 0.81193769,
      "learning_rate": 2.3604273290224253e-07,
      "loss": 0.8335399,
      "num_input_tokens_seen": 152653485,
      "step": 7056,
      "time_per_iteration": 2.6664516925811768
    },
    {
      "auxiliary_loss_clip": 0.0113947,
      "auxiliary_loss_mlp": 0.01026124,
      "balance_loss_clip": 1.04442155,
      "balance_loss_mlp": 1.01858664,
      "epoch": 0.848554079240065,
      "flos": 15013964926080.0,
      "grad_norm": 1.8581342283989246,
      "language_loss": 0.74364591,
      "learning_rate": 2.356757456338039e-07,
      "loss": 0.76530182,
      "num_input_tokens_seen": 152670970,
      "step": 7057,
      "time_per_iteration": 2.6567416191101074
    },
    {
      "auxiliary_loss_clip": 0.01039184,
      "auxiliary_loss_mlp": 0.01002229,
      "balance_loss_clip": 1.00913858,
      "balance_loss_mlp": 1.00117385,
      "epoch": 0.848674322130704,
      "flos": 68060453742720.0,
      "grad_norm": 0.751717422672945,
      "language_loss": 0.59034991,
      "learning_rate": 2.3530902601195147e-07,
      "loss": 0.61076403,
      "num_input_tokens_seen": 152739460,
      "step": 7058,
      "time_per_iteration": 3.301959276199341
    },
    {
      "auxiliary_loss_clip": 0.01150516,
      "auxiliary_loss_mlp": 0.01028518,
      "balance_loss_clip": 1.04451203,
      "balance_loss_mlp": 1.02086449,
      "epoch": 0.8487945650213431,
      "flos": 18475788977280.0,
      "grad_norm": 2.0924921533219027,
      "language_loss": 0.79036587,
      "learning_rate": 2.34942574092317e-07,
      "loss": 0.8121562,
      "num_input_tokens_seen": 152754710,
      "step": 7059,
      "time_per_iteration": 2.6350607872009277
    },
    {
      "auxiliary_loss_clip": 0.01156126,
      "auxiliary_loss_mlp": 0.01023368,
      "balance_loss_clip": 1.04537535,
      "balance_loss_mlp": 1.01639414,
      "epoch": 0.8489148079119821,
      "flos": 23473189405440.0,
      "grad_norm": 1.8922180567389206,
      "language_loss": 0.76386482,
      "learning_rate": 2.3457638993049045e-07,
      "loss": 0.78565979,
      "num_input_tokens_seen": 152772700,
      "step": 7060,
      "time_per_iteration": 2.635366201400757
    },
    {
      "auxiliary_loss_clip": 0.01098113,
      "auxiliary_loss_mlp": 0.01024242,
      "balance_loss_clip": 1.04231179,
      "balance_loss_mlp": 1.01685381,
      "epoch": 0.8490350508026213,
      "flos": 19937604775680.0,
      "grad_norm": 1.9923190405533386,
      "language_loss": 0.63618362,
      "learning_rate": 2.3421047358202252e-07,
      "loss": 0.65740722,
      "num_input_tokens_seen": 152791550,
      "step": 7061,
      "time_per_iteration": 2.800842761993408
    },
    {
      "auxiliary_loss_clip": 0.01153311,
      "auxiliary_loss_mlp": 0.01023833,
      "balance_loss_clip": 1.04460144,
      "balance_loss_mlp": 1.01626873,
      "epoch": 0.8491552936932604,
      "flos": 24279958828800.0,
      "grad_norm": 4.274903663826911,
      "language_loss": 0.83620596,
      "learning_rate": 2.3384482510242144e-07,
      "loss": 0.85797745,
      "num_input_tokens_seen": 152809410,
      "step": 7062,
      "time_per_iteration": 2.669896125793457
    },
    {
      "auxiliary_loss_clip": 0.01166725,
      "auxiliary_loss_mlp": 0.01027486,
      "balance_loss_clip": 1.04463291,
      "balance_loss_mlp": 1.02065849,
      "epoch": 0.8492755365838994,
      "flos": 22522526098560.0,
      "grad_norm": 1.9549202490808835,
      "language_loss": 0.77283823,
      "learning_rate": 2.3347944454715575e-07,
      "loss": 0.79478031,
      "num_input_tokens_seen": 152825800,
      "step": 7063,
      "time_per_iteration": 2.6135904788970947
    },
    {
      "auxiliary_loss_clip": 0.01167854,
      "auxiliary_loss_mlp": 0.01030056,
      "balance_loss_clip": 1.04610193,
      "balance_loss_mlp": 1.02259374,
      "epoch": 0.8493957794745386,
      "flos": 26980441182720.0,
      "grad_norm": 1.6891355432433717,
      "language_loss": 0.67207617,
      "learning_rate": 2.331143319716542e-07,
      "loss": 0.6940552,
      "num_input_tokens_seen": 152845330,
      "step": 7064,
      "time_per_iteration": 2.7460579872131348
    },
    {
      "auxiliary_loss_clip": 0.01128961,
      "auxiliary_loss_mlp": 0.01022693,
      "balance_loss_clip": 1.04186976,
      "balance_loss_mlp": 1.01542735,
      "epoch": 0.8495160223651776,
      "flos": 29861985018240.0,
      "grad_norm": 2.5246985088780667,
      "language_loss": 0.65696877,
      "learning_rate": 2.3274948743130363e-07,
      "loss": 0.67848533,
      "num_input_tokens_seen": 152865165,
      "step": 7065,
      "time_per_iteration": 2.8012807369232178
    },
    {
      "auxiliary_loss_clip": 0.01163962,
      "auxiliary_loss_mlp": 0.01024213,
      "balance_loss_clip": 1.04373598,
      "balance_loss_mlp": 1.01723289,
      "epoch": 0.8496362652558167,
      "flos": 23075443128960.0,
      "grad_norm": 1.8131028880251328,
      "language_loss": 0.79195386,
      "learning_rate": 2.3238491098145085e-07,
      "loss": 0.81383562,
      "num_input_tokens_seen": 152884695,
      "step": 7066,
      "time_per_iteration": 2.6099555492401123
    },
    {
      "auxiliary_loss_clip": 0.01152292,
      "auxiliary_loss_mlp": 0.01023056,
      "balance_loss_clip": 1.04537451,
      "balance_loss_mlp": 1.01535547,
      "epoch": 0.8497565081464559,
      "flos": 14609107756800.0,
      "grad_norm": 2.163188294058064,
      "language_loss": 0.73331773,
      "learning_rate": 2.3202060267740141e-07,
      "loss": 0.75507128,
      "num_input_tokens_seen": 152902220,
      "step": 7067,
      "time_per_iteration": 2.5986642837524414
    },
    {
      "auxiliary_loss_clip": 0.01102972,
      "auxiliary_loss_mlp": 0.0102424,
      "balance_loss_clip": 1.03673494,
      "balance_loss_mlp": 1.01700985,
      "epoch": 0.8498767510370949,
      "flos": 21136446126720.0,
      "grad_norm": 2.1650669952287505,
      "language_loss": 0.77689844,
      "learning_rate": 2.3165656257442044e-07,
      "loss": 0.79817057,
      "num_input_tokens_seen": 152920740,
      "step": 7068,
      "time_per_iteration": 2.736037254333496
    },
    {
      "auxiliary_loss_clip": 0.01147295,
      "auxiliary_loss_mlp": 0.01024972,
      "balance_loss_clip": 1.0445056,
      "balance_loss_mlp": 1.01859462,
      "epoch": 0.849996993927734,
      "flos": 23654538195840.0,
      "grad_norm": 1.970296837928355,
      "language_loss": 0.90339291,
      "learning_rate": 2.31292790727734e-07,
      "loss": 0.92511559,
      "num_input_tokens_seen": 152938305,
      "step": 7069,
      "time_per_iteration": 2.6501681804656982
    },
    {
      "auxiliary_loss_clip": 0.01161699,
      "auxiliary_loss_mlp": 0.01022073,
      "balance_loss_clip": 1.04326999,
      "balance_loss_mlp": 1.0152452,
      "epoch": 0.8501172368183731,
      "flos": 20558069331840.0,
      "grad_norm": 2.7157938259234595,
      "language_loss": 0.80252659,
      "learning_rate": 2.3092928719252392e-07,
      "loss": 0.8243643,
      "num_input_tokens_seen": 152956705,
      "step": 7070,
      "time_per_iteration": 2.617213487625122
    },
    {
      "auxiliary_loss_clip": 0.01149561,
      "auxiliary_loss_mlp": 0.01027915,
      "balance_loss_clip": 1.04398203,
      "balance_loss_mlp": 1.02051854,
      "epoch": 0.8502374797090122,
      "flos": 22272624201600.0,
      "grad_norm": 2.1384330409842507,
      "language_loss": 0.78838897,
      "learning_rate": 2.3056605202393475e-07,
      "loss": 0.81016374,
      "num_input_tokens_seen": 152974265,
      "step": 7071,
      "time_per_iteration": 2.6039540767669678
    },
    {
      "auxiliary_loss_clip": 0.01144646,
      "auxiliary_loss_mlp": 0.00762449,
      "balance_loss_clip": 1.03988528,
      "balance_loss_mlp": 1.00029945,
      "epoch": 0.8503577225996513,
      "flos": 23659817495040.0,
      "grad_norm": 2.141804217037687,
      "language_loss": 0.66768324,
      "learning_rate": 2.3020308527706888e-07,
      "loss": 0.68675423,
      "num_input_tokens_seen": 152993680,
      "step": 7072,
      "time_per_iteration": 2.6783671379089355
    },
    {
      "auxiliary_loss_clip": 0.01140746,
      "auxiliary_loss_mlp": 0.01029438,
      "balance_loss_clip": 1.04054523,
      "balance_loss_mlp": 1.02291739,
      "epoch": 0.8504779654902904,
      "flos": 26758513002240.0,
      "grad_norm": 1.7001826320622906,
      "language_loss": 0.88688457,
      "learning_rate": 2.2984038700698715e-07,
      "loss": 0.90858644,
      "num_input_tokens_seen": 153012990,
      "step": 7073,
      "time_per_iteration": 2.6809561252593994
    },
    {
      "auxiliary_loss_clip": 0.01149177,
      "auxiliary_loss_mlp": 0.01031452,
      "balance_loss_clip": 1.04541969,
      "balance_loss_mlp": 1.02455616,
      "epoch": 0.8505982083809295,
      "flos": 26468247196800.0,
      "grad_norm": 1.5823771237171476,
      "language_loss": 0.78990161,
      "learning_rate": 2.2947795726871222e-07,
      "loss": 0.81170785,
      "num_input_tokens_seen": 153034015,
      "step": 7074,
      "time_per_iteration": 3.6629855632781982
    },
    {
      "auxiliary_loss_clip": 0.01150307,
      "auxiliary_loss_mlp": 0.00762433,
      "balance_loss_clip": 1.04792476,
      "balance_loss_mlp": 1.00032783,
      "epoch": 0.8507184512715685,
      "flos": 20303390926080.0,
      "grad_norm": 1.7353819702728206,
      "language_loss": 0.85607344,
      "learning_rate": 2.2911579611722253e-07,
      "loss": 0.87520087,
      "num_input_tokens_seen": 153053160,
      "step": 7075,
      "time_per_iteration": 2.61295485496521
    },
    {
      "auxiliary_loss_clip": 0.0113381,
      "auxiliary_loss_mlp": 0.01029057,
      "balance_loss_clip": 1.04171646,
      "balance_loss_mlp": 1.02221155,
      "epoch": 0.8508386941622077,
      "flos": 19025186474880.0,
      "grad_norm": 1.8119139228994352,
      "language_loss": 0.8735351,
      "learning_rate": 2.2875390360745905e-07,
      "loss": 0.89516377,
      "num_input_tokens_seen": 153072565,
      "step": 7076,
      "time_per_iteration": 3.5926101207733154
    },
    {
      "auxiliary_loss_clip": 0.0112874,
      "auxiliary_loss_mlp": 0.01025363,
      "balance_loss_clip": 1.04088712,
      "balance_loss_mlp": 1.01793611,
      "epoch": 0.8509589370528468,
      "flos": 16433405654400.0,
      "grad_norm": 1.9384262258085811,
      "language_loss": 0.77594537,
      "learning_rate": 2.2839227979432008e-07,
      "loss": 0.79748642,
      "num_input_tokens_seen": 153090215,
      "step": 7077,
      "time_per_iteration": 2.831124782562256
    },
    {
      "auxiliary_loss_clip": 0.0113754,
      "auxiliary_loss_mlp": 0.01020834,
      "balance_loss_clip": 1.04098701,
      "balance_loss_mlp": 1.01420879,
      "epoch": 0.8510791799434858,
      "flos": 18259714713600.0,
      "grad_norm": 2.5612378665857287,
      "language_loss": 0.85377669,
      "learning_rate": 2.2803092473266373e-07,
      "loss": 0.87536049,
      "num_input_tokens_seen": 153107740,
      "step": 7078,
      "time_per_iteration": 2.7760732173919678
    },
    {
      "auxiliary_loss_clip": 0.01167695,
      "auxiliary_loss_mlp": 0.01030133,
      "balance_loss_clip": 1.04660594,
      "balance_loss_mlp": 1.02268267,
      "epoch": 0.851199422834125,
      "flos": 23441372933760.0,
      "grad_norm": 2.2422804676477193,
      "language_loss": 0.86557108,
      "learning_rate": 2.2766983847730724e-07,
      "loss": 0.8875494,
      "num_input_tokens_seen": 153127410,
      "step": 7079,
      "time_per_iteration": 2.698514699935913
    },
    {
      "auxiliary_loss_clip": 0.01132373,
      "auxiliary_loss_mlp": 0.01029787,
      "balance_loss_clip": 1.04086173,
      "balance_loss_mlp": 1.02166915,
      "epoch": 0.851319665724764,
      "flos": 16289404030080.0,
      "grad_norm": 1.8402620885932484,
      "language_loss": 0.66402376,
      "learning_rate": 2.2730902108302663e-07,
      "loss": 0.68564534,
      "num_input_tokens_seen": 153144325,
      "step": 7080,
      "time_per_iteration": 4.4326677322387695
    },
    {
      "auxiliary_loss_clip": 0.01129637,
      "auxiliary_loss_mlp": 0.01027865,
      "balance_loss_clip": 1.03969181,
      "balance_loss_mlp": 1.02057505,
      "epoch": 0.8514399086154031,
      "flos": 18989347680000.0,
      "grad_norm": 2.5865598695513774,
      "language_loss": 0.68666607,
      "learning_rate": 2.269484726045583e-07,
      "loss": 0.70824111,
      "num_input_tokens_seen": 153163240,
      "step": 7081,
      "time_per_iteration": 2.653665065765381
    },
    {
      "auxiliary_loss_clip": 0.01128071,
      "auxiliary_loss_mlp": 0.01023198,
      "balance_loss_clip": 1.04253709,
      "balance_loss_mlp": 1.01634955,
      "epoch": 0.8515601515060423,
      "flos": 24571194301440.0,
      "grad_norm": 1.6435528000268327,
      "language_loss": 0.79199278,
      "learning_rate": 2.2658819309659672e-07,
      "loss": 0.81350553,
      "num_input_tokens_seen": 153183440,
      "step": 7082,
      "time_per_iteration": 2.714900255203247
    },
    {
      "auxiliary_loss_clip": 0.01134288,
      "auxiliary_loss_mlp": 0.01024306,
      "balance_loss_clip": 1.04427671,
      "balance_loss_mlp": 1.01786315,
      "epoch": 0.8516803943966813,
      "flos": 19529443555200.0,
      "grad_norm": 1.9784495074987636,
      "language_loss": 0.84981954,
      "learning_rate": 2.2622818261379706e-07,
      "loss": 0.87140548,
      "num_input_tokens_seen": 153200460,
      "step": 7083,
      "time_per_iteration": 2.6477298736572266
    },
    {
      "auxiliary_loss_clip": 0.01135917,
      "auxiliary_loss_mlp": 0.01025897,
      "balance_loss_clip": 1.04232275,
      "balance_loss_mlp": 1.01854444,
      "epoch": 0.8518006372873204,
      "flos": 20265792364800.0,
      "grad_norm": 1.705143643502875,
      "language_loss": 0.74859035,
      "learning_rate": 2.2586844121077142e-07,
      "loss": 0.77020854,
      "num_input_tokens_seen": 153218970,
      "step": 7084,
      "time_per_iteration": 2.6346898078918457
    },
    {
      "auxiliary_loss_clip": 0.0111266,
      "auxiliary_loss_mlp": 0.01026998,
      "balance_loss_clip": 1.03935778,
      "balance_loss_mlp": 1.01955295,
      "epoch": 0.8519208801779595,
      "flos": 24133227770880.0,
      "grad_norm": 2.0887545921494977,
      "language_loss": 0.71639442,
      "learning_rate": 2.2550896894209215e-07,
      "loss": 0.73779106,
      "num_input_tokens_seen": 153238485,
      "step": 7085,
      "time_per_iteration": 2.7445759773254395
    },
    {
      "auxiliary_loss_clip": 0.01014058,
      "auxiliary_loss_mlp": 0.01001973,
      "balance_loss_clip": 1.01144266,
      "balance_loss_mlp": 1.00085247,
      "epoch": 0.8520411230685986,
      "flos": 63035223252480.0,
      "grad_norm": 0.6889700179101199,
      "language_loss": 0.56611234,
      "learning_rate": 2.2514976586229184e-07,
      "loss": 0.58627266,
      "num_input_tokens_seen": 153306430,
      "step": 7086,
      "time_per_iteration": 3.5789618492126465
    },
    {
      "auxiliary_loss_clip": 0.01055412,
      "auxiliary_loss_mlp": 0.01001301,
      "balance_loss_clip": 1.01069546,
      "balance_loss_mlp": 1.00024021,
      "epoch": 0.8521613659592376,
      "flos": 65836865283840.0,
      "grad_norm": 0.7493191452507887,
      "language_loss": 0.54590893,
      "learning_rate": 2.247908320258609e-07,
      "loss": 0.56647611,
      "num_input_tokens_seen": 153366520,
      "step": 7087,
      "time_per_iteration": 3.393077850341797
    },
    {
      "auxiliary_loss_clip": 0.01104357,
      "auxiliary_loss_mlp": 0.01024781,
      "balance_loss_clip": 1.04156566,
      "balance_loss_mlp": 1.01749992,
      "epoch": 0.8522816088498768,
      "flos": 23112323418240.0,
      "grad_norm": 2.2504354816639105,
      "language_loss": 0.79599226,
      "learning_rate": 2.2443216748724914e-07,
      "loss": 0.81728369,
      "num_input_tokens_seen": 153387230,
      "step": 7088,
      "time_per_iteration": 2.785992383956909
    },
    {
      "auxiliary_loss_clip": 0.01155361,
      "auxiliary_loss_mlp": 0.00762343,
      "balance_loss_clip": 1.04596686,
      "balance_loss_mlp": 1.00036216,
      "epoch": 0.8524018517405159,
      "flos": 31758140073600.0,
      "grad_norm": 1.8550164917537284,
      "language_loss": 0.74374211,
      "learning_rate": 2.2407377230086588e-07,
      "loss": 0.76291919,
      "num_input_tokens_seen": 153409585,
      "step": 7089,
      "time_per_iteration": 2.7844386100769043
    },
    {
      "auxiliary_loss_clip": 0.01120924,
      "auxiliary_loss_mlp": 0.01024855,
      "balance_loss_clip": 1.04258204,
      "balance_loss_mlp": 1.01760161,
      "epoch": 0.8525220946311549,
      "flos": 18690318956160.0,
      "grad_norm": 2.0523451367161853,
      "language_loss": 0.83916366,
      "learning_rate": 2.23715646521079e-07,
      "loss": 0.86062145,
      "num_input_tokens_seen": 153427105,
      "step": 7090,
      "time_per_iteration": 2.854074239730835
    },
    {
      "auxiliary_loss_clip": 0.01154577,
      "auxiliary_loss_mlp": 0.00762573,
      "balance_loss_clip": 1.04452956,
      "balance_loss_mlp": 1.00036705,
      "epoch": 0.852642337521794,
      "flos": 21793216354560.0,
      "grad_norm": 4.291956274384828,
      "language_loss": 0.84273869,
      "learning_rate": 2.2335779020221724e-07,
      "loss": 0.86191016,
      "num_input_tokens_seen": 153443725,
      "step": 7091,
      "time_per_iteration": 2.7256882190704346
    },
    {
      "auxiliary_loss_clip": 0.01052688,
      "auxiliary_loss_mlp": 0.01000873,
      "balance_loss_clip": 1.01417351,
      "balance_loss_mlp": 0.99979991,
      "epoch": 0.8527625804124331,
      "flos": 69040132260480.0,
      "grad_norm": 0.8048091126206192,
      "language_loss": 0.56422871,
      "learning_rate": 2.2300020339856497e-07,
      "loss": 0.58476436,
      "num_input_tokens_seen": 153506410,
      "step": 7092,
      "time_per_iteration": 3.238832473754883
    },
    {
      "auxiliary_loss_clip": 0.01134335,
      "auxiliary_loss_mlp": 0.01024407,
      "balance_loss_clip": 1.0425086,
      "balance_loss_mlp": 1.01706934,
      "epoch": 0.8528828233030722,
      "flos": 26979399688320.0,
      "grad_norm": 2.262614103193169,
      "language_loss": 0.78400278,
      "learning_rate": 2.2264288616436966e-07,
      "loss": 0.80559015,
      "num_input_tokens_seen": 153526665,
      "step": 7093,
      "time_per_iteration": 2.6890523433685303
    },
    {
      "auxiliary_loss_clip": 0.01134363,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.04337466,
      "balance_loss_mlp": 1.02256227,
      "epoch": 0.8530030661937112,
      "flos": 17487598936320.0,
      "grad_norm": 2.2023646900109033,
      "language_loss": 0.7245028,
      "learning_rate": 2.222858385538351e-07,
      "loss": 0.74614215,
      "num_input_tokens_seen": 153543465,
      "step": 7094,
      "time_per_iteration": 2.7223308086395264
    },
    {
      "auxiliary_loss_clip": 0.0114836,
      "auxiliary_loss_mlp": 0.01021044,
      "balance_loss_clip": 1.04345775,
      "balance_loss_mlp": 1.01439774,
      "epoch": 0.8531233090843504,
      "flos": 22160798184960.0,
      "grad_norm": 1.6399211981645385,
      "language_loss": 0.68004733,
      "learning_rate": 2.2192906062112527e-07,
      "loss": 0.70174134,
      "num_input_tokens_seen": 153563340,
      "step": 7095,
      "time_per_iteration": 2.6311800479888916
    },
    {
      "auxiliary_loss_clip": 0.01164545,
      "auxiliary_loss_mlp": 0.01023573,
      "balance_loss_clip": 1.04381084,
      "balance_loss_mlp": 1.01725459,
      "epoch": 0.8532435519749895,
      "flos": 37635388145280.0,
      "grad_norm": 1.6584030410558421,
      "language_loss": 0.70693767,
      "learning_rate": 2.2157255242036377e-07,
      "loss": 0.72881883,
      "num_input_tokens_seen": 153587005,
      "step": 7096,
      "time_per_iteration": 2.7560884952545166
    },
    {
      "auxiliary_loss_clip": 0.01119321,
      "auxiliary_loss_mlp": 0.01027299,
      "balance_loss_clip": 1.04067373,
      "balance_loss_mlp": 1.01914525,
      "epoch": 0.8533637948656285,
      "flos": 21398163598080.0,
      "grad_norm": 2.076162762638185,
      "language_loss": 0.74634242,
      "learning_rate": 2.2121631400563135e-07,
      "loss": 0.76780868,
      "num_input_tokens_seen": 153606835,
      "step": 7097,
      "time_per_iteration": 2.74774169921875
    },
    {
      "auxiliary_loss_clip": 0.0104845,
      "auxiliary_loss_mlp": 0.01001789,
      "balance_loss_clip": 1.00860381,
      "balance_loss_mlp": 1.00074565,
      "epoch": 0.8534840377562677,
      "flos": 53345122490880.0,
      "grad_norm": 0.7930277828687593,
      "language_loss": 0.52901196,
      "learning_rate": 2.208603454309701e-07,
      "loss": 0.54951441,
      "num_input_tokens_seen": 153664925,
      "step": 7098,
      "time_per_iteration": 3.1556763648986816
    },
    {
      "auxiliary_loss_clip": 0.01108362,
      "auxiliary_loss_mlp": 0.01025402,
      "balance_loss_clip": 1.0405122,
      "balance_loss_mlp": 1.01813912,
      "epoch": 0.8536042806469067,
      "flos": 20814148368000.0,
      "grad_norm": 3.540980250591467,
      "language_loss": 0.7061255,
      "learning_rate": 2.2050464675037994e-07,
      "loss": 0.72746313,
      "num_input_tokens_seen": 153683550,
      "step": 7099,
      "time_per_iteration": 2.7320668697357178
    },
    {
      "auxiliary_loss_clip": 0.01136601,
      "auxiliary_loss_mlp": 0.01029218,
      "balance_loss_clip": 1.04442334,
      "balance_loss_mlp": 1.02215147,
      "epoch": 0.8537245235375458,
      "flos": 24681368292480.0,
      "grad_norm": 2.086214535371896,
      "language_loss": 0.72838306,
      "learning_rate": 2.2014921801782016e-07,
      "loss": 0.75004125,
      "num_input_tokens_seen": 153703040,
      "step": 7100,
      "time_per_iteration": 3.6332132816314697
    },
    {
      "auxiliary_loss_clip": 0.01137397,
      "auxiliary_loss_mlp": 0.01023724,
      "balance_loss_clip": 1.03944874,
      "balance_loss_mlp": 1.01654124,
      "epoch": 0.853844766428185,
      "flos": 24384817607040.0,
      "grad_norm": 1.9552713810402769,
      "language_loss": 0.7394681,
      "learning_rate": 2.1979405928720872e-07,
      "loss": 0.76107937,
      "num_input_tokens_seen": 153722695,
      "step": 7101,
      "time_per_iteration": 2.6510138511657715
    },
    {
      "auxiliary_loss_clip": 0.01138395,
      "auxiliary_loss_mlp": 0.0102267,
      "balance_loss_clip": 1.04154146,
      "balance_loss_mlp": 1.01558876,
      "epoch": 0.853965009318824,
      "flos": 20955707867520.0,
      "grad_norm": 1.4835219857056998,
      "language_loss": 0.79484034,
      "learning_rate": 2.1943917061242257e-07,
      "loss": 0.81645095,
      "num_input_tokens_seen": 153742550,
      "step": 7102,
      "time_per_iteration": 4.112707853317261
    },
    {
      "auxiliary_loss_clip": 0.01158572,
      "auxiliary_loss_mlp": 0.00762855,
      "balance_loss_clip": 1.04527473,
      "balance_loss_mlp": 1.00036097,
      "epoch": 0.8540852522094631,
      "flos": 24201816791040.0,
      "grad_norm": 1.564708585845891,
      "language_loss": 0.66497183,
      "learning_rate": 2.1908455204729903e-07,
      "loss": 0.6841861,
      "num_input_tokens_seen": 153761700,
      "step": 7103,
      "time_per_iteration": 2.779060125350952
    },
    {
      "auxiliary_loss_clip": 0.01135353,
      "auxiliary_loss_mlp": 0.01025456,
      "balance_loss_clip": 1.04021275,
      "balance_loss_mlp": 1.01821375,
      "epoch": 0.8542054951001022,
      "flos": 25082921410560.0,
      "grad_norm": 2.1065440857517164,
      "language_loss": 0.78374982,
      "learning_rate": 2.1873020364563265e-07,
      "loss": 0.80535787,
      "num_input_tokens_seen": 153780765,
      "step": 7104,
      "time_per_iteration": 2.723001003265381
    },
    {
      "auxiliary_loss_clip": 0.01148081,
      "auxiliary_loss_mlp": 0.01021871,
      "balance_loss_clip": 1.04430282,
      "balance_loss_mlp": 1.01515317,
      "epoch": 0.8543257379907413,
      "flos": 24316551809280.0,
      "grad_norm": 2.400422158840647,
      "language_loss": 0.76084709,
      "learning_rate": 2.183761254611789e-07,
      "loss": 0.78254664,
      "num_input_tokens_seen": 153801090,
      "step": 7105,
      "time_per_iteration": 2.7377679347991943
    },
    {
      "auxiliary_loss_clip": 0.01149169,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.04580855,
      "balance_loss_mlp": 1.0206089,
      "epoch": 0.8544459808813804,
      "flos": 55286630467200.0,
      "grad_norm": 2.143652440550313,
      "language_loss": 0.69949925,
      "learning_rate": 2.1802231754764987e-07,
      "loss": 0.72126496,
      "num_input_tokens_seen": 153826530,
      "step": 7106,
      "time_per_iteration": 4.792828559875488
    },
    {
      "auxiliary_loss_clip": 0.01137755,
      "auxiliary_loss_mlp": 0.01024947,
      "balance_loss_clip": 1.04046607,
      "balance_loss_mlp": 1.01799083,
      "epoch": 0.8545662237720195,
      "flos": 25776248705280.0,
      "grad_norm": 1.9615746322677676,
      "language_loss": 0.76389998,
      "learning_rate": 2.17668779958718e-07,
      "loss": 0.78552699,
      "num_input_tokens_seen": 153849110,
      "step": 7107,
      "time_per_iteration": 2.67134428024292
    },
    {
      "auxiliary_loss_clip": 0.01165664,
      "auxiliary_loss_mlp": 0.01024269,
      "balance_loss_clip": 1.04648924,
      "balance_loss_mlp": 1.01675034,
      "epoch": 0.8546864666626586,
      "flos": 11108320427520.0,
      "grad_norm": 3.3725118989668466,
      "language_loss": 0.80724597,
      "learning_rate": 2.1731551274801553e-07,
      "loss": 0.82914531,
      "num_input_tokens_seen": 153865550,
      "step": 7108,
      "time_per_iteration": 2.6561856269836426
    },
    {
      "auxiliary_loss_clip": 0.01138656,
      "auxiliary_loss_mlp": 0.01022654,
      "balance_loss_clip": 1.0438087,
      "balance_loss_mlp": 1.0157876,
      "epoch": 0.8548067095532976,
      "flos": 25520169669120.0,
      "grad_norm": 1.9231641265825687,
      "language_loss": 0.61299968,
      "learning_rate": 2.169625159691324e-07,
      "loss": 0.6346128,
      "num_input_tokens_seen": 153885425,
      "step": 7109,
      "time_per_iteration": 2.714599132537842
    },
    {
      "auxiliary_loss_clip": 0.01116069,
      "auxiliary_loss_mlp": 0.01024428,
      "balance_loss_clip": 1.03891826,
      "balance_loss_mlp": 1.01629782,
      "epoch": 0.8549269524439368,
      "flos": 24717853532160.0,
      "grad_norm": 1.9540620734166119,
      "language_loss": 0.74674201,
      "learning_rate": 2.1660978967561784e-07,
      "loss": 0.76814699,
      "num_input_tokens_seen": 153904760,
      "step": 7110,
      "time_per_iteration": 2.840850591659546
    },
    {
      "auxiliary_loss_clip": 0.01162628,
      "auxiliary_loss_mlp": 0.01024728,
      "balance_loss_clip": 1.04370904,
      "balance_loss_mlp": 1.01795697,
      "epoch": 0.8550471953345758,
      "flos": 19825599191040.0,
      "grad_norm": 8.319640844703247,
      "language_loss": 0.78994304,
      "learning_rate": 2.1625733392098035e-07,
      "loss": 0.81181657,
      "num_input_tokens_seen": 153920370,
      "step": 7111,
      "time_per_iteration": 2.6063499450683594
    },
    {
      "auxiliary_loss_clip": 0.01163846,
      "auxiliary_loss_mlp": 0.01023666,
      "balance_loss_clip": 1.04456425,
      "balance_loss_mlp": 1.01710665,
      "epoch": 0.8551674382252149,
      "flos": 22820441500800.0,
      "grad_norm": 2.0938802725730548,
      "language_loss": 0.79841703,
      "learning_rate": 2.159051487586867e-07,
      "loss": 0.82029212,
      "num_input_tokens_seen": 153940500,
      "step": 7112,
      "time_per_iteration": 2.582059144973755
    },
    {
      "auxiliary_loss_clip": 0.01142514,
      "auxiliary_loss_mlp": 0.01026769,
      "balance_loss_clip": 1.04594505,
      "balance_loss_mlp": 1.01888061,
      "epoch": 0.8552876811158541,
      "flos": 20631255292800.0,
      "grad_norm": 2.394891833624902,
      "language_loss": 0.72912782,
      "learning_rate": 2.155532342421642e-07,
      "loss": 0.75082064,
      "num_input_tokens_seen": 153958500,
      "step": 7113,
      "time_per_iteration": 2.6642394065856934
    },
    {
      "auxiliary_loss_clip": 0.01156171,
      "auxiliary_loss_mlp": 0.01029846,
      "balance_loss_clip": 1.04496002,
      "balance_loss_mlp": 1.02274752,
      "epoch": 0.8554079240064931,
      "flos": 23112359331840.0,
      "grad_norm": 1.6664629764834973,
      "language_loss": 0.78411943,
      "learning_rate": 2.1520159042479636e-07,
      "loss": 0.80597961,
      "num_input_tokens_seen": 153976790,
      "step": 7114,
      "time_per_iteration": 2.664931535720825
    },
    {
      "auxiliary_loss_clip": 0.01149161,
      "auxiliary_loss_mlp": 0.01025135,
      "balance_loss_clip": 1.04433656,
      "balance_loss_mlp": 1.01834917,
      "epoch": 0.8555281668971322,
      "flos": 22128047959680.0,
      "grad_norm": 1.9371589484070706,
      "language_loss": 0.70808703,
      "learning_rate": 2.148502173599287e-07,
      "loss": 0.72982997,
      "num_input_tokens_seen": 153994930,
      "step": 7115,
      "time_per_iteration": 2.603637218475342
    },
    {
      "auxiliary_loss_clip": 0.01132291,
      "auxiliary_loss_mlp": 0.0102597,
      "balance_loss_clip": 1.04285383,
      "balance_loss_mlp": 1.01785481,
      "epoch": 0.8556484097877713,
      "flos": 31139040234240.0,
      "grad_norm": 1.6940619583707974,
      "language_loss": 0.65812707,
      "learning_rate": 2.1449911510086372e-07,
      "loss": 0.67970967,
      "num_input_tokens_seen": 154014400,
      "step": 7116,
      "time_per_iteration": 2.746556043624878
    },
    {
      "auxiliary_loss_clip": 0.0114909,
      "auxiliary_loss_mlp": 0.01025169,
      "balance_loss_clip": 1.04360104,
      "balance_loss_mlp": 1.01854348,
      "epoch": 0.8557686526784104,
      "flos": 24316551809280.0,
      "grad_norm": 2.0868054024809357,
      "language_loss": 0.77025682,
      "learning_rate": 2.141482837008628e-07,
      "loss": 0.7919994,
      "num_input_tokens_seen": 154034940,
      "step": 7117,
      "time_per_iteration": 2.7009174823760986
    },
    {
      "auxiliary_loss_clip": 0.01144076,
      "auxiliary_loss_mlp": 0.01022784,
      "balance_loss_clip": 1.04253578,
      "balance_loss_mlp": 1.01579571,
      "epoch": 0.8558888955690495,
      "flos": 17712723427200.0,
      "grad_norm": 1.8645667339833119,
      "language_loss": 0.72031069,
      "learning_rate": 2.1379772321314826e-07,
      "loss": 0.74197924,
      "num_input_tokens_seen": 154052985,
      "step": 7118,
      "time_per_iteration": 2.709346055984497
    },
    {
      "auxiliary_loss_clip": 0.01083568,
      "auxiliary_loss_mlp": 0.01027137,
      "balance_loss_clip": 1.03798127,
      "balance_loss_mlp": 1.02012157,
      "epoch": 0.8560091384596886,
      "flos": 19171702051200.0,
      "grad_norm": 2.168572080530982,
      "language_loss": 0.82049555,
      "learning_rate": 2.1344743369089802e-07,
      "loss": 0.84160262,
      "num_input_tokens_seen": 154068765,
      "step": 7119,
      "time_per_iteration": 2.8381388187408447
    },
    {
      "auxiliary_loss_clip": 0.01137011,
      "auxiliary_loss_mlp": 0.01027397,
      "balance_loss_clip": 1.04447985,
      "balance_loss_mlp": 1.02016377,
      "epoch": 0.8561293813503277,
      "flos": 23914855036800.0,
      "grad_norm": 1.7845548049348547,
      "language_loss": 0.82039738,
      "learning_rate": 2.130974151872522e-07,
      "loss": 0.84204143,
      "num_input_tokens_seen": 154089100,
      "step": 7120,
      "time_per_iteration": 2.689147472381592
    },
    {
      "auxiliary_loss_clip": 0.01124545,
      "auxiliary_loss_mlp": 0.01020391,
      "balance_loss_clip": 1.04212987,
      "balance_loss_mlp": 1.01365912,
      "epoch": 0.8562496242409667,
      "flos": 22529206028160.0,
      "grad_norm": 1.8358423196909646,
      "language_loss": 0.78501606,
      "learning_rate": 2.1274766775530773e-07,
      "loss": 0.80646539,
      "num_input_tokens_seen": 154108965,
      "step": 7121,
      "time_per_iteration": 2.8025741577148438
    },
    {
      "auxiliary_loss_clip": 0.01169745,
      "auxiliary_loss_mlp": 0.01025096,
      "balance_loss_clip": 1.04653645,
      "balance_loss_mlp": 1.01693296,
      "epoch": 0.8563698671316058,
      "flos": 14712745472640.0,
      "grad_norm": 3.0040132047351253,
      "language_loss": 0.79582512,
      "learning_rate": 2.1239819144812077e-07,
      "loss": 0.81777346,
      "num_input_tokens_seen": 154123425,
      "step": 7122,
      "time_per_iteration": 2.608166456222534
    },
    {
      "auxiliary_loss_clip": 0.0111509,
      "auxiliary_loss_mlp": 0.01028697,
      "balance_loss_clip": 1.03808188,
      "balance_loss_mlp": 1.02119255,
      "epoch": 0.856490110022245,
      "flos": 39167768211840.0,
      "grad_norm": 1.729891099056536,
      "language_loss": 0.69833261,
      "learning_rate": 2.1204898631870716e-07,
      "loss": 0.71977049,
      "num_input_tokens_seen": 154148315,
      "step": 7123,
      "time_per_iteration": 2.915086030960083
    },
    {
      "auxiliary_loss_clip": 0.0113926,
      "auxiliary_loss_mlp": 0.01021249,
      "balance_loss_clip": 1.04589844,
      "balance_loss_mlp": 1.01444852,
      "epoch": 0.856610352912884,
      "flos": 29059345658880.0,
      "grad_norm": 1.740348403264222,
      "language_loss": 0.7593357,
      "learning_rate": 2.1170005242004006e-07,
      "loss": 0.78094077,
      "num_input_tokens_seen": 154169665,
      "step": 7124,
      "time_per_iteration": 2.749047040939331
    },
    {
      "auxiliary_loss_clip": 0.01140945,
      "auxiliary_loss_mlp": 0.01021662,
      "balance_loss_clip": 1.04142749,
      "balance_loss_mlp": 1.01517463,
      "epoch": 0.8567305958035231,
      "flos": 23878333883520.0,
      "grad_norm": 1.777643815422862,
      "language_loss": 0.7831949,
      "learning_rate": 2.1135138980505384e-07,
      "loss": 0.80482101,
      "num_input_tokens_seen": 154190335,
      "step": 7125,
      "time_per_iteration": 3.67966365814209
    },
    {
      "auxiliary_loss_clip": 0.0113193,
      "auxiliary_loss_mlp": 0.01024986,
      "balance_loss_clip": 1.04251599,
      "balance_loss_mlp": 1.0184381,
      "epoch": 0.8568508386941622,
      "flos": 22200120599040.0,
      "grad_norm": 1.637236405821452,
      "language_loss": 0.722251,
      "learning_rate": 2.110029985266395e-07,
      "loss": 0.74382013,
      "num_input_tokens_seen": 154210040,
      "step": 7126,
      "time_per_iteration": 2.73753023147583
    },
    {
      "auxiliary_loss_clip": 0.01142892,
      "auxiliary_loss_mlp": 0.01023662,
      "balance_loss_clip": 1.04302812,
      "balance_loss_mlp": 1.01634896,
      "epoch": 0.8569710815848013,
      "flos": 17307507121920.0,
      "grad_norm": 3.954711404580165,
      "language_loss": 0.73940802,
      "learning_rate": 2.1065487863764787e-07,
      "loss": 0.76107359,
      "num_input_tokens_seen": 154228385,
      "step": 7127,
      "time_per_iteration": 2.7066574096679688
    },
    {
      "auxiliary_loss_clip": 0.01100546,
      "auxiliary_loss_mlp": 0.01023769,
      "balance_loss_clip": 1.0358429,
      "balance_loss_mlp": 1.01662242,
      "epoch": 0.8570913244754403,
      "flos": 23732285184000.0,
      "grad_norm": 1.5619274477599348,
      "language_loss": 0.858603,
      "learning_rate": 2.1030703019088846e-07,
      "loss": 0.87984622,
      "num_input_tokens_seen": 154249015,
      "step": 7128,
      "time_per_iteration": 3.8305954933166504
    },
    {
      "auxiliary_loss_clip": 0.01146414,
      "auxiliary_loss_mlp": 0.01019989,
      "balance_loss_clip": 1.04410338,
      "balance_loss_mlp": 1.01348591,
      "epoch": 0.8572115673660795,
      "flos": 20048748433920.0,
      "grad_norm": 1.9929363539919407,
      "language_loss": 0.7077477,
      "learning_rate": 2.099594532391291e-07,
      "loss": 0.72941172,
      "num_input_tokens_seen": 154267700,
      "step": 7129,
      "time_per_iteration": 2.6667532920837402
    },
    {
      "auxiliary_loss_clip": 0.01142899,
      "auxiliary_loss_mlp": 0.01025046,
      "balance_loss_clip": 1.04275405,
      "balance_loss_mlp": 1.01774466,
      "epoch": 0.8573318102567186,
      "flos": 27160389342720.0,
      "grad_norm": 1.5429759433785004,
      "language_loss": 0.78972203,
      "learning_rate": 2.0961214783509806e-07,
      "loss": 0.81140149,
      "num_input_tokens_seen": 154290580,
      "step": 7130,
      "time_per_iteration": 2.680532217025757
    },
    {
      "auxiliary_loss_clip": 0.01139297,
      "auxiliary_loss_mlp": 0.01023983,
      "balance_loss_clip": 1.04012167,
      "balance_loss_mlp": 1.01700318,
      "epoch": 0.8574520531473576,
      "flos": 24936585402240.0,
      "grad_norm": 1.8476976488686694,
      "language_loss": 0.74702221,
      "learning_rate": 2.0926511403148051e-07,
      "loss": 0.768655,
      "num_input_tokens_seen": 154309545,
      "step": 7131,
      "time_per_iteration": 2.7055540084838867
    },
    {
      "auxiliary_loss_clip": 0.01131429,
      "auxiliary_loss_mlp": 0.01020751,
      "balance_loss_clip": 1.04433978,
      "balance_loss_mlp": 1.01409006,
      "epoch": 0.8575722960379968,
      "flos": 18771154513920.0,
      "grad_norm": 1.8998857565928924,
      "language_loss": 0.7549572,
      "learning_rate": 2.0891835188092143e-07,
      "loss": 0.77647901,
      "num_input_tokens_seen": 154326545,
      "step": 7132,
      "time_per_iteration": 4.546796560287476
    },
    {
      "auxiliary_loss_clip": 0.01130517,
      "auxiliary_loss_mlp": 0.01028405,
      "balance_loss_clip": 1.04222476,
      "balance_loss_mlp": 1.02164006,
      "epoch": 0.8576925389286358,
      "flos": 22200300167040.0,
      "grad_norm": 1.9270937953793448,
      "language_loss": 0.81542766,
      "learning_rate": 2.0857186143602434e-07,
      "loss": 0.83701694,
      "num_input_tokens_seen": 154345190,
      "step": 7133,
      "time_per_iteration": 2.7012603282928467
    },
    {
      "auxiliary_loss_clip": 0.01115542,
      "auxiliary_loss_mlp": 0.01024615,
      "balance_loss_clip": 1.04092193,
      "balance_loss_mlp": 1.01736724,
      "epoch": 0.8578127818192749,
      "flos": 22894345733760.0,
      "grad_norm": 1.7233212797318922,
      "language_loss": 0.67682958,
      "learning_rate": 2.0822564274935094e-07,
      "loss": 0.69823116,
      "num_input_tokens_seen": 154364615,
      "step": 7134,
      "time_per_iteration": 2.6446781158447266
    },
    {
      "auxiliary_loss_clip": 0.01136986,
      "auxiliary_loss_mlp": 0.01028975,
      "balance_loss_clip": 1.04500258,
      "balance_loss_mlp": 1.02078485,
      "epoch": 0.8579330247099141,
      "flos": 34824839541120.0,
      "grad_norm": 2.055907725644711,
      "language_loss": 0.67079622,
      "learning_rate": 2.078796958734239e-07,
      "loss": 0.69245577,
      "num_input_tokens_seen": 154387335,
      "step": 7135,
      "time_per_iteration": 2.792418956756592
    },
    {
      "auxiliary_loss_clip": 0.01150448,
      "auxiliary_loss_mlp": 0.01023773,
      "balance_loss_clip": 1.0451895,
      "balance_loss_mlp": 1.01648366,
      "epoch": 0.8580532676005531,
      "flos": 19755681367680.0,
      "grad_norm": 2.0994298973306997,
      "language_loss": 0.75208074,
      "learning_rate": 2.0753402086072124e-07,
      "loss": 0.77382296,
      "num_input_tokens_seen": 154405965,
      "step": 7136,
      "time_per_iteration": 2.723586082458496
    },
    {
      "auxiliary_loss_clip": 0.0109364,
      "auxiliary_loss_mlp": 0.01028876,
      "balance_loss_clip": 1.03901255,
      "balance_loss_mlp": 1.02185142,
      "epoch": 0.8581735104911922,
      "flos": 22739318634240.0,
      "grad_norm": 7.691147728024715,
      "language_loss": 0.74857146,
      "learning_rate": 2.071886177636828e-07,
      "loss": 0.76979661,
      "num_input_tokens_seen": 154422750,
      "step": 7137,
      "time_per_iteration": 2.9094910621643066
    },
    {
      "auxiliary_loss_clip": 0.01149475,
      "auxiliary_loss_mlp": 0.01024923,
      "balance_loss_clip": 1.04564857,
      "balance_loss_mlp": 1.01760912,
      "epoch": 0.8582937533818313,
      "flos": 23149131880320.0,
      "grad_norm": 1.903500849230241,
      "language_loss": 0.82877076,
      "learning_rate": 2.0684348663470575e-07,
      "loss": 0.85051465,
      "num_input_tokens_seen": 154442930,
      "step": 7138,
      "time_per_iteration": 2.988851547241211
    },
    {
      "auxiliary_loss_clip": 0.01135026,
      "auxiliary_loss_mlp": 0.01022303,
      "balance_loss_clip": 1.03903675,
      "balance_loss_mlp": 1.01420236,
      "epoch": 0.8584139962724704,
      "flos": 19498668577920.0,
      "grad_norm": 1.9087255683714992,
      "language_loss": 0.6134125,
      "learning_rate": 2.0649862752614555e-07,
      "loss": 0.6349858,
      "num_input_tokens_seen": 154461640,
      "step": 7139,
      "time_per_iteration": 2.6334290504455566
    },
    {
      "auxiliary_loss_clip": 0.01044564,
      "auxiliary_loss_mlp": 0.01001559,
      "balance_loss_clip": 1.01009202,
      "balance_loss_mlp": 1.00039637,
      "epoch": 0.8585342391631094,
      "flos": 71276577788160.0,
      "grad_norm": 0.785123412092843,
      "language_loss": 0.56996667,
      "learning_rate": 2.0615404049031838e-07,
      "loss": 0.59042799,
      "num_input_tokens_seen": 154518610,
      "step": 7140,
      "time_per_iteration": 3.229763984680176
    },
    {
      "auxiliary_loss_clip": 0.011524,
      "auxiliary_loss_mlp": 0.010243,
      "balance_loss_clip": 1.045385,
      "balance_loss_mlp": 1.01652753,
      "epoch": 0.8586544820537486,
      "flos": 10815432929280.0,
      "grad_norm": 2.3789058260143565,
      "language_loss": 0.7822358,
      "learning_rate": 2.0580972557949616e-07,
      "loss": 0.80400276,
      "num_input_tokens_seen": 154533700,
      "step": 7141,
      "time_per_iteration": 2.572829484939575
    },
    {
      "auxiliary_loss_clip": 0.0105548,
      "auxiliary_loss_mlp": 0.01000308,
      "balance_loss_clip": 1.00970387,
      "balance_loss_mlp": 0.99925894,
      "epoch": 0.8587747249443877,
      "flos": 64811184422400.0,
      "grad_norm": 0.793529180126071,
      "language_loss": 0.54204392,
      "learning_rate": 2.054656828459125e-07,
      "loss": 0.5626018,
      "num_input_tokens_seen": 154597810,
      "step": 7142,
      "time_per_iteration": 3.251307249069214
    },
    {
      "auxiliary_loss_clip": 0.01106037,
      "auxiliary_loss_mlp": 0.01023177,
      "balance_loss_clip": 1.04059243,
      "balance_loss_mlp": 1.0159831,
      "epoch": 0.8588949678350267,
      "flos": 26834607964800.0,
      "grad_norm": 1.9312799700060816,
      "language_loss": 0.77271008,
      "learning_rate": 2.051219123417578e-07,
      "loss": 0.79400223,
      "num_input_tokens_seen": 154617870,
      "step": 7143,
      "time_per_iteration": 2.7867801189422607
    },
    {
      "auxiliary_loss_clip": 0.01167322,
      "auxiliary_loss_mlp": 0.01024897,
      "balance_loss_clip": 1.04582918,
      "balance_loss_mlp": 1.01688576,
      "epoch": 0.8590152107256659,
      "flos": 26104256726400.0,
      "grad_norm": 2.215683153906872,
      "language_loss": 0.60338986,
      "learning_rate": 2.0477841411918196e-07,
      "loss": 0.62531209,
      "num_input_tokens_seen": 154637395,
      "step": 7144,
      "time_per_iteration": 2.6101300716400146
    },
    {
      "auxiliary_loss_clip": 0.01145377,
      "auxiliary_loss_mlp": 0.01023639,
      "balance_loss_clip": 1.04185092,
      "balance_loss_mlp": 1.01686454,
      "epoch": 0.859135453616305,
      "flos": 26140885620480.0,
      "grad_norm": 1.9594809220030802,
      "language_loss": 0.74833429,
      "learning_rate": 2.0443518823029326e-07,
      "loss": 0.77002442,
      "num_input_tokens_seen": 154657935,
      "step": 7145,
      "time_per_iteration": 2.7533867359161377
    },
    {
      "auxiliary_loss_clip": 0.01115643,
      "auxiliary_loss_mlp": 0.01024488,
      "balance_loss_clip": 1.03999734,
      "balance_loss_mlp": 1.01758575,
      "epoch": 0.859255696506944,
      "flos": 12969319046400.0,
      "grad_norm": 2.2634185528288655,
      "language_loss": 0.76531649,
      "learning_rate": 2.0409223472715854e-07,
      "loss": 0.78671777,
      "num_input_tokens_seen": 154675080,
      "step": 7146,
      "time_per_iteration": 2.6953179836273193
    },
    {
      "auxiliary_loss_clip": 0.01121682,
      "auxiliary_loss_mlp": 0.00761505,
      "balance_loss_clip": 1.04234374,
      "balance_loss_mlp": 1.00034237,
      "epoch": 0.8593759393975832,
      "flos": 18475753063680.0,
      "grad_norm": 2.2651036618379576,
      "language_loss": 0.74923873,
      "learning_rate": 2.0374955366180434e-07,
      "loss": 0.7680707,
      "num_input_tokens_seen": 154692720,
      "step": 7147,
      "time_per_iteration": 2.6443030834198
    },
    {
      "auxiliary_loss_clip": 0.01128121,
      "auxiliary_loss_mlp": 0.01024694,
      "balance_loss_clip": 1.04130733,
      "balance_loss_mlp": 1.01771736,
      "epoch": 0.8594961822882222,
      "flos": 22200156512640.0,
      "grad_norm": 1.698857684184007,
      "language_loss": 0.72930765,
      "learning_rate": 2.034071450862147e-07,
      "loss": 0.7508359,
      "num_input_tokens_seen": 154710190,
      "step": 7148,
      "time_per_iteration": 2.682011842727661
    },
    {
      "auxiliary_loss_clip": 0.01141911,
      "auxiliary_loss_mlp": 0.01030984,
      "balance_loss_clip": 1.04224062,
      "balance_loss_mlp": 1.02324772,
      "epoch": 0.8596164251788613,
      "flos": 23294749616640.0,
      "grad_norm": 1.8538445190991428,
      "language_loss": 0.77062947,
      "learning_rate": 2.030650090523327e-07,
      "loss": 0.7923584,
      "num_input_tokens_seen": 154729380,
      "step": 7149,
      "time_per_iteration": 2.6947977542877197
    },
    {
      "auxiliary_loss_clip": 0.01117665,
      "auxiliary_loss_mlp": 0.01029392,
      "balance_loss_clip": 1.03898692,
      "balance_loss_mlp": 1.02183664,
      "epoch": 0.8597366680695004,
      "flos": 31649905416960.0,
      "grad_norm": 2.2391420338666252,
      "language_loss": 0.59360737,
      "learning_rate": 2.0272314561205995e-07,
      "loss": 0.61507797,
      "num_input_tokens_seen": 154749775,
      "step": 7150,
      "time_per_iteration": 2.7641103267669678
    },
    {
      "auxiliary_loss_clip": 0.0111533,
      "auxiliary_loss_mlp": 0.01022015,
      "balance_loss_clip": 1.03594851,
      "balance_loss_mlp": 1.01482987,
      "epoch": 0.8598569109601395,
      "flos": 21287738211840.0,
      "grad_norm": 1.882668454778394,
      "language_loss": 0.7275129,
      "learning_rate": 2.023815548172567e-07,
      "loss": 0.74888635,
      "num_input_tokens_seen": 154769845,
      "step": 7151,
      "time_per_iteration": 3.66994047164917
    },
    {
      "auxiliary_loss_clip": 0.01149508,
      "auxiliary_loss_mlp": 0.01022741,
      "balance_loss_clip": 1.04240417,
      "balance_loss_mlp": 1.01571095,
      "epoch": 0.8599771538507786,
      "flos": 25447809720960.0,
      "grad_norm": 1.9292665823031423,
      "language_loss": 0.66239846,
      "learning_rate": 2.0204023671974267e-07,
      "loss": 0.68412089,
      "num_input_tokens_seen": 154789230,
      "step": 7152,
      "time_per_iteration": 2.6786062717437744
    },
    {
      "auxiliary_loss_clip": 0.01146366,
      "auxiliary_loss_mlp": 0.01026267,
      "balance_loss_clip": 1.04216826,
      "balance_loss_mlp": 1.01951361,
      "epoch": 0.8600973967414177,
      "flos": 16723958768640.0,
      "grad_norm": 2.1090072728378266,
      "language_loss": 0.80818725,
      "learning_rate": 2.0169919137129532e-07,
      "loss": 0.8299135,
      "num_input_tokens_seen": 154807670,
      "step": 7153,
      "time_per_iteration": 2.5562543869018555
    },
    {
      "auxiliary_loss_clip": 0.01153562,
      "auxiliary_loss_mlp": 0.0102597,
      "balance_loss_clip": 1.04688716,
      "balance_loss_mlp": 1.01822102,
      "epoch": 0.8602176396320568,
      "flos": 25227928615680.0,
      "grad_norm": 2.1889175461436214,
      "language_loss": 0.7088182,
      "learning_rate": 2.013584188236508e-07,
      "loss": 0.73061347,
      "num_input_tokens_seen": 154825575,
      "step": 7154,
      "time_per_iteration": 3.629422903060913
    },
    {
      "auxiliary_loss_clip": 0.01167114,
      "auxiliary_loss_mlp": 0.01027963,
      "balance_loss_clip": 1.04594326,
      "balance_loss_mlp": 1.02078664,
      "epoch": 0.8603378825226958,
      "flos": 20412236113920.0,
      "grad_norm": 1.7833954588102878,
      "language_loss": 0.79483163,
      "learning_rate": 2.0101791912850396e-07,
      "loss": 0.81678236,
      "num_input_tokens_seen": 154845115,
      "step": 7155,
      "time_per_iteration": 2.5893263816833496
    },
    {
      "auxiliary_loss_clip": 0.01139264,
      "auxiliary_loss_mlp": 0.01028248,
      "balance_loss_clip": 1.04567194,
      "balance_loss_mlp": 1.02096176,
      "epoch": 0.8604581254133349,
      "flos": 34930201109760.0,
      "grad_norm": 2.0816786865580155,
      "language_loss": 0.64101493,
      "learning_rate": 2.006776923375082e-07,
      "loss": 0.66269004,
      "num_input_tokens_seen": 154866770,
      "step": 7156,
      "time_per_iteration": 2.758556604385376
    },
    {
      "auxiliary_loss_clip": 0.01163413,
      "auxiliary_loss_mlp": 0.01019268,
      "balance_loss_clip": 1.04401803,
      "balance_loss_mlp": 1.01216316,
      "epoch": 0.860578368303974,
      "flos": 22596538072320.0,
      "grad_norm": 1.7339618308165046,
      "language_loss": 0.71121514,
      "learning_rate": 2.003377385022764e-07,
      "loss": 0.733042,
      "num_input_tokens_seen": 154885595,
      "step": 7157,
      "time_per_iteration": 3.5030624866485596
    },
    {
      "auxiliary_loss_clip": 0.01137622,
      "auxiliary_loss_mlp": 0.01025614,
      "balance_loss_clip": 1.04236126,
      "balance_loss_mlp": 1.01883352,
      "epoch": 0.8606986111946131,
      "flos": 21324331192320.0,
      "grad_norm": 2.145581417479779,
      "language_loss": 0.77588087,
      "learning_rate": 1.9999805767437826e-07,
      "loss": 0.79751325,
      "num_input_tokens_seen": 154904485,
      "step": 7158,
      "time_per_iteration": 2.6859681606292725
    },
    {
      "auxiliary_loss_clip": 0.01129234,
      "auxiliary_loss_mlp": 0.01022119,
      "balance_loss_clip": 1.04004693,
      "balance_loss_mlp": 1.01522541,
      "epoch": 0.8608188540852522,
      "flos": 28877206769280.0,
      "grad_norm": 1.716529933886509,
      "language_loss": 0.71930081,
      "learning_rate": 1.9965864990534386e-07,
      "loss": 0.74081433,
      "num_input_tokens_seen": 154925010,
      "step": 7159,
      "time_per_iteration": 3.5790255069732666
    },
    {
      "auxiliary_loss_clip": 0.01116766,
      "auxiliary_loss_mlp": 0.01022545,
      "balance_loss_clip": 1.03792489,
      "balance_loss_mlp": 1.01536274,
      "epoch": 0.8609390969758913,
      "flos": 29716187713920.0,
      "grad_norm": 1.9911643598488806,
      "language_loss": 0.77608061,
      "learning_rate": 1.9931951524666092e-07,
      "loss": 0.79747367,
      "num_input_tokens_seen": 154946100,
      "step": 7160,
      "time_per_iteration": 2.75137996673584
    },
    {
      "auxiliary_loss_clip": 0.01155995,
      "auxiliary_loss_mlp": 0.00762244,
      "balance_loss_clip": 1.04518652,
      "balance_loss_mlp": 1.00033522,
      "epoch": 0.8610593398665304,
      "flos": 21249349551360.0,
      "grad_norm": 1.845537916430274,
      "language_loss": 0.81207287,
      "learning_rate": 1.9898065374977534e-07,
      "loss": 0.8312552,
      "num_input_tokens_seen": 154966305,
      "step": 7161,
      "time_per_iteration": 2.657897710800171
    },
    {
      "auxiliary_loss_clip": 0.01121344,
      "auxiliary_loss_mlp": 0.01017375,
      "balance_loss_clip": 1.03978992,
      "balance_loss_mlp": 1.01117277,
      "epoch": 0.8611795827571694,
      "flos": 14830102183680.0,
      "grad_norm": 2.084375488838697,
      "language_loss": 0.72797596,
      "learning_rate": 1.9864206546609342e-07,
      "loss": 0.74936318,
      "num_input_tokens_seen": 154985145,
      "step": 7162,
      "time_per_iteration": 2.6752803325653076
    },
    {
      "auxiliary_loss_clip": 0.0116406,
      "auxiliary_loss_mlp": 0.01022987,
      "balance_loss_clip": 1.04373062,
      "balance_loss_mlp": 1.01629877,
      "epoch": 0.8612998256478086,
      "flos": 24243258107520.0,
      "grad_norm": 1.831737609914979,
      "language_loss": 0.84219527,
      "learning_rate": 1.983037504469771e-07,
      "loss": 0.86406577,
      "num_input_tokens_seen": 155003855,
      "step": 7163,
      "time_per_iteration": 2.5494441986083984
    },
    {
      "auxiliary_loss_clip": 0.01151161,
      "auxiliary_loss_mlp": 0.01029651,
      "balance_loss_clip": 1.04430914,
      "balance_loss_mlp": 1.02259135,
      "epoch": 0.8614200685384477,
      "flos": 21252653602560.0,
      "grad_norm": 1.832595670008331,
      "language_loss": 0.66648054,
      "learning_rate": 1.9796570874374984e-07,
      "loss": 0.68828863,
      "num_input_tokens_seen": 155023960,
      "step": 7164,
      "time_per_iteration": 2.5694549083709717
    },
    {
      "auxiliary_loss_clip": 0.01140723,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 1.0433042,
      "balance_loss_mlp": 1.02004445,
      "epoch": 0.8615403114290867,
      "flos": 20007738080640.0,
      "grad_norm": 1.6617486799965422,
      "language_loss": 0.77754819,
      "learning_rate": 1.976279404076917e-07,
      "loss": 0.79922551,
      "num_input_tokens_seen": 155043360,
      "step": 7165,
      "time_per_iteration": 2.5954620838165283
    },
    {
      "auxiliary_loss_clip": 0.01122546,
      "auxiliary_loss_mlp": 0.01026582,
      "balance_loss_clip": 1.04318786,
      "balance_loss_mlp": 1.01891696,
      "epoch": 0.8616605543197259,
      "flos": 29789373674880.0,
      "grad_norm": 1.856242811945421,
      "language_loss": 0.76453954,
      "learning_rate": 1.9729044549004193e-07,
      "loss": 0.78603083,
      "num_input_tokens_seen": 155064745,
      "step": 7166,
      "time_per_iteration": 2.6513164043426514
    },
    {
      "auxiliary_loss_clip": 0.01149459,
      "auxiliary_loss_mlp": 0.01025069,
      "balance_loss_clip": 1.043733,
      "balance_loss_mlp": 1.01859629,
      "epoch": 0.8617807972103649,
      "flos": 28911609020160.0,
      "grad_norm": 1.7675755679526302,
      "language_loss": 0.70370877,
      "learning_rate": 1.9695322404199822e-07,
      "loss": 0.72545397,
      "num_input_tokens_seen": 155086790,
      "step": 7167,
      "time_per_iteration": 2.6939780712127686
    },
    {
      "auxiliary_loss_clip": 0.01134043,
      "auxiliary_loss_mlp": 0.010261,
      "balance_loss_clip": 1.04174268,
      "balance_loss_mlp": 1.01837277,
      "epoch": 0.861901040101004,
      "flos": 27673804391040.0,
      "grad_norm": 1.9456807754740857,
      "language_loss": 0.82074732,
      "learning_rate": 1.9661627611471654e-07,
      "loss": 0.84234869,
      "num_input_tokens_seen": 155106585,
      "step": 7168,
      "time_per_iteration": 2.7579081058502197
    },
    {
      "auxiliary_loss_clip": 0.01144509,
      "auxiliary_loss_mlp": 0.0102288,
      "balance_loss_clip": 1.04288769,
      "balance_loss_mlp": 1.01531577,
      "epoch": 0.8620212829916432,
      "flos": 49748056755840.0,
      "grad_norm": 16.158414512943875,
      "language_loss": 0.70402759,
      "learning_rate": 1.9627960175931246e-07,
      "loss": 0.72570145,
      "num_input_tokens_seen": 155131285,
      "step": 7169,
      "time_per_iteration": 2.940899610519409
    },
    {
      "auxiliary_loss_clip": 0.01152602,
      "auxiliary_loss_mlp": 0.01027342,
      "balance_loss_clip": 1.04576457,
      "balance_loss_mlp": 1.02081847,
      "epoch": 0.8621415258822822,
      "flos": 21138672769920.0,
      "grad_norm": 2.6496488080516363,
      "language_loss": 0.74570143,
      "learning_rate": 1.9594320102685847e-07,
      "loss": 0.76750082,
      "num_input_tokens_seen": 155150555,
      "step": 7170,
      "time_per_iteration": 2.6414034366607666
    },
    {
      "auxiliary_loss_clip": 0.01127143,
      "auxiliary_loss_mlp": 0.00762062,
      "balance_loss_clip": 1.04004812,
      "balance_loss_mlp": 1.00034177,
      "epoch": 0.8622617687729213,
      "flos": 21689039934720.0,
      "grad_norm": 2.6527758173417766,
      "language_loss": 0.64341938,
      "learning_rate": 1.956070739683864e-07,
      "loss": 0.66231143,
      "num_input_tokens_seen": 155169890,
      "step": 7171,
      "time_per_iteration": 2.6548707485198975
    },
    {
      "auxiliary_loss_clip": 0.01106596,
      "auxiliary_loss_mlp": 0.01019376,
      "balance_loss_clip": 1.03721023,
      "balance_loss_mlp": 1.0127598,
      "epoch": 0.8623820116635604,
      "flos": 26250592734720.0,
      "grad_norm": 1.5085956852081983,
      "language_loss": 0.74213332,
      "learning_rate": 1.9527122063488678e-07,
      "loss": 0.76339298,
      "num_input_tokens_seen": 155191005,
      "step": 7172,
      "time_per_iteration": 2.7063567638397217
    },
    {
      "auxiliary_loss_clip": 0.01133647,
      "auxiliary_loss_mlp": 0.01022562,
      "balance_loss_clip": 1.0385344,
      "balance_loss_mlp": 1.01534057,
      "epoch": 0.8625022545541995,
      "flos": 19647554451840.0,
      "grad_norm": 1.7140280899912888,
      "language_loss": 0.80425847,
      "learning_rate": 1.9493564107730755e-07,
      "loss": 0.82582057,
      "num_input_tokens_seen": 155211005,
      "step": 7173,
      "time_per_iteration": 2.6357126235961914
    },
    {
      "auxiliary_loss_clip": 0.01131145,
      "auxiliary_loss_mlp": 0.01023876,
      "balance_loss_clip": 1.03975296,
      "balance_loss_mlp": 1.01717353,
      "epoch": 0.8626224974448385,
      "flos": 21908382336000.0,
      "grad_norm": 1.9049073428580094,
      "language_loss": 0.60879517,
      "learning_rate": 1.9460033534655684e-07,
      "loss": 0.63034534,
      "num_input_tokens_seen": 155230365,
      "step": 7174,
      "time_per_iteration": 2.6813619136810303
    },
    {
      "auxiliary_loss_clip": 0.01130539,
      "auxiliary_loss_mlp": 0.01025889,
      "balance_loss_clip": 1.03774881,
      "balance_loss_mlp": 1.0189271,
      "epoch": 0.8627427403354777,
      "flos": 23331198942720.0,
      "grad_norm": 1.635754109214876,
      "language_loss": 0.84502518,
      "learning_rate": 1.9426530349349978e-07,
      "loss": 0.86658955,
      "num_input_tokens_seen": 155250815,
      "step": 7175,
      "time_per_iteration": 2.686558485031128
    },
    {
      "auxiliary_loss_clip": 0.01148936,
      "auxiliary_loss_mlp": 0.00761872,
      "balance_loss_clip": 1.042647,
      "balance_loss_mlp": 1.00031614,
      "epoch": 0.8628629832261168,
      "flos": 16362877299840.0,
      "grad_norm": 1.855064150178447,
      "language_loss": 0.64787221,
      "learning_rate": 1.9393054556896038e-07,
      "loss": 0.66698027,
      "num_input_tokens_seen": 155268515,
      "step": 7176,
      "time_per_iteration": 2.6345980167388916
    },
    {
      "auxiliary_loss_clip": 0.01121205,
      "auxiliary_loss_mlp": 0.01025425,
      "balance_loss_clip": 1.04006135,
      "balance_loss_mlp": 1.01824558,
      "epoch": 0.8629832261167558,
      "flos": 28103941756800.0,
      "grad_norm": 2.336250220930642,
      "language_loss": 0.68790728,
      "learning_rate": 1.9359606162372133e-07,
      "loss": 0.70937353,
      "num_input_tokens_seen": 155290120,
      "step": 7177,
      "time_per_iteration": 3.771299123764038
    },
    {
      "auxiliary_loss_clip": 0.01165493,
      "auxiliary_loss_mlp": 0.01020642,
      "balance_loss_clip": 1.0470736,
      "balance_loss_mlp": 1.01319122,
      "epoch": 0.863103469007395,
      "flos": 20230061310720.0,
      "grad_norm": 1.7739054333130502,
      "language_loss": 0.70558107,
      "learning_rate": 1.9326185170852293e-07,
      "loss": 0.72744244,
      "num_input_tokens_seen": 155309085,
      "step": 7178,
      "time_per_iteration": 2.6541996002197266
    },
    {
      "auxiliary_loss_clip": 0.01150113,
      "auxiliary_loss_mlp": 0.01026302,
      "balance_loss_clip": 1.04409266,
      "balance_loss_mlp": 1.0191052,
      "epoch": 0.863223711898034,
      "flos": 24498547044480.0,
      "grad_norm": 1.9369858039736345,
      "language_loss": 0.72229439,
      "learning_rate": 1.9292791587406598e-07,
      "loss": 0.74405855,
      "num_input_tokens_seen": 155327945,
      "step": 7179,
      "time_per_iteration": 2.672797203063965
    },
    {
      "auxiliary_loss_clip": 0.01149387,
      "auxiliary_loss_mlp": 0.00762354,
      "balance_loss_clip": 1.04306555,
      "balance_loss_mlp": 1.00031364,
      "epoch": 0.8633439547886731,
      "flos": 17675376261120.0,
      "grad_norm": 2.09837599603976,
      "language_loss": 0.86848027,
      "learning_rate": 1.9259425417100661e-07,
      "loss": 0.88759762,
      "num_input_tokens_seen": 155344060,
      "step": 7180,
      "time_per_iteration": 3.5816879272460938
    },
    {
      "auxiliary_loss_clip": 0.01088739,
      "auxiliary_loss_mlp": 0.01026757,
      "balance_loss_clip": 1.03139722,
      "balance_loss_mlp": 1.01927078,
      "epoch": 0.8634641976793123,
      "flos": 12895055677440.0,
      "grad_norm": 1.9428146577784717,
      "language_loss": 0.74645793,
      "learning_rate": 1.9226086664996234e-07,
      "loss": 0.76761281,
      "num_input_tokens_seen": 155362305,
      "step": 7181,
      "time_per_iteration": 2.718672752380371
    },
    {
      "auxiliary_loss_clip": 0.011415,
      "auxiliary_loss_mlp": 0.01018964,
      "balance_loss_clip": 1.04517245,
      "balance_loss_mlp": 1.01209772,
      "epoch": 0.8635844405699513,
      "flos": 23878980328320.0,
      "grad_norm": 3.613124617743116,
      "language_loss": 0.7414974,
      "learning_rate": 1.9192775336150712e-07,
      "loss": 0.76310205,
      "num_input_tokens_seen": 155382605,
      "step": 7182,
      "time_per_iteration": 2.690648078918457
    },
    {
      "auxiliary_loss_clip": 0.0104943,
      "auxiliary_loss_mlp": 0.01002038,
      "balance_loss_clip": 1.00822604,
      "balance_loss_mlp": 1.00095963,
      "epoch": 0.8637046834605904,
      "flos": 60453387521280.0,
      "grad_norm": 0.7622576332722774,
      "language_loss": 0.56289518,
      "learning_rate": 1.915949143561739e-07,
      "loss": 0.58340991,
      "num_input_tokens_seen": 155437280,
      "step": 7183,
      "time_per_iteration": 4.05260443687439
    },
    {
      "auxiliary_loss_clip": 0.0115187,
      "auxiliary_loss_mlp": 0.01030617,
      "balance_loss_clip": 1.0453651,
      "balance_loss_mlp": 1.02299023,
      "epoch": 0.8638249263512295,
      "flos": 20558751690240.0,
      "grad_norm": 1.7627645974046182,
      "language_loss": 0.7767421,
      "learning_rate": 1.9126234968445498e-07,
      "loss": 0.79856694,
      "num_input_tokens_seen": 155456970,
      "step": 7184,
      "time_per_iteration": 3.543426513671875
    },
    {
      "auxiliary_loss_clip": 0.01167325,
      "auxiliary_loss_mlp": 0.01025116,
      "balance_loss_clip": 1.04744649,
      "balance_loss_mlp": 1.01844347,
      "epoch": 0.8639451692418686,
      "flos": 26615768353920.0,
      "grad_norm": 1.5606417652399565,
      "language_loss": 0.67630446,
      "learning_rate": 1.9093005939679884e-07,
      "loss": 0.69822884,
      "num_input_tokens_seen": 155478925,
      "step": 7185,
      "time_per_iteration": 2.6614227294921875
    },
    {
      "auxiliary_loss_clip": 0.01151543,
      "auxiliary_loss_mlp": 0.01027733,
      "balance_loss_clip": 1.04621899,
      "balance_loss_mlp": 1.0204103,
      "epoch": 0.8640654121325076,
      "flos": 15122450977920.0,
      "grad_norm": 1.8831870948660911,
      "language_loss": 0.76740342,
      "learning_rate": 1.9059804354361452e-07,
      "loss": 0.78919613,
      "num_input_tokens_seen": 155496700,
      "step": 7186,
      "time_per_iteration": 2.5526468753814697
    },
    {
      "auxiliary_loss_clip": 0.01129767,
      "auxiliary_loss_mlp": 0.01022783,
      "balance_loss_clip": 1.03927016,
      "balance_loss_mlp": 1.0160979,
      "epoch": 0.8641856550231467,
      "flos": 31869068250240.0,
      "grad_norm": 2.3492456012954324,
      "language_loss": 0.70625961,
      "learning_rate": 1.902663021752684e-07,
      "loss": 0.72778517,
      "num_input_tokens_seen": 155518130,
      "step": 7187,
      "time_per_iteration": 2.7872605323791504
    },
    {
      "auxiliary_loss_clip": 0.01167846,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.04684472,
      "balance_loss_mlp": 1.01898789,
      "epoch": 0.8643058979137859,
      "flos": 14976545932800.0,
      "grad_norm": 2.2005799720253862,
      "language_loss": 0.82768297,
      "learning_rate": 1.8993483534208556e-07,
      "loss": 0.8496263,
      "num_input_tokens_seen": 155537040,
      "step": 7188,
      "time_per_iteration": 2.691890001296997
    },
    {
      "auxiliary_loss_clip": 0.01130549,
      "auxiliary_loss_mlp": 0.01021462,
      "balance_loss_clip": 1.04274797,
      "balance_loss_mlp": 1.01439333,
      "epoch": 0.8644261408044249,
      "flos": 13115726881920.0,
      "grad_norm": 2.495818442137677,
      "language_loss": 0.75188887,
      "learning_rate": 1.8960364309434884e-07,
      "loss": 0.77340901,
      "num_input_tokens_seen": 155554535,
      "step": 7189,
      "time_per_iteration": 2.6328091621398926
    },
    {
      "auxiliary_loss_clip": 0.01091966,
      "auxiliary_loss_mlp": 0.0076175,
      "balance_loss_clip": 1.03875136,
      "balance_loss_mlp": 1.00033545,
      "epoch": 0.864546383695064,
      "flos": 20850920916480.0,
      "grad_norm": 1.7993057065347131,
      "language_loss": 0.77976131,
      "learning_rate": 1.8927272548229967e-07,
      "loss": 0.79829854,
      "num_input_tokens_seen": 155574225,
      "step": 7190,
      "time_per_iteration": 2.794088125228882
    },
    {
      "auxiliary_loss_clip": 0.01107659,
      "auxiliary_loss_mlp": 0.01029736,
      "balance_loss_clip": 1.0389297,
      "balance_loss_mlp": 1.02262259,
      "epoch": 0.8646666265857031,
      "flos": 21324582587520.0,
      "grad_norm": 1.7727385853435105,
      "language_loss": 0.83027005,
      "learning_rate": 1.8894208255613876e-07,
      "loss": 0.85164404,
      "num_input_tokens_seen": 155593540,
      "step": 7191,
      "time_per_iteration": 2.7616939544677734
    },
    {
      "auxiliary_loss_clip": 0.01164931,
      "auxiliary_loss_mlp": 0.01023697,
      "balance_loss_clip": 1.04640579,
      "balance_loss_mlp": 1.01684237,
      "epoch": 0.8647868694763422,
      "flos": 19750833031680.0,
      "grad_norm": 1.858189988932058,
      "language_loss": 0.7791276,
      "learning_rate": 1.8861171436602397e-07,
      "loss": 0.80101389,
      "num_input_tokens_seen": 155610655,
      "step": 7192,
      "time_per_iteration": 2.594428777694702
    },
    {
      "auxiliary_loss_clip": 0.01153888,
      "auxiliary_loss_mlp": 0.01024697,
      "balance_loss_clip": 1.04623413,
      "balance_loss_mlp": 1.01759553,
      "epoch": 0.8649071123669813,
      "flos": 26176760328960.0,
      "grad_norm": 2.4900749878375836,
      "language_loss": 0.80438733,
      "learning_rate": 1.882816209620719e-07,
      "loss": 0.82617325,
      "num_input_tokens_seen": 155627365,
      "step": 7193,
      "time_per_iteration": 2.7147698402404785
    },
    {
      "auxiliary_loss_clip": 0.01141652,
      "auxiliary_loss_mlp": 0.01028338,
      "balance_loss_clip": 1.04749024,
      "balance_loss_mlp": 1.02108717,
      "epoch": 0.8650273552576204,
      "flos": 20302888135680.0,
      "grad_norm": 1.901977928179352,
      "language_loss": 0.76730353,
      "learning_rate": 1.8795180239435738e-07,
      "loss": 0.78900349,
      "num_input_tokens_seen": 155646220,
      "step": 7194,
      "time_per_iteration": 2.7062807083129883
    },
    {
      "auxiliary_loss_clip": 0.0114419,
      "auxiliary_loss_mlp": 0.01026291,
      "balance_loss_clip": 1.04384005,
      "balance_loss_mlp": 1.01916862,
      "epoch": 0.8651475981482595,
      "flos": 23951088881280.0,
      "grad_norm": 3.9397989089396757,
      "language_loss": 0.7558893,
      "learning_rate": 1.8762225871291348e-07,
      "loss": 0.77759415,
      "num_input_tokens_seen": 155662095,
      "step": 7195,
      "time_per_iteration": 2.7234749794006348
    },
    {
      "auxiliary_loss_clip": 0.01164864,
      "auxiliary_loss_mlp": 0.00762002,
      "balance_loss_clip": 1.04532468,
      "balance_loss_mlp": 1.00028634,
      "epoch": 0.8652678410388985,
      "flos": 21684622561920.0,
      "grad_norm": 1.8001298630211915,
      "language_loss": 0.8074317,
      "learning_rate": 1.8729298996773201e-07,
      "loss": 0.82670027,
      "num_input_tokens_seen": 155680845,
      "step": 7196,
      "time_per_iteration": 2.663961410522461
    },
    {
      "auxiliary_loss_clip": 0.01049082,
      "auxiliary_loss_mlp": 0.01001778,
      "balance_loss_clip": 1.00929379,
      "balance_loss_mlp": 1.0007056,
      "epoch": 0.8653880839295377,
      "flos": 65224660855680.0,
      "grad_norm": 0.830508887732276,
      "language_loss": 0.60882175,
      "learning_rate": 1.8696399620876301e-07,
      "loss": 0.6293304,
      "num_input_tokens_seen": 155737875,
      "step": 7197,
      "time_per_iteration": 3.157330274581909
    },
    {
      "auxiliary_loss_clip": 0.0111843,
      "auxiliary_loss_mlp": 0.01026771,
      "balance_loss_clip": 1.03613019,
      "balance_loss_mlp": 1.01905274,
      "epoch": 0.8655083268201768,
      "flos": 17749172753280.0,
      "grad_norm": 2.6192545803710505,
      "language_loss": 0.78811312,
      "learning_rate": 1.866352774859141e-07,
      "loss": 0.80956513,
      "num_input_tokens_seen": 155753100,
      "step": 7198,
      "time_per_iteration": 2.7168991565704346
    },
    {
      "auxiliary_loss_clip": 0.0112618,
      "auxiliary_loss_mlp": 0.01021117,
      "balance_loss_clip": 1.03973413,
      "balance_loss_mlp": 1.01481056,
      "epoch": 0.8656285697108158,
      "flos": 20703974376960.0,
      "grad_norm": 2.195829167103854,
      "language_loss": 0.69223666,
      "learning_rate": 1.8630683384905188e-07,
      "loss": 0.71370959,
      "num_input_tokens_seen": 155772430,
      "step": 7199,
      "time_per_iteration": 2.795522689819336
    },
    {
      "auxiliary_loss_clip": 0.01165234,
      "auxiliary_loss_mlp": 0.00762218,
      "balance_loss_clip": 1.04666197,
      "balance_loss_mlp": 1.00031996,
      "epoch": 0.865748812601455,
      "flos": 18653833716480.0,
      "grad_norm": 1.995233305081438,
      "language_loss": 0.88604444,
      "learning_rate": 1.8597866534800045e-07,
      "loss": 0.90531898,
      "num_input_tokens_seen": 155787545,
      "step": 7200,
      "time_per_iteration": 2.6190578937530518
    },
    {
      "auxiliary_loss_clip": 0.01153167,
      "auxiliary_loss_mlp": 0.00762255,
      "balance_loss_clip": 1.04418015,
      "balance_loss_mlp": 1.0003581,
      "epoch": 0.865869055492094,
      "flos": 70652554807680.0,
      "grad_norm": 2.1468728449959924,
      "language_loss": 0.74912989,
      "learning_rate": 1.8565077203254398e-07,
      "loss": 0.76828414,
      "num_input_tokens_seen": 155813005,
      "step": 7201,
      "time_per_iteration": 3.031463146209717
    },
    {
      "auxiliary_loss_clip": 0.0112532,
      "auxiliary_loss_mlp": 0.01023572,
      "balance_loss_clip": 1.04522526,
      "balance_loss_mlp": 1.01571035,
      "epoch": 0.8659892983827331,
      "flos": 17383961220480.0,
      "grad_norm": 3.655888372550851,
      "language_loss": 0.72210425,
      "learning_rate": 1.8532315395242203e-07,
      "loss": 0.74359322,
      "num_input_tokens_seen": 155829455,
      "step": 7202,
      "time_per_iteration": 2.664135456085205
    },
    {
      "auxiliary_loss_clip": 0.01127116,
      "auxiliary_loss_mlp": 0.01022045,
      "balance_loss_clip": 1.0418725,
      "balance_loss_mlp": 1.01562858,
      "epoch": 0.8661095412733723,
      "flos": 17895221452800.0,
      "grad_norm": 2.0166007637412906,
      "language_loss": 0.72529364,
      "learning_rate": 1.849958111573353e-07,
      "loss": 0.74678528,
      "num_input_tokens_seen": 155848060,
      "step": 7203,
      "time_per_iteration": 3.616074800491333
    },
    {
      "auxiliary_loss_clip": 0.01162221,
      "auxiliary_loss_mlp": 0.01027761,
      "balance_loss_clip": 1.04477561,
      "balance_loss_mlp": 1.02032518,
      "epoch": 0.8662297841640113,
      "flos": 18224163227520.0,
      "grad_norm": 1.931478183790679,
      "language_loss": 0.64171195,
      "learning_rate": 1.8466874369694074e-07,
      "loss": 0.66361177,
      "num_input_tokens_seen": 155865755,
      "step": 7204,
      "time_per_iteration": 2.596609592437744
    },
    {
      "auxiliary_loss_clip": 0.01122692,
      "auxiliary_loss_mlp": 0.01022803,
      "balance_loss_clip": 1.03742886,
      "balance_loss_mlp": 1.01584756,
      "epoch": 0.8663500270546504,
      "flos": 16362159027840.0,
      "grad_norm": 4.022737778391404,
      "language_loss": 0.6999073,
      "learning_rate": 1.843419516208542e-07,
      "loss": 0.72136235,
      "num_input_tokens_seen": 155882680,
      "step": 7205,
      "time_per_iteration": 2.7297699451446533
    },
    {
      "auxiliary_loss_clip": 0.01154197,
      "auxiliary_loss_mlp": 0.01024433,
      "balance_loss_clip": 1.04617679,
      "balance_loss_mlp": 1.01660061,
      "epoch": 0.8664702699452895,
      "flos": 17894431353600.0,
      "grad_norm": 2.1236357310180662,
      "language_loss": 0.79577732,
      "learning_rate": 1.8401543497865047e-07,
      "loss": 0.81756359,
      "num_input_tokens_seen": 155900680,
      "step": 7206,
      "time_per_iteration": 3.646923065185547
    },
    {
      "auxiliary_loss_clip": 0.01152389,
      "auxiliary_loss_mlp": 0.00762054,
      "balance_loss_clip": 1.04297161,
      "balance_loss_mlp": 1.00031734,
      "epoch": 0.8665905128359286,
      "flos": 30736373794560.0,
      "grad_norm": 2.6056777347025983,
      "language_loss": 0.64004844,
      "learning_rate": 1.836891938198608e-07,
      "loss": 0.65919286,
      "num_input_tokens_seen": 155921105,
      "step": 7207,
      "time_per_iteration": 2.723679304122925
    },
    {
      "auxiliary_loss_clip": 0.01136782,
      "auxiliary_loss_mlp": 0.01022211,
      "balance_loss_clip": 1.04313457,
      "balance_loss_mlp": 1.01520395,
      "epoch": 0.8667107557265676,
      "flos": 18656419495680.0,
      "grad_norm": 2.2396151388134906,
      "language_loss": 0.7114675,
      "learning_rate": 1.8336322819397677e-07,
      "loss": 0.7330575,
      "num_input_tokens_seen": 155938640,
      "step": 7208,
      "time_per_iteration": 2.6658427715301514
    },
    {
      "auxiliary_loss_clip": 0.01125988,
      "auxiliary_loss_mlp": 0.01025288,
      "balance_loss_clip": 1.03821874,
      "balance_loss_mlp": 1.01765847,
      "epoch": 0.8668309986172068,
      "flos": 20083725302400.0,
      "grad_norm": 1.7757728704907143,
      "language_loss": 0.62684739,
      "learning_rate": 1.8303753815044654e-07,
      "loss": 0.64836007,
      "num_input_tokens_seen": 155957945,
      "step": 7209,
      "time_per_iteration": 3.5742831230163574
    },
    {
      "auxiliary_loss_clip": 0.01148961,
      "auxiliary_loss_mlp": 0.0102705,
      "balance_loss_clip": 1.04438746,
      "balance_loss_mlp": 1.01922143,
      "epoch": 0.8669512415078459,
      "flos": 21615099788160.0,
      "grad_norm": 2.3340052585193938,
      "language_loss": 0.70265317,
      "learning_rate": 1.827121237386773e-07,
      "loss": 0.72441328,
      "num_input_tokens_seen": 155975390,
      "step": 7210,
      "time_per_iteration": 3.5693185329437256
    },
    {
      "auxiliary_loss_clip": 0.01138295,
      "auxiliary_loss_mlp": 0.01022976,
      "balance_loss_clip": 1.04270625,
      "balance_loss_mlp": 1.0154537,
      "epoch": 0.8670714843984849,
      "flos": 17703601372800.0,
      "grad_norm": 2.872965685661625,
      "language_loss": 0.7505762,
      "learning_rate": 1.8238698500803374e-07,
      "loss": 0.7721889,
      "num_input_tokens_seen": 155988155,
      "step": 7211,
      "time_per_iteration": 2.6012954711914062
    },
    {
      "auxiliary_loss_clip": 0.01056229,
      "auxiliary_loss_mlp": 0.0100083,
      "balance_loss_clip": 1.01030433,
      "balance_loss_mlp": 0.99973947,
      "epoch": 0.8671917272891241,
      "flos": 60705483125760.0,
      "grad_norm": 0.7205851947943182,
      "language_loss": 0.56258273,
      "learning_rate": 1.820621220078391e-07,
      "loss": 0.58315331,
      "num_input_tokens_seen": 156052065,
      "step": 7212,
      "time_per_iteration": 3.2570459842681885
    },
    {
      "auxiliary_loss_clip": 0.01166263,
      "auxiliary_loss_mlp": 0.0102452,
      "balance_loss_clip": 1.04535317,
      "balance_loss_mlp": 1.01686716,
      "epoch": 0.8673119701797631,
      "flos": 20451881750400.0,
      "grad_norm": 1.6422282465976248,
      "language_loss": 0.67696047,
      "learning_rate": 1.8173753478737553e-07,
      "loss": 0.69886827,
      "num_input_tokens_seen": 156072500,
      "step": 7213,
      "time_per_iteration": 2.6624302864074707
    },
    {
      "auxiliary_loss_clip": 0.01167599,
      "auxiliary_loss_mlp": 0.01023269,
      "balance_loss_clip": 1.04588246,
      "balance_loss_mlp": 1.01527572,
      "epoch": 0.8674322130704022,
      "flos": 19647410797440.0,
      "grad_norm": 2.2210793817714927,
      "language_loss": 0.79784238,
      "learning_rate": 1.8141322339588205e-07,
      "loss": 0.81975108,
      "num_input_tokens_seen": 156089840,
      "step": 7214,
      "time_per_iteration": 2.553856372833252
    },
    {
      "auxiliary_loss_clip": 0.01164467,
      "auxiliary_loss_mlp": 0.01024759,
      "balance_loss_clip": 1.04610479,
      "balance_loss_mlp": 1.01754737,
      "epoch": 0.8675524559610414,
      "flos": 26025001367040.0,
      "grad_norm": 2.240022714926397,
      "language_loss": 0.70142567,
      "learning_rate": 1.810891878825569e-07,
      "loss": 0.72331792,
      "num_input_tokens_seen": 156109815,
      "step": 7215,
      "time_per_iteration": 2.6321849822998047
    },
    {
      "auxiliary_loss_clip": 0.01137748,
      "auxiliary_loss_mlp": 0.01029971,
      "balance_loss_clip": 1.04171181,
      "balance_loss_mlp": 1.02256823,
      "epoch": 0.8676726988516804,
      "flos": 15049444584960.0,
      "grad_norm": 2.002437494060461,
      "language_loss": 0.71926832,
      "learning_rate": 1.8076542829655561e-07,
      "loss": 0.74094546,
      "num_input_tokens_seen": 156128620,
      "step": 7216,
      "time_per_iteration": 2.6315650939941406
    },
    {
      "auxiliary_loss_clip": 0.01138441,
      "auxiliary_loss_mlp": 0.01024903,
      "balance_loss_clip": 1.04438508,
      "balance_loss_mlp": 1.01757753,
      "epoch": 0.8677929417423195,
      "flos": 16288111140480.0,
      "grad_norm": 2.132253532979228,
      "language_loss": 0.79346716,
      "learning_rate": 1.8044194468699203e-07,
      "loss": 0.81510061,
      "num_input_tokens_seen": 156145930,
      "step": 7217,
      "time_per_iteration": 2.6922640800476074
    },
    {
      "auxiliary_loss_clip": 0.01136824,
      "auxiliary_loss_mlp": 0.01022512,
      "balance_loss_clip": 1.04491985,
      "balance_loss_mlp": 1.01522875,
      "epoch": 0.8679131846329585,
      "flos": 18844160906880.0,
      "grad_norm": 2.8580772045960576,
      "language_loss": 0.75840545,
      "learning_rate": 1.8011873710293912e-07,
      "loss": 0.77999878,
      "num_input_tokens_seen": 156164435,
      "step": 7218,
      "time_per_iteration": 2.6042444705963135
    },
    {
      "auxiliary_loss_clip": 0.01149284,
      "auxiliary_loss_mlp": 0.01029951,
      "balance_loss_clip": 1.04486108,
      "balance_loss_mlp": 1.02225924,
      "epoch": 0.8680334275235977,
      "flos": 33620718890880.0,
      "grad_norm": 3.8647837018938676,
      "language_loss": 0.6914556,
      "learning_rate": 1.7979580559342677e-07,
      "loss": 0.71324795,
      "num_input_tokens_seen": 156185165,
      "step": 7219,
      "time_per_iteration": 2.7507784366607666
    },
    {
      "auxiliary_loss_clip": 0.01135776,
      "auxiliary_loss_mlp": 0.01025354,
      "balance_loss_clip": 1.04287779,
      "balance_loss_mlp": 1.01808548,
      "epoch": 0.8681536704142367,
      "flos": 24681152810880.0,
      "grad_norm": 1.6907430229472833,
      "language_loss": 0.66703421,
      "learning_rate": 1.7947315020744358e-07,
      "loss": 0.68864548,
      "num_input_tokens_seen": 156206260,
      "step": 7220,
      "time_per_iteration": 2.720121383666992
    },
    {
      "auxiliary_loss_clip": 0.0113593,
      "auxiliary_loss_mlp": 0.01021411,
      "balance_loss_clip": 1.04156065,
      "balance_loss_mlp": 1.01510811,
      "epoch": 0.8682739133048758,
      "flos": 20011042131840.0,
      "grad_norm": 2.212574643950438,
      "language_loss": 0.80232763,
      "learning_rate": 1.7915077099393594e-07,
      "loss": 0.82390106,
      "num_input_tokens_seen": 156222860,
      "step": 7221,
      "time_per_iteration": 2.745957136154175
    },
    {
      "auxiliary_loss_clip": 0.01153586,
      "auxiliary_loss_mlp": 0.01023046,
      "balance_loss_clip": 1.04350495,
      "balance_loss_mlp": 1.01583123,
      "epoch": 0.868394156195515,
      "flos": 16654759217280.0,
      "grad_norm": 2.0511217319288586,
      "language_loss": 0.73335361,
      "learning_rate": 1.788286680018083e-07,
      "loss": 0.75511992,
      "num_input_tokens_seen": 156241570,
      "step": 7222,
      "time_per_iteration": 2.718672275543213
    },
    {
      "auxiliary_loss_clip": 0.01140234,
      "auxiliary_loss_mlp": 0.01023217,
      "balance_loss_clip": 1.044065,
      "balance_loss_mlp": 1.016047,
      "epoch": 0.868514399086154,
      "flos": 28001381448960.0,
      "grad_norm": 1.6637833227843744,
      "language_loss": 0.72353375,
      "learning_rate": 1.7850684127992443e-07,
      "loss": 0.74516827,
      "num_input_tokens_seen": 156261315,
      "step": 7223,
      "time_per_iteration": 2.8501052856445312
    },
    {
      "auxiliary_loss_clip": 0.01125389,
      "auxiliary_loss_mlp": 0.0102582,
      "balance_loss_clip": 1.04443371,
      "balance_loss_mlp": 1.0184201,
      "epoch": 0.8686346419767931,
      "flos": 20084587228800.0,
      "grad_norm": 2.0484443985792167,
      "language_loss": 0.70362473,
      "learning_rate": 1.7818529087710378e-07,
      "loss": 0.72513688,
      "num_input_tokens_seen": 156281670,
      "step": 7224,
      "time_per_iteration": 2.725290060043335
    },
    {
      "auxiliary_loss_clip": 0.01150527,
      "auxiliary_loss_mlp": 0.00762105,
      "balance_loss_clip": 1.04430628,
      "balance_loss_mlp": 1.00031936,
      "epoch": 0.8687548848674322,
      "flos": 18223516782720.0,
      "grad_norm": 1.7681518154399882,
      "language_loss": 0.84138501,
      "learning_rate": 1.7786401684212637e-07,
      "loss": 0.86051136,
      "num_input_tokens_seen": 156300500,
      "step": 7225,
      "time_per_iteration": 2.6578962802886963
    },
    {
      "auxiliary_loss_clip": 0.01034196,
      "auxiliary_loss_mlp": 0.01001643,
      "balance_loss_clip": 1.01321983,
      "balance_loss_mlp": 1.00058222,
      "epoch": 0.8688751277580713,
      "flos": 70457885049600.0,
      "grad_norm": 0.7337689220900175,
      "language_loss": 0.55913556,
      "learning_rate": 1.7754301922372883e-07,
      "loss": 0.579494,
      "num_input_tokens_seen": 156350145,
      "step": 7226,
      "time_per_iteration": 3.1123263835906982
    },
    {
      "auxiliary_loss_clip": 0.01100191,
      "auxiliary_loss_mlp": 0.01023015,
      "balance_loss_clip": 1.03839636,
      "balance_loss_mlp": 1.01631296,
      "epoch": 0.8689953706487104,
      "flos": 26906788344960.0,
      "grad_norm": 1.955616906942337,
      "language_loss": 0.8109175,
      "learning_rate": 1.7722229807060617e-07,
      "loss": 0.83214962,
      "num_input_tokens_seen": 156368725,
      "step": 7227,
      "time_per_iteration": 2.8001925945281982
    },
    {
      "auxiliary_loss_clip": 0.01112941,
      "auxiliary_loss_mlp": 0.01025517,
      "balance_loss_clip": 1.03805697,
      "balance_loss_mlp": 1.01820302,
      "epoch": 0.8691156135393495,
      "flos": 34637385438720.0,
      "grad_norm": 1.9755535299927285,
      "language_loss": 0.81624377,
      "learning_rate": 1.7690185343141172e-07,
      "loss": 0.83762836,
      "num_input_tokens_seen": 156388640,
      "step": 7228,
      "time_per_iteration": 2.8757705688476562
    },
    {
      "auxiliary_loss_clip": 0.01135437,
      "auxiliary_loss_mlp": 0.01027087,
      "balance_loss_clip": 1.03868949,
      "balance_loss_mlp": 1.02036619,
      "epoch": 0.8692358564299886,
      "flos": 18989814556800.0,
      "grad_norm": 2.7504030986500427,
      "language_loss": 0.70152318,
      "learning_rate": 1.7658168535475615e-07,
      "loss": 0.72314847,
      "num_input_tokens_seen": 156406425,
      "step": 7229,
      "time_per_iteration": 3.6369171142578125
    },
    {
      "auxiliary_loss_clip": 0.01141683,
      "auxiliary_loss_mlp": 0.0103175,
      "balance_loss_clip": 1.04413331,
      "balance_loss_mlp": 1.02438831,
      "epoch": 0.8693560993206276,
      "flos": 30370839039360.0,
      "grad_norm": 1.5100824508337627,
      "language_loss": 0.64298582,
      "learning_rate": 1.7626179388920948e-07,
      "loss": 0.66472018,
      "num_input_tokens_seen": 156427705,
      "step": 7230,
      "time_per_iteration": 2.7803828716278076
    },
    {
      "auxiliary_loss_clip": 0.01138281,
      "auxiliary_loss_mlp": 0.00762263,
      "balance_loss_clip": 1.04307604,
      "balance_loss_mlp": 1.0003047,
      "epoch": 0.8694763422112668,
      "flos": 27200430028800.0,
      "grad_norm": 1.6504425114459482,
      "language_loss": 0.80373943,
      "learning_rate": 1.7594217908329866e-07,
      "loss": 0.82274485,
      "num_input_tokens_seen": 156449890,
      "step": 7231,
      "time_per_iteration": 2.756636142730713
    },
    {
      "auxiliary_loss_clip": 0.01129687,
      "auxiliary_loss_mlp": 0.01024563,
      "balance_loss_clip": 1.04222608,
      "balance_loss_mlp": 1.01769674,
      "epoch": 0.8695965851019059,
      "flos": 26139161767680.0,
      "grad_norm": 2.0008207769322444,
      "language_loss": 0.74344909,
      "learning_rate": 1.7562284098550895e-07,
      "loss": 0.76499152,
      "num_input_tokens_seen": 156469600,
      "step": 7232,
      "time_per_iteration": 2.8242292404174805
    },
    {
      "auxiliary_loss_clip": 0.01035967,
      "auxiliary_loss_mlp": 0.01000965,
      "balance_loss_clip": 1.00848031,
      "balance_loss_mlp": 0.9999755,
      "epoch": 0.8697168279925449,
      "flos": 67332616456320.0,
      "grad_norm": 0.8339608653136719,
      "language_loss": 0.62236714,
      "learning_rate": 1.753037796442838e-07,
      "loss": 0.64273643,
      "num_input_tokens_seen": 156529040,
      "step": 7233,
      "time_per_iteration": 4.549138307571411
    },
    {
      "auxiliary_loss_clip": 0.0116257,
      "auxiliary_loss_mlp": 0.01026912,
      "balance_loss_clip": 1.04277492,
      "balance_loss_mlp": 1.01962793,
      "epoch": 0.8698370708831841,
      "flos": 19718693337600.0,
      "grad_norm": 2.149235800836467,
      "language_loss": 0.75099307,
      "learning_rate": 1.74984995108024e-07,
      "loss": 0.77288795,
      "num_input_tokens_seen": 156546970,
      "step": 7234,
      "time_per_iteration": 2.6407880783081055
    },
    {
      "auxiliary_loss_clip": 0.01153715,
      "auxiliary_loss_mlp": 0.01026798,
      "balance_loss_clip": 1.04458392,
      "balance_loss_mlp": 1.01929998,
      "epoch": 0.8699573137738231,
      "flos": 12859971068160.0,
      "grad_norm": 2.651492010645613,
      "language_loss": 0.83160651,
      "learning_rate": 1.7466648742508981e-07,
      "loss": 0.85341161,
      "num_input_tokens_seen": 156563155,
      "step": 7235,
      "time_per_iteration": 3.7609012126922607
    },
    {
      "auxiliary_loss_clip": 0.01135787,
      "auxiliary_loss_mlp": 0.01023268,
      "balance_loss_clip": 1.04352427,
      "balance_loss_mlp": 1.01633644,
      "epoch": 0.8700775566644622,
      "flos": 17420733768960.0,
      "grad_norm": 3.1864419325609523,
      "language_loss": 0.84572554,
      "learning_rate": 1.7434825664379837e-07,
      "loss": 0.86731607,
      "num_input_tokens_seen": 156581660,
      "step": 7236,
      "time_per_iteration": 2.6378731727600098
    },
    {
      "auxiliary_loss_clip": 0.01153978,
      "auxiliary_loss_mlp": 0.01028861,
      "balance_loss_clip": 1.0454092,
      "balance_loss_mlp": 1.02161944,
      "epoch": 0.8701977995551013,
      "flos": 13735221770880.0,
      "grad_norm": 2.6545530333264846,
      "language_loss": 0.86445045,
      "learning_rate": 1.740303028124246e-07,
      "loss": 0.88627887,
      "num_input_tokens_seen": 156597720,
      "step": 7237,
      "time_per_iteration": 3.5886640548706055
    },
    {
      "auxiliary_loss_clip": 0.01082294,
      "auxiliary_loss_mlp": 0.01022426,
      "balance_loss_clip": 1.03384101,
      "balance_loss_mlp": 1.01522613,
      "epoch": 0.8703180424457404,
      "flos": 30555707362560.0,
      "grad_norm": 2.0125780664581963,
      "language_loss": 0.75790465,
      "learning_rate": 1.7371262597920212e-07,
      "loss": 0.77895194,
      "num_input_tokens_seen": 156619780,
      "step": 7238,
      "time_per_iteration": 2.837843894958496
    },
    {
      "auxiliary_loss_clip": 0.01107736,
      "auxiliary_loss_mlp": 0.01023909,
      "balance_loss_clip": 1.0428133,
      "balance_loss_mlp": 1.01687312,
      "epoch": 0.8704382853363795,
      "flos": 19608986223360.0,
      "grad_norm": 1.650254867927733,
      "language_loss": 0.7632935,
      "learning_rate": 1.7339522619232195e-07,
      "loss": 0.78460991,
      "num_input_tokens_seen": 156638160,
      "step": 7239,
      "time_per_iteration": 2.702130079269409
    },
    {
      "auxiliary_loss_clip": 0.0114643,
      "auxiliary_loss_mlp": 0.0102498,
      "balance_loss_clip": 1.04256904,
      "balance_loss_mlp": 1.0174247,
      "epoch": 0.8705585282270186,
      "flos": 26613900846720.0,
      "grad_norm": 1.842181176382965,
      "language_loss": 0.75373083,
      "learning_rate": 1.730781034999338e-07,
      "loss": 0.77544498,
      "num_input_tokens_seen": 156659740,
      "step": 7240,
      "time_per_iteration": 2.714712619781494
    },
    {
      "auxiliary_loss_clip": 0.01161791,
      "auxiliary_loss_mlp": 0.01022555,
      "balance_loss_clip": 1.04601431,
      "balance_loss_mlp": 1.01582289,
      "epoch": 0.8706787711176577,
      "flos": 34090465979520.0,
      "grad_norm": 1.7260359105662368,
      "language_loss": 0.73121929,
      "learning_rate": 1.7276125795014497e-07,
      "loss": 0.75306273,
      "num_input_tokens_seen": 156678190,
      "step": 7241,
      "time_per_iteration": 2.6628572940826416
    },
    {
      "auxiliary_loss_clip": 0.01140314,
      "auxiliary_loss_mlp": 0.01026252,
      "balance_loss_clip": 1.04115093,
      "balance_loss_mlp": 1.01844931,
      "epoch": 0.8707990140082967,
      "flos": 14611513968000.0,
      "grad_norm": 3.111027694617384,
      "language_loss": 0.67362064,
      "learning_rate": 1.7244468959102054e-07,
      "loss": 0.69528627,
      "num_input_tokens_seen": 156695245,
      "step": 7242,
      "time_per_iteration": 2.704679250717163
    },
    {
      "auxiliary_loss_clip": 0.01150679,
      "auxiliary_loss_mlp": 0.01023879,
      "balance_loss_clip": 1.04519892,
      "balance_loss_mlp": 1.0170033,
      "epoch": 0.8709192568989359,
      "flos": 20084156265600.0,
      "grad_norm": 2.121500619976293,
      "language_loss": 0.85637796,
      "learning_rate": 1.7212839847058348e-07,
      "loss": 0.87812364,
      "num_input_tokens_seen": 156710375,
      "step": 7243,
      "time_per_iteration": 2.7049427032470703
    },
    {
      "auxiliary_loss_clip": 0.01103923,
      "auxiliary_loss_mlp": 0.0102499,
      "balance_loss_clip": 1.03817129,
      "balance_loss_mlp": 1.01793003,
      "epoch": 0.871039499789575,
      "flos": 16727083251840.0,
      "grad_norm": 5.602977235714113,
      "language_loss": 0.74029577,
      "learning_rate": 1.718123846368147e-07,
      "loss": 0.76158494,
      "num_input_tokens_seen": 156729420,
      "step": 7244,
      "time_per_iteration": 2.7712364196777344
    },
    {
      "auxiliary_loss_clip": 0.01135145,
      "auxiliary_loss_mlp": 0.00762053,
      "balance_loss_clip": 1.0429002,
      "balance_loss_mlp": 1.0002805,
      "epoch": 0.871159742680214,
      "flos": 21068790860160.0,
      "grad_norm": 1.729561588735784,
      "language_loss": 0.71534711,
      "learning_rate": 1.714966481376543e-07,
      "loss": 0.73431909,
      "num_input_tokens_seen": 156746100,
      "step": 7245,
      "time_per_iteration": 2.6597390174865723
    },
    {
      "auxiliary_loss_clip": 0.01150242,
      "auxiliary_loss_mlp": 0.0102393,
      "balance_loss_clip": 1.04326999,
      "balance_loss_mlp": 1.016572,
      "epoch": 0.8712799855708532,
      "flos": 28256526731520.0,
      "grad_norm": 1.9541527202810542,
      "language_loss": 0.83348507,
      "learning_rate": 1.7118118902099797e-07,
      "loss": 0.85522681,
      "num_input_tokens_seen": 156764185,
      "step": 7246,
      "time_per_iteration": 2.7947535514831543
    },
    {
      "auxiliary_loss_clip": 0.01150411,
      "auxiliary_loss_mlp": 0.01025654,
      "balance_loss_clip": 1.04417133,
      "balance_loss_mlp": 1.01869226,
      "epoch": 0.8714002284614922,
      "flos": 22236677665920.0,
      "grad_norm": 1.7324966375242503,
      "language_loss": 0.80830705,
      "learning_rate": 1.7086600733470146e-07,
      "loss": 0.83006775,
      "num_input_tokens_seen": 156784855,
      "step": 7247,
      "time_per_iteration": 2.6020960807800293
    },
    {
      "auxiliary_loss_clip": 0.01146129,
      "auxiliary_loss_mlp": 0.01018448,
      "balance_loss_clip": 1.04227912,
      "balance_loss_mlp": 1.01222551,
      "epoch": 0.8715204713521313,
      "flos": 21431919404160.0,
      "grad_norm": 2.079136764264898,
      "language_loss": 0.77104324,
      "learning_rate": 1.7055110312657738e-07,
      "loss": 0.79268897,
      "num_input_tokens_seen": 156804350,
      "step": 7248,
      "time_per_iteration": 2.6741738319396973
    },
    {
      "auxiliary_loss_clip": 0.01131874,
      "auxiliary_loss_mlp": 0.01027688,
      "balance_loss_clip": 1.04178238,
      "balance_loss_mlp": 1.02042508,
      "epoch": 0.8716407142427703,
      "flos": 23440439180160.0,
      "grad_norm": 3.4059993289215647,
      "language_loss": 0.73671281,
      "learning_rate": 1.702364764443962e-07,
      "loss": 0.75830847,
      "num_input_tokens_seen": 156823425,
      "step": 7249,
      "time_per_iteration": 2.640627145767212
    },
    {
      "auxiliary_loss_clip": 0.01092552,
      "auxiliary_loss_mlp": 0.01023438,
      "balance_loss_clip": 1.03645611,
      "balance_loss_mlp": 1.01614881,
      "epoch": 0.8717609571334095,
      "flos": 27958683156480.0,
      "grad_norm": 1.8956472465721956,
      "language_loss": 0.72415388,
      "learning_rate": 1.6992212733588685e-07,
      "loss": 0.74531376,
      "num_input_tokens_seen": 156843090,
      "step": 7250,
      "time_per_iteration": 2.778545379638672
    },
    {
      "auxiliary_loss_clip": 0.01134993,
      "auxiliary_loss_mlp": 0.01024177,
      "balance_loss_clip": 1.04272413,
      "balance_loss_mlp": 1.0172267,
      "epoch": 0.8718812000240486,
      "flos": 25479482538240.0,
      "grad_norm": 1.9153419237576017,
      "language_loss": 0.75269252,
      "learning_rate": 1.6960805584873538e-07,
      "loss": 0.77428418,
      "num_input_tokens_seen": 156861090,
      "step": 7251,
      "time_per_iteration": 2.6828668117523193
    },
    {
      "auxiliary_loss_clip": 0.01110826,
      "auxiliary_loss_mlp": 0.01022842,
      "balance_loss_clip": 1.03825974,
      "balance_loss_mlp": 1.01575792,
      "epoch": 0.8720014429146876,
      "flos": 23403056100480.0,
      "grad_norm": 1.5621852482781111,
      "language_loss": 0.78255045,
      "learning_rate": 1.6929426203058684e-07,
      "loss": 0.80388719,
      "num_input_tokens_seen": 156881515,
      "step": 7252,
      "time_per_iteration": 2.7120277881622314
    },
    {
      "auxiliary_loss_clip": 0.01166542,
      "auxiliary_loss_mlp": 0.00762962,
      "balance_loss_clip": 1.04314256,
      "balance_loss_mlp": 1.00033724,
      "epoch": 0.8721216858053268,
      "flos": 24352821567360.0,
      "grad_norm": 2.481391710407773,
      "language_loss": 0.80192971,
      "learning_rate": 1.689807459290431e-07,
      "loss": 0.82122481,
      "num_input_tokens_seen": 156900170,
      "step": 7253,
      "time_per_iteration": 2.660639524459839
    },
    {
      "auxiliary_loss_clip": 0.01139246,
      "auxiliary_loss_mlp": 0.01027833,
      "balance_loss_clip": 1.04419446,
      "balance_loss_mlp": 1.02075529,
      "epoch": 0.8722419286959658,
      "flos": 33869687034240.0,
      "grad_norm": 2.1118929973916423,
      "language_loss": 0.70486695,
      "learning_rate": 1.6866750759166437e-07,
      "loss": 0.72653776,
      "num_input_tokens_seen": 156920150,
      "step": 7254,
      "time_per_iteration": 2.7880444526672363
    },
    {
      "auxiliary_loss_clip": 0.01118129,
      "auxiliary_loss_mlp": 0.01023806,
      "balance_loss_clip": 1.03703618,
      "balance_loss_mlp": 1.01653099,
      "epoch": 0.8723621715866049,
      "flos": 18369385914240.0,
      "grad_norm": 2.722148056159217,
      "language_loss": 0.77344733,
      "learning_rate": 1.6835454706596865e-07,
      "loss": 0.79486668,
      "num_input_tokens_seen": 156937980,
      "step": 7255,
      "time_per_iteration": 3.7212982177734375
    },
    {
      "auxiliary_loss_clip": 0.01166938,
      "auxiliary_loss_mlp": 0.01025918,
      "balance_loss_clip": 1.04768622,
      "balance_loss_mlp": 1.01846147,
      "epoch": 0.8724824144772441,
      "flos": 22013348855040.0,
      "grad_norm": 1.841066691904238,
      "language_loss": 0.73459494,
      "learning_rate": 1.680418643994317e-07,
      "loss": 0.75652349,
      "num_input_tokens_seen": 156956550,
      "step": 7256,
      "time_per_iteration": 2.613248109817505
    },
    {
      "auxiliary_loss_clip": 0.01063779,
      "auxiliary_loss_mlp": 0.01000399,
      "balance_loss_clip": 1.00982738,
      "balance_loss_mlp": 0.99928439,
      "epoch": 0.8726026573678831,
      "flos": 66698720213760.0,
      "grad_norm": 0.8866463219472269,
      "language_loss": 0.64473844,
      "learning_rate": 1.6772945963948738e-07,
      "loss": 0.66538012,
      "num_input_tokens_seen": 157014715,
      "step": 7257,
      "time_per_iteration": 3.1832034587860107
    },
    {
      "auxiliary_loss_clip": 0.01136705,
      "auxiliary_loss_mlp": 0.0102477,
      "balance_loss_clip": 1.04436493,
      "balance_loss_mlp": 1.01748335,
      "epoch": 0.8727229002585222,
      "flos": 13370908078080.0,
      "grad_norm": 2.5237168916348076,
      "language_loss": 0.7744329,
      "learning_rate": 1.6741733283352733e-07,
      "loss": 0.79604769,
      "num_input_tokens_seen": 157032320,
      "step": 7258,
      "time_per_iteration": 2.6153995990753174
    },
    {
      "auxiliary_loss_clip": 0.01111996,
      "auxiliary_loss_mlp": 0.0102604,
      "balance_loss_clip": 1.03997302,
      "balance_loss_mlp": 1.01864934,
      "epoch": 0.8728431431491613,
      "flos": 21796987282560.0,
      "grad_norm": 1.5407311822511656,
      "language_loss": 0.83775401,
      "learning_rate": 1.6710548402890102e-07,
      "loss": 0.85913444,
      "num_input_tokens_seen": 157052845,
      "step": 7259,
      "time_per_iteration": 3.683941602706909
    },
    {
      "auxiliary_loss_clip": 0.01169614,
      "auxiliary_loss_mlp": 0.01026806,
      "balance_loss_clip": 1.04579294,
      "balance_loss_mlp": 1.01900995,
      "epoch": 0.8729633860398004,
      "flos": 36173823742080.0,
      "grad_norm": 1.9671246123920263,
      "language_loss": 0.6712265,
      "learning_rate": 1.6679391327291527e-07,
      "loss": 0.69319069,
      "num_input_tokens_seen": 157074050,
      "step": 7260,
      "time_per_iteration": 2.754448413848877
    },
    {
      "auxiliary_loss_clip": 0.01137933,
      "auxiliary_loss_mlp": 0.01024907,
      "balance_loss_clip": 1.04058909,
      "balance_loss_mlp": 1.01787984,
      "epoch": 0.8730836289304394,
      "flos": 16359680989440.0,
      "grad_norm": 3.3832735154254676,
      "language_loss": 0.68257892,
      "learning_rate": 1.6648262061283492e-07,
      "loss": 0.7042073,
      "num_input_tokens_seen": 157089350,
      "step": 7261,
      "time_per_iteration": 3.443157434463501
    },
    {
      "auxiliary_loss_clip": 0.01122288,
      "auxiliary_loss_mlp": 0.01019692,
      "balance_loss_clip": 1.03860259,
      "balance_loss_mlp": 1.01318038,
      "epoch": 0.8732038718210786,
      "flos": 21215126868480.0,
      "grad_norm": 1.7893717377168012,
      "language_loss": 0.73836637,
      "learning_rate": 1.6617160609588353e-07,
      "loss": 0.75978613,
      "num_input_tokens_seen": 157108525,
      "step": 7262,
      "time_per_iteration": 2.699214458465576
    },
    {
      "auxiliary_loss_clip": 0.01140241,
      "auxiliary_loss_mlp": 0.01024862,
      "balance_loss_clip": 1.04215324,
      "balance_loss_mlp": 1.01767695,
      "epoch": 0.8733241147117177,
      "flos": 16610696208000.0,
      "grad_norm": 2.106505762121692,
      "language_loss": 0.72605437,
      "learning_rate": 1.6586086976924163e-07,
      "loss": 0.7477054,
      "num_input_tokens_seen": 157124025,
      "step": 7263,
      "time_per_iteration": 3.587695598602295
    },
    {
      "auxiliary_loss_clip": 0.01151199,
      "auxiliary_loss_mlp": 0.01024519,
      "balance_loss_clip": 1.04277718,
      "balance_loss_mlp": 1.01761961,
      "epoch": 0.8734443576023567,
      "flos": 20193935207040.0,
      "grad_norm": 2.072181274733871,
      "language_loss": 0.78584707,
      "learning_rate": 1.6555041168004747e-07,
      "loss": 0.80760419,
      "num_input_tokens_seen": 157143345,
      "step": 7264,
      "time_per_iteration": 2.577897310256958
    },
    {
      "auxiliary_loss_clip": 0.01133803,
      "auxiliary_loss_mlp": 0.0103179,
      "balance_loss_clip": 1.04284883,
      "balance_loss_mlp": 1.02482867,
      "epoch": 0.8735646004929959,
      "flos": 18041162411520.0,
      "grad_norm": 2.0580591432551896,
      "language_loss": 0.69129175,
      "learning_rate": 1.6524023187539715e-07,
      "loss": 0.71294767,
      "num_input_tokens_seen": 157161630,
      "step": 7265,
      "time_per_iteration": 2.698284387588501
    },
    {
      "auxiliary_loss_clip": 0.01134231,
      "auxiliary_loss_mlp": 0.01024626,
      "balance_loss_clip": 1.04109156,
      "balance_loss_mlp": 1.01741362,
      "epoch": 0.873684843383635,
      "flos": 20262344659200.0,
      "grad_norm": 2.132302142650764,
      "language_loss": 0.75218976,
      "learning_rate": 1.649303304023446e-07,
      "loss": 0.77377832,
      "num_input_tokens_seen": 157181385,
      "step": 7266,
      "time_per_iteration": 2.621762752532959
    },
    {
      "auxiliary_loss_clip": 0.011181,
      "auxiliary_loss_mlp": 0.0102518,
      "balance_loss_clip": 1.04323363,
      "balance_loss_mlp": 1.01839066,
      "epoch": 0.873805086274274,
      "flos": 16947287579520.0,
      "grad_norm": 1.9028484610913756,
      "language_loss": 0.7883181,
      "learning_rate": 1.6462070730790246e-07,
      "loss": 0.80975091,
      "num_input_tokens_seen": 157200545,
      "step": 7267,
      "time_per_iteration": 2.6563141345977783
    },
    {
      "auxiliary_loss_clip": 0.01131091,
      "auxiliary_loss_mlp": 0.01022847,
      "balance_loss_clip": 1.03855038,
      "balance_loss_mlp": 1.01599836,
      "epoch": 0.8739253291649132,
      "flos": 18041270152320.0,
      "grad_norm": 2.5014693423972068,
      "language_loss": 0.78546011,
      "learning_rate": 1.6431136263903912e-07,
      "loss": 0.80699944,
      "num_input_tokens_seen": 157219545,
      "step": 7268,
      "time_per_iteration": 2.609837055206299
    },
    {
      "auxiliary_loss_clip": 0.01152752,
      "auxiliary_loss_mlp": 0.0076242,
      "balance_loss_clip": 1.04156637,
      "balance_loss_mlp": 1.00030661,
      "epoch": 0.8740455720555522,
      "flos": 21325085377920.0,
      "grad_norm": 3.18531652504852,
      "language_loss": 0.73314148,
      "learning_rate": 1.6400229644268282e-07,
      "loss": 0.75229323,
      "num_input_tokens_seen": 157237900,
      "step": 7269,
      "time_per_iteration": 2.628361701965332
    },
    {
      "auxiliary_loss_clip": 0.01120832,
      "auxiliary_loss_mlp": 0.0102603,
      "balance_loss_clip": 1.04594326,
      "balance_loss_mlp": 1.01887119,
      "epoch": 0.8741658149461913,
      "flos": 15158684822400.0,
      "grad_norm": 2.0070455872733595,
      "language_loss": 0.81079817,
      "learning_rate": 1.6369350876571852e-07,
      "loss": 0.83226675,
      "num_input_tokens_seen": 157256055,
      "step": 7270,
      "time_per_iteration": 2.630439281463623
    },
    {
      "auxiliary_loss_clip": 0.01105183,
      "auxiliary_loss_mlp": 0.01026293,
      "balance_loss_clip": 1.03800571,
      "balance_loss_mlp": 1.01937246,
      "epoch": 0.8742860578368304,
      "flos": 23039855729280.0,
      "grad_norm": 1.967326702373711,
      "language_loss": 0.81462806,
      "learning_rate": 1.6338499965498874e-07,
      "loss": 0.83594286,
      "num_input_tokens_seen": 157274785,
      "step": 7271,
      "time_per_iteration": 2.7288546562194824
    },
    {
      "auxiliary_loss_clip": 0.01117921,
      "auxiliary_loss_mlp": 0.0102348,
      "balance_loss_clip": 1.03998494,
      "balance_loss_mlp": 1.01601148,
      "epoch": 0.8744063007274695,
      "flos": 28145347159680.0,
      "grad_norm": 1.5531408435442666,
      "language_loss": 0.77211964,
      "learning_rate": 1.630767691572943e-07,
      "loss": 0.79353362,
      "num_input_tokens_seen": 157294805,
      "step": 7272,
      "time_per_iteration": 2.708362340927124
    },
    {
      "auxiliary_loss_clip": 0.0104614,
      "auxiliary_loss_mlp": 0.01000423,
      "balance_loss_clip": 1.00983977,
      "balance_loss_mlp": 0.99931461,
      "epoch": 0.8745265436181086,
      "flos": 64034076654720.0,
      "grad_norm": 0.7395824641449439,
      "language_loss": 0.53510451,
      "learning_rate": 1.6276881731939306e-07,
      "loss": 0.55557013,
      "num_input_tokens_seen": 157356695,
      "step": 7273,
      "time_per_iteration": 3.2982938289642334
    },
    {
      "auxiliary_loss_clip": 0.01147267,
      "auxiliary_loss_mlp": 0.01023166,
      "balance_loss_clip": 1.0440836,
      "balance_loss_mlp": 1.0160135,
      "epoch": 0.8746467865087477,
      "flos": 28658618553600.0,
      "grad_norm": 2.8046853219473564,
      "language_loss": 0.75306726,
      "learning_rate": 1.6246114418800193e-07,
      "loss": 0.77477157,
      "num_input_tokens_seen": 157376975,
      "step": 7274,
      "time_per_iteration": 2.6496031284332275
    },
    {
      "auxiliary_loss_clip": 0.0114845,
      "auxiliary_loss_mlp": 0.01030917,
      "balance_loss_clip": 1.04471493,
      "balance_loss_mlp": 1.02374625,
      "epoch": 0.8747670293993868,
      "flos": 23985850268160.0,
      "grad_norm": 1.6831792379573156,
      "language_loss": 0.76723397,
      "learning_rate": 1.6215374980979423e-07,
      "loss": 0.78902763,
      "num_input_tokens_seen": 157397385,
      "step": 7275,
      "time_per_iteration": 2.636554002761841
    },
    {
      "auxiliary_loss_clip": 0.01149804,
      "auxiliary_loss_mlp": 0.01024321,
      "balance_loss_clip": 1.04678965,
      "balance_loss_mlp": 1.01728177,
      "epoch": 0.8748872722900258,
      "flos": 45221624478720.0,
      "grad_norm": 2.3883594851199224,
      "language_loss": 0.68456042,
      "learning_rate": 1.6184663423140133e-07,
      "loss": 0.70630169,
      "num_input_tokens_seen": 157417685,
      "step": 7276,
      "time_per_iteration": 2.8852758407592773
    },
    {
      "auxiliary_loss_clip": 0.01111369,
      "auxiliary_loss_mlp": 0.01032495,
      "balance_loss_clip": 1.0406177,
      "balance_loss_mlp": 1.0258106,
      "epoch": 0.875007515180665,
      "flos": 19754280737280.0,
      "grad_norm": 3.426875908680601,
      "language_loss": 0.63856876,
      "learning_rate": 1.615397974994126e-07,
      "loss": 0.66000742,
      "num_input_tokens_seen": 157435490,
      "step": 7277,
      "time_per_iteration": 2.7091591358184814
    },
    {
      "auxiliary_loss_clip": 0.0116426,
      "auxiliary_loss_mlp": 0.01024694,
      "balance_loss_clip": 1.04661441,
      "balance_loss_mlp": 1.01808119,
      "epoch": 0.875127758071304,
      "flos": 22710734386560.0,
      "grad_norm": 1.5091558039609847,
      "language_loss": 0.80714977,
      "learning_rate": 1.6123323966037438e-07,
      "loss": 0.82903922,
      "num_input_tokens_seen": 157454010,
      "step": 7278,
      "time_per_iteration": 2.6624274253845215
    },
    {
      "auxiliary_loss_clip": 0.01165226,
      "auxiliary_loss_mlp": 0.0102558,
      "balance_loss_clip": 1.04691386,
      "balance_loss_mlp": 1.01844549,
      "epoch": 0.8752480009619431,
      "flos": 23403846199680.0,
      "grad_norm": 2.362465263535424,
      "language_loss": 0.78463286,
      "learning_rate": 1.6092696076079216e-07,
      "loss": 0.80654097,
      "num_input_tokens_seen": 157472385,
      "step": 7279,
      "time_per_iteration": 2.585635185241699
    },
    {
      "auxiliary_loss_clip": 0.01110085,
      "auxiliary_loss_mlp": 0.01023621,
      "balance_loss_clip": 1.03865314,
      "balance_loss_mlp": 1.01689792,
      "epoch": 0.8753682438525822,
      "flos": 26213101914240.0,
      "grad_norm": 1.6492561958061493,
      "language_loss": 0.73807371,
      "learning_rate": 1.6062096084712785e-07,
      "loss": 0.75941074,
      "num_input_tokens_seen": 157493735,
      "step": 7280,
      "time_per_iteration": 2.8308045864105225
    },
    {
      "auxiliary_loss_clip": 0.01127536,
      "auxiliary_loss_mlp": 0.00762087,
      "balance_loss_clip": 1.0392319,
      "balance_loss_mlp": 1.00034499,
      "epoch": 0.8754884867432213,
      "flos": 23326745656320.0,
      "grad_norm": 2.180551441267642,
      "language_loss": 0.70590103,
      "learning_rate": 1.6031523996580098e-07,
      "loss": 0.72479725,
      "num_input_tokens_seen": 157511295,
      "step": 7281,
      "time_per_iteration": 3.6046698093414307
    },
    {
      "auxiliary_loss_clip": 0.01129718,
      "auxiliary_loss_mlp": 0.01025158,
      "balance_loss_clip": 1.04172122,
      "balance_loss_mlp": 1.01733172,
      "epoch": 0.8756087296338604,
      "flos": 12495226412160.0,
      "grad_norm": 2.0295094836304406,
      "language_loss": 0.66268319,
      "learning_rate": 1.6000979816318981e-07,
      "loss": 0.68423194,
      "num_input_tokens_seen": 157529760,
      "step": 7282,
      "time_per_iteration": 2.618992567062378
    },
    {
      "auxiliary_loss_clip": 0.01146471,
      "auxiliary_loss_mlp": 0.01024356,
      "balance_loss_clip": 1.04449463,
      "balance_loss_mlp": 1.01699483,
      "epoch": 0.8757289725244994,
      "flos": 18952898353920.0,
      "grad_norm": 2.3779890626651072,
      "language_loss": 0.75254142,
      "learning_rate": 1.5970463548562886e-07,
      "loss": 0.77424967,
      "num_input_tokens_seen": 157548915,
      "step": 7283,
      "time_per_iteration": 2.7662734985351562
    },
    {
      "auxiliary_loss_clip": 0.01132961,
      "auxiliary_loss_mlp": 0.01021469,
      "balance_loss_clip": 1.04234564,
      "balance_loss_mlp": 1.01414979,
      "epoch": 0.8758492154151386,
      "flos": 25265958140160.0,
      "grad_norm": 1.8332834981617228,
      "language_loss": 0.71480161,
      "learning_rate": 1.5939975197941192e-07,
      "loss": 0.73634589,
      "num_input_tokens_seen": 157570570,
      "step": 7284,
      "time_per_iteration": 2.7201197147369385
    },
    {
      "auxiliary_loss_clip": 0.01045485,
      "auxiliary_loss_mlp": 0.01000387,
      "balance_loss_clip": 1.01040816,
      "balance_loss_mlp": 0.99928993,
      "epoch": 0.8759694583057777,
      "flos": 65571664193280.0,
      "grad_norm": 0.8068913526430259,
      "language_loss": 0.53328645,
      "learning_rate": 1.5909514769078892e-07,
      "loss": 0.55374515,
      "num_input_tokens_seen": 157635675,
      "step": 7285,
      "time_per_iteration": 4.465670824050903
    },
    {
      "auxiliary_loss_clip": 0.01114719,
      "auxiliary_loss_mlp": 0.01029068,
      "balance_loss_clip": 1.04323983,
      "balance_loss_mlp": 1.02235317,
      "epoch": 0.8760897011964167,
      "flos": 25446193608960.0,
      "grad_norm": 1.518160326874458,
      "language_loss": 0.77572083,
      "learning_rate": 1.5879082266596867e-07,
      "loss": 0.79715872,
      "num_input_tokens_seen": 157657015,
      "step": 7286,
      "time_per_iteration": 2.7442522048950195
    },
    {
      "auxiliary_loss_clip": 0.01127476,
      "auxiliary_loss_mlp": 0.01023479,
      "balance_loss_clip": 1.03642857,
      "balance_loss_mlp": 1.01663077,
      "epoch": 0.8762099440870559,
      "flos": 28984830894720.0,
      "grad_norm": 2.175116705612869,
      "language_loss": 0.72145331,
      "learning_rate": 1.5848677695111645e-07,
      "loss": 0.7429629,
      "num_input_tokens_seen": 157678615,
      "step": 7287,
      "time_per_iteration": 3.588203191757202
    },
    {
      "auxiliary_loss_clip": 0.01131906,
      "auxiliary_loss_mlp": 0.0102708,
      "balance_loss_clip": 1.0443902,
      "balance_loss_mlp": 1.01956725,
      "epoch": 0.8763301869776949,
      "flos": 21609461352960.0,
      "grad_norm": 2.153063874875102,
      "language_loss": 0.6894412,
      "learning_rate": 1.5818301059235562e-07,
      "loss": 0.71103108,
      "num_input_tokens_seen": 157693790,
      "step": 7288,
      "time_per_iteration": 2.6855738162994385
    },
    {
      "auxiliary_loss_clip": 0.0114031,
      "auxiliary_loss_mlp": 0.01032621,
      "balance_loss_clip": 1.04395461,
      "balance_loss_mlp": 1.0256263,
      "epoch": 0.876450429868334,
      "flos": 24644416176000.0,
      "grad_norm": 1.6316562276810045,
      "language_loss": 0.81172907,
      "learning_rate": 1.578795236357684e-07,
      "loss": 0.83345842,
      "num_input_tokens_seen": 157715255,
      "step": 7289,
      "time_per_iteration": 3.6084144115448
    },
    {
      "auxiliary_loss_clip": 0.01139863,
      "auxiliary_loss_mlp": 0.01023525,
      "balance_loss_clip": 1.04496586,
      "balance_loss_mlp": 1.01647687,
      "epoch": 0.8765706727589732,
      "flos": 20260046188800.0,
      "grad_norm": 2.164144388502614,
      "language_loss": 0.85402048,
      "learning_rate": 1.5757631612739218e-07,
      "loss": 0.87565434,
      "num_input_tokens_seen": 157728800,
      "step": 7290,
      "time_per_iteration": 2.6464099884033203
    },
    {
      "auxiliary_loss_clip": 0.01063688,
      "auxiliary_loss_mlp": 0.01001535,
      "balance_loss_clip": 1.00987816,
      "balance_loss_mlp": 1.00046194,
      "epoch": 0.8766909156496122,
      "flos": 71371165276800.0,
      "grad_norm": 0.7780697193850279,
      "language_loss": 0.61451524,
      "learning_rate": 1.572733881132242e-07,
      "loss": 0.63516754,
      "num_input_tokens_seen": 157789445,
      "step": 7291,
      "time_per_iteration": 3.277047872543335
    },
    {
      "auxiliary_loss_clip": 0.01028041,
      "auxiliary_loss_mlp": 0.01001151,
      "balance_loss_clip": 1.00982809,
      "balance_loss_mlp": 1.0001322,
      "epoch": 0.8768111585402513,
      "flos": 69523490603520.0,
      "grad_norm": 0.7776909159034786,
      "language_loss": 0.5852173,
      "learning_rate": 1.5697073963921814e-07,
      "loss": 0.60550922,
      "num_input_tokens_seen": 157848685,
      "step": 7292,
      "time_per_iteration": 3.2369918823242188
    },
    {
      "auxiliary_loss_clip": 0.01150697,
      "auxiliary_loss_mlp": 0.01024459,
      "balance_loss_clip": 1.04483223,
      "balance_loss_mlp": 1.01706839,
      "epoch": 0.8769314014308904,
      "flos": 18838558385280.0,
      "grad_norm": 8.11853627589704,
      "language_loss": 0.85482097,
      "learning_rate": 1.566683707512857e-07,
      "loss": 0.87657249,
      "num_input_tokens_seen": 157866360,
      "step": 7293,
      "time_per_iteration": 2.6047167778015137
    },
    {
      "auxiliary_loss_clip": 0.01132545,
      "auxiliary_loss_mlp": 0.01025128,
      "balance_loss_clip": 1.04087603,
      "balance_loss_mlp": 1.01779985,
      "epoch": 0.8770516443215295,
      "flos": 14976402278400.0,
      "grad_norm": 1.9239766604471977,
      "language_loss": 0.79340392,
      "learning_rate": 1.5636628149529553e-07,
      "loss": 0.81498069,
      "num_input_tokens_seen": 157884150,
      "step": 7294,
      "time_per_iteration": 2.6527507305145264
    },
    {
      "auxiliary_loss_clip": 0.01135643,
      "auxiliary_loss_mlp": 0.01024819,
      "balance_loss_clip": 1.04136741,
      "balance_loss_mlp": 1.01852143,
      "epoch": 0.8771718872121685,
      "flos": 31649654021760.0,
      "grad_norm": 2.155165612399429,
      "language_loss": 0.79473811,
      "learning_rate": 1.560644719170743e-07,
      "loss": 0.81634271,
      "num_input_tokens_seen": 157905020,
      "step": 7295,
      "time_per_iteration": 2.693279266357422
    },
    {
      "auxiliary_loss_clip": 0.01121506,
      "auxiliary_loss_mlp": 0.01025904,
      "balance_loss_clip": 1.03929377,
      "balance_loss_mlp": 1.01829505,
      "epoch": 0.8772921301028077,
      "flos": 36095466222720.0,
      "grad_norm": 1.9870419563178132,
      "language_loss": 0.720456,
      "learning_rate": 1.5576294206240692e-07,
      "loss": 0.74193013,
      "num_input_tokens_seen": 157924545,
      "step": 7296,
      "time_per_iteration": 2.8292758464813232
    },
    {
      "auxiliary_loss_clip": 0.01133083,
      "auxiliary_loss_mlp": 0.01025522,
      "balance_loss_clip": 1.04272318,
      "balance_loss_mlp": 1.01822364,
      "epoch": 0.8774123729934468,
      "flos": 57116961849600.0,
      "grad_norm": 1.7109750929582392,
      "language_loss": 0.680282,
      "learning_rate": 1.5546169197703507e-07,
      "loss": 0.70186806,
      "num_input_tokens_seen": 157950820,
      "step": 7297,
      "time_per_iteration": 3.017703056335449
    },
    {
      "auxiliary_loss_clip": 0.01141063,
      "auxiliary_loss_mlp": 0.01029456,
      "balance_loss_clip": 1.04113734,
      "balance_loss_mlp": 1.02262509,
      "epoch": 0.8775326158840858,
      "flos": 23914495900800.0,
      "grad_norm": 2.8667188378766455,
      "language_loss": 0.77309406,
      "learning_rate": 1.5516072170665774e-07,
      "loss": 0.79479921,
      "num_input_tokens_seen": 157968790,
      "step": 7298,
      "time_per_iteration": 2.6317033767700195
    },
    {
      "auxiliary_loss_clip": 0.01153034,
      "auxiliary_loss_mlp": 0.01023293,
      "balance_loss_clip": 1.04476976,
      "balance_loss_mlp": 1.01659298,
      "epoch": 0.877652858774725,
      "flos": 17123285243520.0,
      "grad_norm": 1.812087348607546,
      "language_loss": 0.86755735,
      "learning_rate": 1.5486003129693214e-07,
      "loss": 0.88932061,
      "num_input_tokens_seen": 157986155,
      "step": 7299,
      "time_per_iteration": 2.555802822113037
    },
    {
      "auxiliary_loss_clip": 0.01154185,
      "auxiliary_loss_mlp": 0.01022068,
      "balance_loss_clip": 1.04455984,
      "balance_loss_mlp": 1.0148797,
      "epoch": 0.877773101665364,
      "flos": 16508961912960.0,
      "grad_norm": 2.2403401874713147,
      "language_loss": 0.7796911,
      "learning_rate": 1.545596207934725e-07,
      "loss": 0.80145359,
      "num_input_tokens_seen": 158004640,
      "step": 7300,
      "time_per_iteration": 2.6984965801239014
    },
    {
      "auxiliary_loss_clip": 0.01128927,
      "auxiliary_loss_mlp": 0.01023965,
      "balance_loss_clip": 1.03950858,
      "balance_loss_mlp": 1.01722062,
      "epoch": 0.8778933445560031,
      "flos": 22053209973120.0,
      "grad_norm": 1.7561554186803003,
      "language_loss": 0.77897739,
      "learning_rate": 1.5425949024185147e-07,
      "loss": 0.80050635,
      "num_input_tokens_seen": 158024665,
      "step": 7301,
      "time_per_iteration": 2.6678168773651123
    },
    {
      "auxiliary_loss_clip": 0.0113636,
      "auxiliary_loss_mlp": 0.01033852,
      "balance_loss_clip": 1.03970563,
      "balance_loss_mlp": 1.0269289,
      "epoch": 0.8780135874466423,
      "flos": 22564757514240.0,
      "grad_norm": 1.8473028573438084,
      "language_loss": 0.67654943,
      "learning_rate": 1.5395963968759818e-07,
      "loss": 0.69825149,
      "num_input_tokens_seen": 158044940,
      "step": 7302,
      "time_per_iteration": 2.6924238204956055
    },
    {
      "auxiliary_loss_clip": 0.01137139,
      "auxiliary_loss_mlp": 0.01025727,
      "balance_loss_clip": 1.04152226,
      "balance_loss_mlp": 1.01868165,
      "epoch": 0.8781338303372813,
      "flos": 61531999073280.0,
      "grad_norm": 1.8836446766541677,
      "language_loss": 0.643996,
      "learning_rate": 1.536600691761998e-07,
      "loss": 0.66562462,
      "num_input_tokens_seen": 158070770,
      "step": 7303,
      "time_per_iteration": 3.019775152206421
    },
    {
      "auxiliary_loss_clip": 0.01126032,
      "auxiliary_loss_mlp": 0.01022588,
      "balance_loss_clip": 1.04357266,
      "balance_loss_mlp": 1.01602244,
      "epoch": 0.8782540732279204,
      "flos": 22674751937280.0,
      "grad_norm": 1.709293467872228,
      "language_loss": 0.71925914,
      "learning_rate": 1.5336077875310084e-07,
      "loss": 0.74074537,
      "num_input_tokens_seen": 158089995,
      "step": 7304,
      "time_per_iteration": 2.684757709503174
    },
    {
      "auxiliary_loss_clip": 0.01111025,
      "auxiliary_loss_mlp": 0.0102203,
      "balance_loss_clip": 1.03975415,
      "balance_loss_mlp": 1.01536906,
      "epoch": 0.8783743161185595,
      "flos": 16070348937600.0,
      "grad_norm": 1.856272974650716,
      "language_loss": 0.74417388,
      "learning_rate": 1.5306176846370321e-07,
      "loss": 0.76550448,
      "num_input_tokens_seen": 158108140,
      "step": 7305,
      "time_per_iteration": 2.7901532649993896
    },
    {
      "auxiliary_loss_clip": 0.01144548,
      "auxiliary_loss_mlp": 0.01024718,
      "balance_loss_clip": 1.04204941,
      "balance_loss_mlp": 1.0177412,
      "epoch": 0.8784945590091986,
      "flos": 26067879227520.0,
      "grad_norm": 1.8957695556898477,
      "language_loss": 0.74132109,
      "learning_rate": 1.5276303835336712e-07,
      "loss": 0.76301378,
      "num_input_tokens_seen": 158128680,
      "step": 7306,
      "time_per_iteration": 2.722118854522705
    },
    {
      "auxiliary_loss_clip": 0.01055053,
      "auxiliary_loss_mlp": 0.01001821,
      "balance_loss_clip": 1.01033807,
      "balance_loss_mlp": 1.00074196,
      "epoch": 0.8786148018998376,
      "flos": 62720643939840.0,
      "grad_norm": 0.7599929587460502,
      "language_loss": 0.53532767,
      "learning_rate": 1.524645884674094e-07,
      "loss": 0.55589646,
      "num_input_tokens_seen": 158185610,
      "step": 7307,
      "time_per_iteration": 4.130972862243652
    },
    {
      "auxiliary_loss_clip": 0.01168498,
      "auxiliary_loss_mlp": 0.00762748,
      "balance_loss_clip": 1.04665542,
      "balance_loss_mlp": 1.00033689,
      "epoch": 0.8787350447904768,
      "flos": 21652734263040.0,
      "grad_norm": 1.9801854911587087,
      "language_loss": 0.7901355,
      "learning_rate": 1.521664188511047e-07,
      "loss": 0.809448,
      "num_input_tokens_seen": 158205635,
      "step": 7308,
      "time_per_iteration": 2.6811304092407227
    },
    {
      "auxiliary_loss_clip": 0.01135825,
      "auxiliary_loss_mlp": 0.00761913,
      "balance_loss_clip": 1.04574776,
      "balance_loss_mlp": 1.00032067,
      "epoch": 0.8788552876811159,
      "flos": 25478476957440.0,
      "grad_norm": 1.8161530161664998,
      "language_loss": 0.80703342,
      "learning_rate": 1.518685295496851e-07,
      "loss": 0.82601082,
      "num_input_tokens_seen": 158223495,
      "step": 7309,
      "time_per_iteration": 2.680771827697754
    },
    {
      "auxiliary_loss_clip": 0.01150382,
      "auxiliary_loss_mlp": 0.01025273,
      "balance_loss_clip": 1.04215217,
      "balance_loss_mlp": 1.01821232,
      "epoch": 0.8789755305717549,
      "flos": 22310222762880.0,
      "grad_norm": 1.842008748910858,
      "language_loss": 0.85548961,
      "learning_rate": 1.5157092060833975e-07,
      "loss": 0.87724614,
      "num_input_tokens_seen": 158243145,
      "step": 7310,
      "time_per_iteration": 2.646902084350586
    },
    {
      "auxiliary_loss_clip": 0.01134012,
      "auxiliary_loss_mlp": 0.01022448,
      "balance_loss_clip": 1.04064727,
      "balance_loss_mlp": 1.01546526,
      "epoch": 0.879095773462394,
      "flos": 29310971408640.0,
      "grad_norm": 1.710386695610177,
      "language_loss": 0.65955323,
      "learning_rate": 1.5127359207221658e-07,
      "loss": 0.68111777,
      "num_input_tokens_seen": 158262625,
      "step": 7311,
      "time_per_iteration": 3.7048845291137695
    },
    {
      "auxiliary_loss_clip": 0.01082797,
      "auxiliary_loss_mlp": 0.01024492,
      "balance_loss_clip": 1.03319788,
      "balance_loss_mlp": 1.01727974,
      "epoch": 0.8792160163530331,
      "flos": 16690023394560.0,
      "grad_norm": 1.872200122242252,
      "language_loss": 0.73467815,
      "learning_rate": 1.5097654398641923e-07,
      "loss": 0.75575107,
      "num_input_tokens_seen": 158280530,
      "step": 7312,
      "time_per_iteration": 2.8412325382232666
    },
    {
      "auxiliary_loss_clip": 0.01155188,
      "auxiliary_loss_mlp": 0.01024128,
      "balance_loss_clip": 1.04535174,
      "balance_loss_mlp": 1.01757717,
      "epoch": 0.8793362592436722,
      "flos": 24499301230080.0,
      "grad_norm": 1.3956736803658787,
      "language_loss": 0.731996,
      "learning_rate": 1.5067977639601014e-07,
      "loss": 0.75378919,
      "num_input_tokens_seen": 158303290,
      "step": 7313,
      "time_per_iteration": 3.5221035480499268
    },
    {
      "auxiliary_loss_clip": 0.01134914,
      "auxiliary_loss_mlp": 0.01023593,
      "balance_loss_clip": 1.04295158,
      "balance_loss_mlp": 1.01642871,
      "epoch": 0.8794565021343113,
      "flos": 14538399834240.0,
      "grad_norm": 2.022052663508477,
      "language_loss": 0.70458853,
      "learning_rate": 1.5038328934600864e-07,
      "loss": 0.72617358,
      "num_input_tokens_seen": 158319925,
      "step": 7314,
      "time_per_iteration": 3.540907621383667
    },
    {
      "auxiliary_loss_clip": 0.01137812,
      "auxiliary_loss_mlp": 0.01025539,
      "balance_loss_clip": 1.04503047,
      "balance_loss_mlp": 1.01868451,
      "epoch": 0.8795767450249504,
      "flos": 39530286224640.0,
      "grad_norm": 1.8319473022924495,
      "language_loss": 0.6992141,
      "learning_rate": 1.5008708288139161e-07,
      "loss": 0.72084755,
      "num_input_tokens_seen": 158342285,
      "step": 7315,
      "time_per_iteration": 2.796792507171631
    },
    {
      "auxiliary_loss_clip": 0.01150985,
      "auxiliary_loss_mlp": 0.01022016,
      "balance_loss_clip": 1.04665208,
      "balance_loss_mlp": 1.01499486,
      "epoch": 0.8796969879155895,
      "flos": 22960672197120.0,
      "grad_norm": 2.3463255076044236,
      "language_loss": 0.73192626,
      "learning_rate": 1.497911570470931e-07,
      "loss": 0.75365627,
      "num_input_tokens_seen": 158362290,
      "step": 7316,
      "time_per_iteration": 2.619607448577881
    },
    {
      "auxiliary_loss_clip": 0.01111408,
      "auxiliary_loss_mlp": 0.0102369,
      "balance_loss_clip": 1.03940582,
      "balance_loss_mlp": 1.0164839,
      "epoch": 0.8798172308062285,
      "flos": 28362427004160.0,
      "grad_norm": 1.6201314254989154,
      "language_loss": 0.85595918,
      "learning_rate": 1.494955118880048e-07,
      "loss": 0.87731016,
      "num_input_tokens_seen": 158383275,
      "step": 7317,
      "time_per_iteration": 2.7967560291290283
    },
    {
      "auxiliary_loss_clip": 0.01150308,
      "auxiliary_loss_mlp": 0.01021643,
      "balance_loss_clip": 1.04371107,
      "balance_loss_mlp": 1.0151844,
      "epoch": 0.8799374736968677,
      "flos": 23988974751360.0,
      "grad_norm": 1.8625832051271416,
      "language_loss": 0.72841799,
      "learning_rate": 1.4920014744897634e-07,
      "loss": 0.75013757,
      "num_input_tokens_seen": 158402690,
      "step": 7318,
      "time_per_iteration": 2.68392014503479
    },
    {
      "auxiliary_loss_clip": 0.01129351,
      "auxiliary_loss_mlp": 0.01027001,
      "balance_loss_clip": 1.04272437,
      "balance_loss_mlp": 1.01978898,
      "epoch": 0.8800577165875068,
      "flos": 25630271832960.0,
      "grad_norm": 1.943495049244269,
      "language_loss": 0.86672628,
      "learning_rate": 1.4890506377481392e-07,
      "loss": 0.88828981,
      "num_input_tokens_seen": 158421780,
      "step": 7319,
      "time_per_iteration": 2.7370026111602783
    },
    {
      "auxiliary_loss_clip": 0.01090704,
      "auxiliary_loss_mlp": 0.01023551,
      "balance_loss_clip": 1.03720713,
      "balance_loss_mlp": 1.01742101,
      "epoch": 0.8801779594781458,
      "flos": 23440331439360.0,
      "grad_norm": 1.7030834533872041,
      "language_loss": 0.64356792,
      "learning_rate": 1.486102609102815e-07,
      "loss": 0.66471052,
      "num_input_tokens_seen": 158442330,
      "step": 7320,
      "time_per_iteration": 2.89265513420105
    },
    {
      "auxiliary_loss_clip": 0.01132573,
      "auxiliary_loss_mlp": 0.0102223,
      "balance_loss_clip": 1.04236782,
      "balance_loss_mlp": 1.01546466,
      "epoch": 0.880298202368785,
      "flos": 11508580656000.0,
      "grad_norm": 2.634447978451085,
      "language_loss": 0.85916805,
      "learning_rate": 1.483157389001004e-07,
      "loss": 0.88071609,
      "num_input_tokens_seen": 158459890,
      "step": 7321,
      "time_per_iteration": 2.8615598678588867
    },
    {
      "auxiliary_loss_clip": 0.01136212,
      "auxiliary_loss_mlp": 0.0102064,
      "balance_loss_clip": 1.04082918,
      "balance_loss_mlp": 1.01291561,
      "epoch": 0.880418445259424,
      "flos": 22671447886080.0,
      "grad_norm": 2.0660175655852324,
      "language_loss": 0.79083145,
      "learning_rate": 1.4802149778894933e-07,
      "loss": 0.81239998,
      "num_input_tokens_seen": 158478680,
      "step": 7322,
      "time_per_iteration": 2.7057135105133057
    },
    {
      "auxiliary_loss_clip": 0.01138257,
      "auxiliary_loss_mlp": 0.01023973,
      "balance_loss_clip": 1.03821075,
      "balance_loss_mlp": 1.01768219,
      "epoch": 0.8805386881500631,
      "flos": 20522158709760.0,
      "grad_norm": 1.6663683610417948,
      "language_loss": 0.87769306,
      "learning_rate": 1.4772753762146484e-07,
      "loss": 0.89931536,
      "num_input_tokens_seen": 158497935,
      "step": 7323,
      "time_per_iteration": 2.591421127319336
    },
    {
      "auxiliary_loss_clip": 0.0114452,
      "auxiliary_loss_mlp": 0.01021278,
      "balance_loss_clip": 1.04076314,
      "balance_loss_mlp": 1.01430142,
      "epoch": 0.8806589310407023,
      "flos": 36538891620480.0,
      "grad_norm": 1.581750167520751,
      "language_loss": 0.70491219,
      "learning_rate": 1.474338584422401e-07,
      "loss": 0.72657013,
      "num_input_tokens_seen": 158523145,
      "step": 7324,
      "time_per_iteration": 2.770631790161133
    },
    {
      "auxiliary_loss_clip": 0.01147048,
      "auxiliary_loss_mlp": 0.01022299,
      "balance_loss_clip": 1.04438233,
      "balance_loss_mlp": 1.01538801,
      "epoch": 0.8807791739313413,
      "flos": 23440187784960.0,
      "grad_norm": 3.0589357484176602,
      "language_loss": 0.7582376,
      "learning_rate": 1.4714046029582595e-07,
      "loss": 0.77993107,
      "num_input_tokens_seen": 158542210,
      "step": 7325,
      "time_per_iteration": 2.6987175941467285
    },
    {
      "auxiliary_loss_clip": 0.01125955,
      "auxiliary_loss_mlp": 0.01023172,
      "balance_loss_clip": 1.040025,
      "balance_loss_mlp": 1.01621342,
      "epoch": 0.8808994168219804,
      "flos": 25956843310080.0,
      "grad_norm": 1.7545723290700608,
      "language_loss": 0.75767952,
      "learning_rate": 1.46847343226731e-07,
      "loss": 0.77917081,
      "num_input_tokens_seen": 158563250,
      "step": 7326,
      "time_per_iteration": 2.716244697570801
    },
    {
      "auxiliary_loss_clip": 0.01155225,
      "auxiliary_loss_mlp": 0.010264,
      "balance_loss_clip": 1.04429901,
      "balance_loss_mlp": 1.01941395,
      "epoch": 0.8810196597126195,
      "flos": 17092079303040.0,
      "grad_norm": 2.0978701804729387,
      "language_loss": 0.69515151,
      "learning_rate": 1.465545072794203e-07,
      "loss": 0.7169677,
      "num_input_tokens_seen": 158581125,
      "step": 7327,
      "time_per_iteration": 2.6343047618865967
    },
    {
      "auxiliary_loss_clip": 0.011042,
      "auxiliary_loss_mlp": 0.01022801,
      "balance_loss_clip": 1.04166448,
      "balance_loss_mlp": 1.01592278,
      "epoch": 0.8811399026032586,
      "flos": 23002831785600.0,
      "grad_norm": 1.674097396754825,
      "language_loss": 0.76057196,
      "learning_rate": 1.4626195249831774e-07,
      "loss": 0.78184199,
      "num_input_tokens_seen": 158602025,
      "step": 7328,
      "time_per_iteration": 2.824004650115967
    },
    {
      "auxiliary_loss_clip": 0.01146759,
      "auxiliary_loss_mlp": 0.01025543,
      "balance_loss_clip": 1.04159081,
      "balance_loss_mlp": 1.01844358,
      "epoch": 0.8812601454938976,
      "flos": 14463813242880.0,
      "grad_norm": 1.78945363535501,
      "language_loss": 0.71841609,
      "learning_rate": 1.4596967892780244e-07,
      "loss": 0.74013913,
      "num_input_tokens_seen": 158618355,
      "step": 7329,
      "time_per_iteration": 2.6211071014404297
    },
    {
      "auxiliary_loss_clip": 0.0116262,
      "auxiliary_loss_mlp": 0.01025428,
      "balance_loss_clip": 1.04487669,
      "balance_loss_mlp": 1.01906562,
      "epoch": 0.8813803883845368,
      "flos": 22493223578880.0,
      "grad_norm": 1.6654149330345132,
      "language_loss": 0.74991429,
      "learning_rate": 1.4567768661221314e-07,
      "loss": 0.7717948,
      "num_input_tokens_seen": 158638925,
      "step": 7330,
      "time_per_iteration": 2.602931022644043
    },
    {
      "auxiliary_loss_clip": 0.01153674,
      "auxiliary_loss_mlp": 0.00763165,
      "balance_loss_clip": 1.04493344,
      "balance_loss_mlp": 1.00031853,
      "epoch": 0.8815006312751759,
      "flos": 21506901045120.0,
      "grad_norm": 1.9997754687501357,
      "language_loss": 0.74582636,
      "learning_rate": 1.4538597559584442e-07,
      "loss": 0.76499474,
      "num_input_tokens_seen": 158656715,
      "step": 7331,
      "time_per_iteration": 2.7275524139404297
    },
    {
      "auxiliary_loss_clip": 0.01131506,
      "auxiliary_loss_mlp": 0.01025426,
      "balance_loss_clip": 1.04142213,
      "balance_loss_mlp": 1.01807642,
      "epoch": 0.8816208741658149,
      "flos": 22784566792320.0,
      "grad_norm": 1.7754792428530966,
      "language_loss": 0.78790021,
      "learning_rate": 1.4509454592294823e-07,
      "loss": 0.80946952,
      "num_input_tokens_seen": 158677200,
      "step": 7332,
      "time_per_iteration": 2.748170852661133
    },
    {
      "auxiliary_loss_clip": 0.01125001,
      "auxiliary_loss_mlp": 0.00761745,
      "balance_loss_clip": 1.04391074,
      "balance_loss_mlp": 1.00036526,
      "epoch": 0.8817411170564541,
      "flos": 17779409026560.0,
      "grad_norm": 1.875482080855588,
      "language_loss": 0.78980607,
      "learning_rate": 1.448033976377354e-07,
      "loss": 0.8086735,
      "num_input_tokens_seen": 158692185,
      "step": 7333,
      "time_per_iteration": 3.6027653217315674
    },
    {
      "auxiliary_loss_clip": 0.01151279,
      "auxiliary_loss_mlp": 0.01024031,
      "balance_loss_clip": 1.0421375,
      "balance_loss_mlp": 1.01730752,
      "epoch": 0.8818613599470931,
      "flos": 18551812112640.0,
      "grad_norm": 2.5822906913678882,
      "language_loss": 0.74189335,
      "learning_rate": 1.445125307843713e-07,
      "loss": 0.76364648,
      "num_input_tokens_seen": 158710410,
      "step": 7334,
      "time_per_iteration": 2.6831414699554443
    },
    {
      "auxiliary_loss_clip": 0.0114729,
      "auxiliary_loss_mlp": 0.01022895,
      "balance_loss_clip": 1.04379129,
      "balance_loss_mlp": 1.01598716,
      "epoch": 0.8819816028377322,
      "flos": 27599792417280.0,
      "grad_norm": 1.5681874134326286,
      "language_loss": 0.75440156,
      "learning_rate": 1.442219454069813e-07,
      "loss": 0.77610338,
      "num_input_tokens_seen": 158731435,
      "step": 7335,
      "time_per_iteration": 2.6399829387664795
    },
    {
      "auxiliary_loss_clip": 0.0111172,
      "auxiliary_loss_mlp": 0.01026191,
      "balance_loss_clip": 1.03918386,
      "balance_loss_mlp": 1.01859117,
      "epoch": 0.8821018457283714,
      "flos": 23404600385280.0,
      "grad_norm": 1.805800720670328,
      "language_loss": 0.6637519,
      "learning_rate": 1.4393164154964676e-07,
      "loss": 0.68513095,
      "num_input_tokens_seen": 158750965,
      "step": 7336,
      "time_per_iteration": 2.7591781616210938
    },
    {
      "auxiliary_loss_clip": 0.01148691,
      "auxiliary_loss_mlp": 0.01023947,
      "balance_loss_clip": 1.04549265,
      "balance_loss_mlp": 1.01717901,
      "epoch": 0.8822220886190104,
      "flos": 29132459792640.0,
      "grad_norm": 1.8413131493111239,
      "language_loss": 0.9398514,
      "learning_rate": 1.4364161925640649e-07,
      "loss": 0.96157777,
      "num_input_tokens_seen": 158772365,
      "step": 7337,
      "time_per_iteration": 3.6458804607391357
    },
    {
      "auxiliary_loss_clip": 0.01162947,
      "auxiliary_loss_mlp": 0.01027558,
      "balance_loss_clip": 1.04454565,
      "balance_loss_mlp": 1.02119195,
      "epoch": 0.8823423315096495,
      "flos": 20485422074880.0,
      "grad_norm": 1.8130564564364458,
      "language_loss": 0.85298634,
      "learning_rate": 1.4335187857125663e-07,
      "loss": 0.87489134,
      "num_input_tokens_seen": 158791065,
      "step": 7338,
      "time_per_iteration": 2.6147055625915527
    },
    {
      "auxiliary_loss_clip": 0.01152828,
      "auxiliary_loss_mlp": 0.01022221,
      "balance_loss_clip": 1.04498184,
      "balance_loss_mlp": 1.01513386,
      "epoch": 0.8824625744002886,
      "flos": 24206377818240.0,
      "grad_norm": 2.4640067973388304,
      "language_loss": 0.75609112,
      "learning_rate": 1.4306241953815023e-07,
      "loss": 0.77784157,
      "num_input_tokens_seen": 158812125,
      "step": 7339,
      "time_per_iteration": 3.616642475128174
    },
    {
      "auxiliary_loss_clip": 0.01152349,
      "auxiliary_loss_mlp": 0.01031783,
      "balance_loss_clip": 1.04412913,
      "balance_loss_mlp": 1.02455878,
      "epoch": 0.8825828172909277,
      "flos": 24679500785280.0,
      "grad_norm": 1.6382755399289484,
      "language_loss": 0.70902061,
      "learning_rate": 1.4277324220099862e-07,
      "loss": 0.73086202,
      "num_input_tokens_seen": 158834035,
      "step": 7340,
      "time_per_iteration": 3.551053524017334
    },
    {
      "auxiliary_loss_clip": 0.01116226,
      "auxiliary_loss_mlp": 0.01023607,
      "balance_loss_clip": 1.03718746,
      "balance_loss_mlp": 1.0169518,
      "epoch": 0.8827030601815667,
      "flos": 22456163721600.0,
      "grad_norm": 1.967168713989707,
      "language_loss": 0.74310088,
      "learning_rate": 1.4248434660366938e-07,
      "loss": 0.76449925,
      "num_input_tokens_seen": 158853510,
      "step": 7341,
      "time_per_iteration": 2.743844985961914
    },
    {
      "auxiliary_loss_clip": 0.0113756,
      "auxiliary_loss_mlp": 0.0102826,
      "balance_loss_clip": 1.04477715,
      "balance_loss_mlp": 1.02093506,
      "epoch": 0.8828233030722058,
      "flos": 19865639877120.0,
      "grad_norm": 1.9758013145910316,
      "language_loss": 0.70900273,
      "learning_rate": 1.4219573278998808e-07,
      "loss": 0.73066092,
      "num_input_tokens_seen": 158871970,
      "step": 7342,
      "time_per_iteration": 2.707883596420288
    },
    {
      "auxiliary_loss_clip": 0.01133696,
      "auxiliary_loss_mlp": 0.01023658,
      "balance_loss_clip": 1.04004216,
      "balance_loss_mlp": 1.01608551,
      "epoch": 0.882943545962845,
      "flos": 39347213581440.0,
      "grad_norm": 2.5088557478473077,
      "language_loss": 0.6489374,
      "learning_rate": 1.4190740080373685e-07,
      "loss": 0.67051095,
      "num_input_tokens_seen": 158892250,
      "step": 7343,
      "time_per_iteration": 2.7806074619293213
    },
    {
      "auxiliary_loss_clip": 0.01108178,
      "auxiliary_loss_mlp": 0.01026609,
      "balance_loss_clip": 1.04177892,
      "balance_loss_mlp": 1.01940584,
      "epoch": 0.883063788853484,
      "flos": 19054524908160.0,
      "grad_norm": 1.850247342726599,
      "language_loss": 0.83969223,
      "learning_rate": 1.4161935068865538e-07,
      "loss": 0.86104012,
      "num_input_tokens_seen": 158907395,
      "step": 7344,
      "time_per_iteration": 2.950233221054077
    },
    {
      "auxiliary_loss_clip": 0.01162829,
      "auxiliary_loss_mlp": 0.01025502,
      "balance_loss_clip": 1.04365587,
      "balance_loss_mlp": 1.01830792,
      "epoch": 0.8831840317441231,
      "flos": 18733196816640.0,
      "grad_norm": 2.221890758178015,
      "language_loss": 0.75689912,
      "learning_rate": 1.4133158248844113e-07,
      "loss": 0.77878249,
      "num_input_tokens_seen": 158926300,
      "step": 7345,
      "time_per_iteration": 2.6039254665374756
    },
    {
      "auxiliary_loss_clip": 0.01126108,
      "auxiliary_loss_mlp": 0.01023585,
      "balance_loss_clip": 1.04221749,
      "balance_loss_mlp": 1.01604533,
      "epoch": 0.8833042746347622,
      "flos": 26827712553600.0,
      "grad_norm": 1.7954049940125767,
      "language_loss": 0.73467028,
      "learning_rate": 1.4104409624674785e-07,
      "loss": 0.75616717,
      "num_input_tokens_seen": 158946085,
      "step": 7346,
      "time_per_iteration": 2.768541097640991
    },
    {
      "auxiliary_loss_clip": 0.01152698,
      "auxiliary_loss_mlp": 0.01033663,
      "balance_loss_clip": 1.04663908,
      "balance_loss_mlp": 1.02676678,
      "epoch": 0.8834245175254013,
      "flos": 26104077158400.0,
      "grad_norm": 1.8418977718763025,
      "language_loss": 0.78609478,
      "learning_rate": 1.407568920071873e-07,
      "loss": 0.80795842,
      "num_input_tokens_seen": 158964950,
      "step": 7347,
      "time_per_iteration": 2.709946632385254
    },
    {
      "auxiliary_loss_clip": 0.01169526,
      "auxiliary_loss_mlp": 0.01020615,
      "balance_loss_clip": 1.04660726,
      "balance_loss_mlp": 1.01322377,
      "epoch": 0.8835447604160404,
      "flos": 30629036977920.0,
      "grad_norm": 3.1903350754573427,
      "language_loss": 0.68383771,
      "learning_rate": 1.4046996981332782e-07,
      "loss": 0.70573914,
      "num_input_tokens_seen": 158984835,
      "step": 7348,
      "time_per_iteration": 2.6688780784606934
    },
    {
      "auxiliary_loss_clip": 0.01122275,
      "auxiliary_loss_mlp": 0.01022748,
      "balance_loss_clip": 1.03988028,
      "balance_loss_mlp": 1.01537776,
      "epoch": 0.8836650033066795,
      "flos": 24718356322560.0,
      "grad_norm": 2.1526581492396475,
      "language_loss": 0.78630245,
      "learning_rate": 1.4018332970869516e-07,
      "loss": 0.80775267,
      "num_input_tokens_seen": 159002775,
      "step": 7349,
      "time_per_iteration": 2.7519280910491943
    },
    {
      "auxiliary_loss_clip": 0.01128344,
      "auxiliary_loss_mlp": 0.01023141,
      "balance_loss_clip": 1.04196525,
      "balance_loss_mlp": 1.01639056,
      "epoch": 0.8837852461973186,
      "flos": 25413371556480.0,
      "grad_norm": 1.8540547039930992,
      "language_loss": 0.84751916,
      "learning_rate": 1.398969717367733e-07,
      "loss": 0.86903405,
      "num_input_tokens_seen": 159024100,
      "step": 7350,
      "time_per_iteration": 2.784491539001465
    },
    {
      "auxiliary_loss_clip": 0.01107498,
      "auxiliary_loss_mlp": 0.01025199,
      "balance_loss_clip": 1.04319572,
      "balance_loss_mlp": 1.01899743,
      "epoch": 0.8839054890879576,
      "flos": 17822574195840.0,
      "grad_norm": 1.773543566915141,
      "language_loss": 0.76390219,
      "learning_rate": 1.396108959410014e-07,
      "loss": 0.78522921,
      "num_input_tokens_seen": 159043315,
      "step": 7351,
      "time_per_iteration": 2.7038753032684326
    },
    {
      "auxiliary_loss_clip": 0.01148445,
      "auxiliary_loss_mlp": 0.00762621,
      "balance_loss_clip": 1.04500461,
      "balance_loss_mlp": 1.00032449,
      "epoch": 0.8840257319785968,
      "flos": 23769021818880.0,
      "grad_norm": 1.6672788801270013,
      "language_loss": 0.81081527,
      "learning_rate": 1.3932510236477745e-07,
      "loss": 0.82992595,
      "num_input_tokens_seen": 159063985,
      "step": 7352,
      "time_per_iteration": 2.7274186611175537
    },
    {
      "auxiliary_loss_clip": 0.0114843,
      "auxiliary_loss_mlp": 0.01024221,
      "balance_loss_clip": 1.04169989,
      "balance_loss_mlp": 1.01686621,
      "epoch": 0.8841459748692359,
      "flos": 29059776622080.0,
      "grad_norm": 2.103254462457511,
      "language_loss": 0.56121218,
      "learning_rate": 1.3903959105145636e-07,
      "loss": 0.58293867,
      "num_input_tokens_seen": 159084475,
      "step": 7353,
      "time_per_iteration": 2.6778509616851807
    },
    {
      "auxiliary_loss_clip": 0.01165703,
      "auxiliary_loss_mlp": 0.01023482,
      "balance_loss_clip": 1.04520071,
      "balance_loss_mlp": 1.01664805,
      "epoch": 0.8842662177598749,
      "flos": 24311523905280.0,
      "grad_norm": 2.0493406692423997,
      "language_loss": 0.83196855,
      "learning_rate": 1.387543620443492e-07,
      "loss": 0.8538605,
      "num_input_tokens_seen": 159101320,
      "step": 7354,
      "time_per_iteration": 2.5573666095733643
    },
    {
      "auxiliary_loss_clip": 0.01163211,
      "auxiliary_loss_mlp": 0.01025649,
      "balance_loss_clip": 1.04571044,
      "balance_loss_mlp": 1.01912475,
      "epoch": 0.8843864606505141,
      "flos": 25007867942400.0,
      "grad_norm": 1.7246239052273389,
      "language_loss": 0.84399056,
      "learning_rate": 1.3846941538672606e-07,
      "loss": 0.86587912,
      "num_input_tokens_seen": 159120025,
      "step": 7355,
      "time_per_iteration": 2.562065839767456
    },
    {
      "auxiliary_loss_clip": 0.01113544,
      "auxiliary_loss_mlp": 0.01022807,
      "balance_loss_clip": 1.04273701,
      "balance_loss_mlp": 1.01596439,
      "epoch": 0.8845067035411531,
      "flos": 28183915388160.0,
      "grad_norm": 2.1635010084432094,
      "language_loss": 0.80948305,
      "learning_rate": 1.3818475112181193e-07,
      "loss": 0.83084661,
      "num_input_tokens_seen": 159138820,
      "step": 7356,
      "time_per_iteration": 2.777174472808838
    },
    {
      "auxiliary_loss_clip": 0.01135126,
      "auxiliary_loss_mlp": 0.01019693,
      "balance_loss_clip": 1.04207063,
      "balance_loss_mlp": 1.01291859,
      "epoch": 0.8846269464317922,
      "flos": 12853219311360.0,
      "grad_norm": 2.284630376184178,
      "language_loss": 0.7978726,
      "learning_rate": 1.3790036929279091e-07,
      "loss": 0.8194207,
      "num_input_tokens_seen": 159155975,
      "step": 7357,
      "time_per_iteration": 2.600107192993164
    },
    {
      "auxiliary_loss_clip": 0.011514,
      "auxiliary_loss_mlp": 0.00761869,
      "balance_loss_clip": 1.04509449,
      "balance_loss_mlp": 1.00028908,
      "epoch": 0.8847471893224313,
      "flos": 18624351628800.0,
      "grad_norm": 2.219289943254991,
      "language_loss": 0.58664739,
      "learning_rate": 1.3761626994280363e-07,
      "loss": 0.60578012,
      "num_input_tokens_seen": 159173445,
      "step": 7358,
      "time_per_iteration": 2.617225408554077
    },
    {
      "auxiliary_loss_clip": 0.01127864,
      "auxiliary_loss_mlp": 0.01020232,
      "balance_loss_clip": 1.04177177,
      "balance_loss_mlp": 1.01345778,
      "epoch": 0.8848674322130704,
      "flos": 35769433449600.0,
      "grad_norm": 2.1394910609561,
      "language_loss": 0.7356993,
      "learning_rate": 1.3733245311494735e-07,
      "loss": 0.75718021,
      "num_input_tokens_seen": 159196100,
      "step": 7359,
      "time_per_iteration": 3.7582991123199463
    },
    {
      "auxiliary_loss_clip": 0.01152327,
      "auxiliary_loss_mlp": 0.01026149,
      "balance_loss_clip": 1.04488754,
      "balance_loss_mlp": 1.01926517,
      "epoch": 0.8849876751037095,
      "flos": 24243760897920.0,
      "grad_norm": 2.2790324635833143,
      "language_loss": 0.70360363,
      "learning_rate": 1.3704891885227676e-07,
      "loss": 0.72538841,
      "num_input_tokens_seen": 159216145,
      "step": 7360,
      "time_per_iteration": 2.697330951690674
    },
    {
      "auxiliary_loss_clip": 0.01121819,
      "auxiliary_loss_mlp": 0.01029825,
      "balance_loss_clip": 1.03827369,
      "balance_loss_mlp": 1.02216029,
      "epoch": 0.8851079179943486,
      "flos": 21500580251520.0,
      "grad_norm": 1.8611025010608084,
      "language_loss": 0.78432369,
      "learning_rate": 1.367656671978037e-07,
      "loss": 0.80584019,
      "num_input_tokens_seen": 159233610,
      "step": 7361,
      "time_per_iteration": 2.783442974090576
    },
    {
      "auxiliary_loss_clip": 0.01143257,
      "auxiliary_loss_mlp": 0.01021491,
      "balance_loss_clip": 1.04248333,
      "balance_loss_mlp": 1.01492882,
      "epoch": 0.8852281608849877,
      "flos": 15300711198720.0,
      "grad_norm": 2.010836178451157,
      "language_loss": 0.73684764,
      "learning_rate": 1.36482698194498e-07,
      "loss": 0.75849515,
      "num_input_tokens_seen": 159250155,
      "step": 7362,
      "time_per_iteration": 3.643077850341797
    },
    {
      "auxiliary_loss_clip": 0.01133831,
      "auxiliary_loss_mlp": 0.01023013,
      "balance_loss_clip": 1.03919125,
      "balance_loss_mlp": 1.0158937,
      "epoch": 0.8853484037756267,
      "flos": 23295719283840.0,
      "grad_norm": 1.9332913168759227,
      "language_loss": 0.72161436,
      "learning_rate": 1.3620001188528506e-07,
      "loss": 0.74318278,
      "num_input_tokens_seen": 159270875,
      "step": 7363,
      "time_per_iteration": 2.769676923751831
    },
    {
      "auxiliary_loss_clip": 0.01155121,
      "auxiliary_loss_mlp": 0.01025147,
      "balance_loss_clip": 1.04482973,
      "balance_loss_mlp": 1.01802468,
      "epoch": 0.8854686466662659,
      "flos": 25114773795840.0,
      "grad_norm": 7.312806073921662,
      "language_loss": 0.73645079,
      "learning_rate": 1.3591760831304865e-07,
      "loss": 0.75825351,
      "num_input_tokens_seen": 159288565,
      "step": 7364,
      "time_per_iteration": 3.544142723083496
    },
    {
      "auxiliary_loss_clip": 0.01164339,
      "auxiliary_loss_mlp": 0.0102499,
      "balance_loss_clip": 1.04581857,
      "balance_loss_mlp": 1.01784301,
      "epoch": 0.885588889556905,
      "flos": 21390873137280.0,
      "grad_norm": 1.8225185455261672,
      "language_loss": 0.79230481,
      "learning_rate": 1.356354875206287e-07,
      "loss": 0.81419814,
      "num_input_tokens_seen": 159306400,
      "step": 7365,
      "time_per_iteration": 2.5810978412628174
    },
    {
      "auxiliary_loss_clip": 0.01124068,
      "auxiliary_loss_mlp": 0.0102798,
      "balance_loss_clip": 1.04460168,
      "balance_loss_mlp": 1.02100921,
      "epoch": 0.885709132447544,
      "flos": 26906752431360.0,
      "grad_norm": 1.9013590449566793,
      "language_loss": 0.70093209,
      "learning_rate": 1.3535364955082296e-07,
      "loss": 0.72245252,
      "num_input_tokens_seen": 159326250,
      "step": 7366,
      "time_per_iteration": 2.732025146484375
    },
    {
      "auxiliary_loss_clip": 0.01163115,
      "auxiliary_loss_mlp": 0.01027907,
      "balance_loss_clip": 1.04573703,
      "balance_loss_mlp": 1.02103186,
      "epoch": 0.8858293753381832,
      "flos": 26103394800000.0,
      "grad_norm": 1.8603167090734614,
      "language_loss": 0.64635551,
      "learning_rate": 1.3507209444638613e-07,
      "loss": 0.6682657,
      "num_input_tokens_seen": 159348250,
      "step": 7367,
      "time_per_iteration": 3.595496654510498
    },
    {
      "auxiliary_loss_clip": 0.01149152,
      "auxiliary_loss_mlp": 0.0101986,
      "balance_loss_clip": 1.04338408,
      "balance_loss_mlp": 1.01268101,
      "epoch": 0.8859496182288222,
      "flos": 23292810282240.0,
      "grad_norm": 1.8646209441098294,
      "language_loss": 0.73738873,
      "learning_rate": 1.347908222500298e-07,
      "loss": 0.7590788,
      "num_input_tokens_seen": 159368325,
      "step": 7368,
      "time_per_iteration": 2.643167018890381
    },
    {
      "auxiliary_loss_clip": 0.01107453,
      "auxiliary_loss_mlp": 0.01025938,
      "balance_loss_clip": 1.03955829,
      "balance_loss_mlp": 1.01929832,
      "epoch": 0.8860698611194613,
      "flos": 16872916469760.0,
      "grad_norm": 1.792211356794916,
      "language_loss": 0.69607604,
      "learning_rate": 1.3450983300442276e-07,
      "loss": 0.71740997,
      "num_input_tokens_seen": 159387555,
      "step": 7369,
      "time_per_iteration": 2.6939852237701416
    },
    {
      "auxiliary_loss_clip": 0.01153874,
      "auxiliary_loss_mlp": 0.01022192,
      "balance_loss_clip": 1.04477048,
      "balance_loss_mlp": 1.01558447,
      "epoch": 0.8861901040101005,
      "flos": 24681404206080.0,
      "grad_norm": 1.9348951312408165,
      "language_loss": 0.73389578,
      "learning_rate": 1.3422912675219068e-07,
      "loss": 0.75565648,
      "num_input_tokens_seen": 159407310,
      "step": 7370,
      "time_per_iteration": 2.695908308029175
    },
    {
      "auxiliary_loss_clip": 0.01163779,
      "auxiliary_loss_mlp": 0.01026474,
      "balance_loss_clip": 1.04623079,
      "balance_loss_mlp": 1.02013254,
      "epoch": 0.8863103469007395,
      "flos": 24423026699520.0,
      "grad_norm": 1.7604677451137405,
      "language_loss": 0.790034,
      "learning_rate": 1.339487035359166e-07,
      "loss": 0.81193656,
      "num_input_tokens_seen": 159427680,
      "step": 7371,
      "time_per_iteration": 2.6354246139526367
    },
    {
      "auxiliary_loss_clip": 0.01139103,
      "auxiliary_loss_mlp": 0.0076147,
      "balance_loss_clip": 1.04451656,
      "balance_loss_mlp": 1.00032055,
      "epoch": 0.8864305897913786,
      "flos": 22053964158720.0,
      "grad_norm": 1.5272410789901996,
      "language_loss": 0.84692836,
      "learning_rate": 1.336685633981409e-07,
      "loss": 0.86593407,
      "num_input_tokens_seen": 159448765,
      "step": 7372,
      "time_per_iteration": 2.6439337730407715
    },
    {
      "auxiliary_loss_clip": 0.0115295,
      "auxiliary_loss_mlp": 0.01022818,
      "balance_loss_clip": 1.04420316,
      "balance_loss_mlp": 1.01570153,
      "epoch": 0.8865508326820177,
      "flos": 19099449843840.0,
      "grad_norm": 1.8028232393706307,
      "language_loss": 0.75050825,
      "learning_rate": 1.333887063813597e-07,
      "loss": 0.77226591,
      "num_input_tokens_seen": 159466870,
      "step": 7373,
      "time_per_iteration": 2.6081502437591553
    },
    {
      "auxiliary_loss_clip": 0.01138833,
      "auxiliary_loss_mlp": 0.01023412,
      "balance_loss_clip": 1.04127204,
      "balance_loss_mlp": 1.01663446,
      "epoch": 0.8866710755726568,
      "flos": 15414189240960.0,
      "grad_norm": 1.8511753317731303,
      "language_loss": 0.66225541,
      "learning_rate": 1.331091325280278e-07,
      "loss": 0.68387783,
      "num_input_tokens_seen": 159485840,
      "step": 7374,
      "time_per_iteration": 2.7389426231384277
    },
    {
      "auxiliary_loss_clip": 0.01099943,
      "auxiliary_loss_mlp": 0.01024738,
      "balance_loss_clip": 1.03805161,
      "balance_loss_mlp": 1.01729333,
      "epoch": 0.8867913184632958,
      "flos": 20083689388800.0,
      "grad_norm": 1.6701040773265599,
      "language_loss": 0.78623968,
      "learning_rate": 1.3282984188055625e-07,
      "loss": 0.80748653,
      "num_input_tokens_seen": 159505630,
      "step": 7375,
      "time_per_iteration": 2.795354127883911
    },
    {
      "auxiliary_loss_clip": 0.01164006,
      "auxiliary_loss_mlp": 0.01023322,
      "balance_loss_clip": 1.0448277,
      "balance_loss_mlp": 1.0162828,
      "epoch": 0.8869115613539349,
      "flos": 23365852588800.0,
      "grad_norm": 2.028403330495967,
      "language_loss": 0.79512191,
      "learning_rate": 1.3255083448131288e-07,
      "loss": 0.8169952,
      "num_input_tokens_seen": 159524675,
      "step": 7376,
      "time_per_iteration": 2.6071555614471436
    },
    {
      "auxiliary_loss_clip": 0.01153872,
      "auxiliary_loss_mlp": 0.01024493,
      "balance_loss_clip": 1.04262555,
      "balance_loss_mlp": 1.01771879,
      "epoch": 0.8870318042445741,
      "flos": 21286840371840.0,
      "grad_norm": 1.97674660023906,
      "language_loss": 0.79323483,
      "learning_rate": 1.3227211037262365e-07,
      "loss": 0.81501853,
      "num_input_tokens_seen": 159541915,
      "step": 7377,
      "time_per_iteration": 2.6521496772766113
    },
    {
      "auxiliary_loss_clip": 0.01111302,
      "auxiliary_loss_mlp": 0.01025049,
      "balance_loss_clip": 1.03887558,
      "balance_loss_mlp": 1.01797354,
      "epoch": 0.8871520471352131,
      "flos": 20010862563840.0,
      "grad_norm": 2.3412854355689694,
      "language_loss": 0.8546561,
      "learning_rate": 1.319936695967696e-07,
      "loss": 0.8760196,
      "num_input_tokens_seen": 159559740,
      "step": 7378,
      "time_per_iteration": 2.7119977474212646
    },
    {
      "auxiliary_loss_clip": 0.01170949,
      "auxiliary_loss_mlp": 0.01026749,
      "balance_loss_clip": 1.04608703,
      "balance_loss_mlp": 1.01923025,
      "epoch": 0.8872722900258522,
      "flos": 22601422321920.0,
      "grad_norm": 2.4394303057842177,
      "language_loss": 0.82174218,
      "learning_rate": 1.3171551219599097e-07,
      "loss": 0.84371918,
      "num_input_tokens_seen": 159578265,
      "step": 7379,
      "time_per_iteration": 2.5848402976989746
    },
    {
      "auxiliary_loss_clip": 0.01167089,
      "auxiliary_loss_mlp": 0.01029529,
      "balance_loss_clip": 1.04820573,
      "balance_loss_mlp": 1.02212608,
      "epoch": 0.8873925329164913,
      "flos": 22163276223360.0,
      "grad_norm": 2.2537615883002973,
      "language_loss": 0.78105557,
      "learning_rate": 1.3143763821248377e-07,
      "loss": 0.80302179,
      "num_input_tokens_seen": 159595350,
      "step": 7380,
      "time_per_iteration": 2.603397846221924
    },
    {
      "auxiliary_loss_clip": 0.01163759,
      "auxiliary_loss_mlp": 0.01023578,
      "balance_loss_clip": 1.04611015,
      "balance_loss_mlp": 1.01701295,
      "epoch": 0.8875127758071304,
      "flos": 19208223204480.0,
      "grad_norm": 1.865175826780953,
      "language_loss": 0.72029305,
      "learning_rate": 1.3116004768840118e-07,
      "loss": 0.74216646,
      "num_input_tokens_seen": 159613725,
      "step": 7381,
      "time_per_iteration": 2.6400012969970703
    },
    {
      "auxiliary_loss_clip": 0.01165454,
      "auxiliary_loss_mlp": 0.0102699,
      "balance_loss_clip": 1.04528236,
      "balance_loss_mlp": 1.01995969,
      "epoch": 0.8876330186977694,
      "flos": 18110900666880.0,
      "grad_norm": 1.673894447149643,
      "language_loss": 0.74009037,
      "learning_rate": 1.3088274066585348e-07,
      "loss": 0.76201487,
      "num_input_tokens_seen": 159631335,
      "step": 7382,
      "time_per_iteration": 2.6107428073883057
    },
    {
      "auxiliary_loss_clip": 0.01126876,
      "auxiliary_loss_mlp": 0.01022011,
      "balance_loss_clip": 1.04068291,
      "balance_loss_mlp": 1.01486993,
      "epoch": 0.8877532615884086,
      "flos": 22009434272640.0,
      "grad_norm": 2.2974769830762467,
      "language_loss": 0.90249467,
      "learning_rate": 1.3060571718690749e-07,
      "loss": 0.92398351,
      "num_input_tokens_seen": 159648830,
      "step": 7383,
      "time_per_iteration": 2.7659108638763428
    },
    {
      "auxiliary_loss_clip": 0.01035105,
      "auxiliary_loss_mlp": 0.00753812,
      "balance_loss_clip": 1.00983679,
      "balance_loss_mlp": 1.00020826,
      "epoch": 0.8878735044790477,
      "flos": 72136924346880.0,
      "grad_norm": 0.787896062108697,
      "language_loss": 0.56860304,
      "learning_rate": 1.3032897729358805e-07,
      "loss": 0.58649218,
      "num_input_tokens_seen": 159709785,
      "step": 7384,
      "time_per_iteration": 3.31854248046875
    },
    {
      "auxiliary_loss_clip": 0.0108274,
      "auxiliary_loss_mlp": 0.00762478,
      "balance_loss_clip": 1.03461778,
      "balance_loss_mlp": 1.00040317,
      "epoch": 0.8879937473696867,
      "flos": 27526355061120.0,
      "grad_norm": 2.6961100516643857,
      "language_loss": 0.80360633,
      "learning_rate": 1.3005252102787645e-07,
      "loss": 0.8220585,
      "num_input_tokens_seen": 159728725,
      "step": 7385,
      "time_per_iteration": 3.675999164581299
    },
    {
      "auxiliary_loss_clip": 0.01154768,
      "auxiliary_loss_mlp": 0.01022629,
      "balance_loss_clip": 1.0451169,
      "balance_loss_mlp": 1.01560116,
      "epoch": 0.8881139902603259,
      "flos": 22234091886720.0,
      "grad_norm": 1.5266200932494387,
      "language_loss": 0.73249799,
      "learning_rate": 1.297763484317105e-07,
      "loss": 0.75427192,
      "num_input_tokens_seen": 159747020,
      "step": 7386,
      "time_per_iteration": 2.6809632778167725
    },
    {
      "auxiliary_loss_clip": 0.01105892,
      "auxiliary_loss_mlp": 0.00762652,
      "balance_loss_clip": 1.0379231,
      "balance_loss_mlp": 1.00037611,
      "epoch": 0.888234233150965,
      "flos": 20299548170880.0,
      "grad_norm": 2.3694038341931183,
      "language_loss": 0.70259523,
      "learning_rate": 1.2950045954698551e-07,
      "loss": 0.72128069,
      "num_input_tokens_seen": 159764855,
      "step": 7387,
      "time_per_iteration": 2.7491867542266846
    },
    {
      "auxiliary_loss_clip": 0.0111607,
      "auxiliary_loss_mlp": 0.01025384,
      "balance_loss_clip": 1.04188037,
      "balance_loss_mlp": 1.0185746,
      "epoch": 0.888354476041604,
      "flos": 18147996437760.0,
      "grad_norm": 1.6221960725881368,
      "language_loss": 0.7580508,
      "learning_rate": 1.2922485441555343e-07,
      "loss": 0.77946532,
      "num_input_tokens_seen": 159783935,
      "step": 7388,
      "time_per_iteration": 3.7265148162841797
    },
    {
      "auxiliary_loss_clip": 0.01163191,
      "auxiliary_loss_mlp": 0.01022048,
      "balance_loss_clip": 1.0442189,
      "balance_loss_mlp": 1.0156343,
      "epoch": 0.8884747189322432,
      "flos": 22014282608640.0,
      "grad_norm": 1.7435569515816058,
      "language_loss": 0.81899333,
      "learning_rate": 1.2894953307922363e-07,
      "loss": 0.8408457,
      "num_input_tokens_seen": 159802895,
      "step": 7389,
      "time_per_iteration": 2.636754035949707
    },
    {
      "auxiliary_loss_clip": 0.01116072,
      "auxiliary_loss_mlp": 0.01024128,
      "balance_loss_clip": 1.03983819,
      "balance_loss_mlp": 1.0171634,
      "epoch": 0.8885949618228822,
      "flos": 19786779567360.0,
      "grad_norm": 3.3929040804211277,
      "language_loss": 0.84267282,
      "learning_rate": 1.2867449557976208e-07,
      "loss": 0.86407483,
      "num_input_tokens_seen": 159820995,
      "step": 7390,
      "time_per_iteration": 2.7651143074035645
    },
    {
      "auxiliary_loss_clip": 0.01149059,
      "auxiliary_loss_mlp": 0.01025448,
      "balance_loss_clip": 1.04450893,
      "balance_loss_mlp": 1.01807213,
      "epoch": 0.8887152047135213,
      "flos": 20047599198720.0,
      "grad_norm": 1.9039674970027378,
      "language_loss": 0.75708103,
      "learning_rate": 1.283997419588916e-07,
      "loss": 0.77882612,
      "num_input_tokens_seen": 159840465,
      "step": 7391,
      "time_per_iteration": 3.5311691761016846
    },
    {
      "auxiliary_loss_clip": 0.01153123,
      "auxiliary_loss_mlp": 0.01023488,
      "balance_loss_clip": 1.04279423,
      "balance_loss_mlp": 1.01626086,
      "epoch": 0.8888354476041604,
      "flos": 18588117784320.0,
      "grad_norm": 1.9259934589587973,
      "language_loss": 0.62112403,
      "learning_rate": 1.2812527225829216e-07,
      "loss": 0.64289016,
      "num_input_tokens_seen": 159858690,
      "step": 7392,
      "time_per_iteration": 2.6355276107788086
    },
    {
      "auxiliary_loss_clip": 0.01156231,
      "auxiliary_loss_mlp": 0.01030551,
      "balance_loss_clip": 1.04504287,
      "balance_loss_mlp": 1.02259946,
      "epoch": 0.8889556904947995,
      "flos": 21689794120320.0,
      "grad_norm": 1.937035248817835,
      "language_loss": 0.76479155,
      "learning_rate": 1.2785108651960052e-07,
      "loss": 0.78665936,
      "num_input_tokens_seen": 159880325,
      "step": 7393,
      "time_per_iteration": 3.454632520675659
    },
    {
      "auxiliary_loss_clip": 0.01151753,
      "auxiliary_loss_mlp": 0.01024198,
      "balance_loss_clip": 1.04273808,
      "balance_loss_mlp": 1.01742685,
      "epoch": 0.8890759333854386,
      "flos": 27381204201600.0,
      "grad_norm": 2.304090441109236,
      "language_loss": 0.8080591,
      "learning_rate": 1.2757718478441094e-07,
      "loss": 0.82981861,
      "num_input_tokens_seen": 159901070,
      "step": 7394,
      "time_per_iteration": 2.694096088409424
    },
    {
      "auxiliary_loss_clip": 0.01133868,
      "auxiliary_loss_mlp": 0.01026454,
      "balance_loss_clip": 1.04091358,
      "balance_loss_mlp": 1.0193193,
      "epoch": 0.8891961762760777,
      "flos": 24498834353280.0,
      "grad_norm": 2.116831459498727,
      "language_loss": 0.77399015,
      "learning_rate": 1.2730356709427302e-07,
      "loss": 0.79559332,
      "num_input_tokens_seen": 159919750,
      "step": 7395,
      "time_per_iteration": 2.6911487579345703
    },
    {
      "auxiliary_loss_clip": 0.01148982,
      "auxiliary_loss_mlp": 0.01027635,
      "balance_loss_clip": 1.04610872,
      "balance_loss_mlp": 1.02071774,
      "epoch": 0.8893164191667168,
      "flos": 41499770895360.0,
      "grad_norm": 1.643883142257371,
      "language_loss": 0.59863508,
      "learning_rate": 1.2703023349069542e-07,
      "loss": 0.62040126,
      "num_input_tokens_seen": 159944600,
      "step": 7396,
      "time_per_iteration": 2.861041307449341
    },
    {
      "auxiliary_loss_clip": 0.01145512,
      "auxiliary_loss_mlp": 0.01027683,
      "balance_loss_clip": 1.04308057,
      "balance_loss_mlp": 1.02069163,
      "epoch": 0.8894366620573558,
      "flos": 33583623120000.0,
      "grad_norm": 1.8091600979618832,
      "language_loss": 0.61711574,
      "learning_rate": 1.2675718401514223e-07,
      "loss": 0.63884765,
      "num_input_tokens_seen": 159968780,
      "step": 7397,
      "time_per_iteration": 2.697268486022949
    },
    {
      "auxiliary_loss_clip": 0.01136209,
      "auxiliary_loss_mlp": 0.01023013,
      "balance_loss_clip": 1.04240656,
      "balance_loss_mlp": 1.01581025,
      "epoch": 0.889556904947995,
      "flos": 16909832672640.0,
      "grad_norm": 2.245038101603853,
      "language_loss": 0.74476713,
      "learning_rate": 1.264844187090346e-07,
      "loss": 0.76635939,
      "num_input_tokens_seen": 159985905,
      "step": 7398,
      "time_per_iteration": 2.630293607711792
    },
    {
      "auxiliary_loss_clip": 0.01130249,
      "auxiliary_loss_mlp": 0.0102663,
      "balance_loss_clip": 1.0392344,
      "balance_loss_mlp": 1.01960266,
      "epoch": 0.889677147838634,
      "flos": 26030855283840.0,
      "grad_norm": 1.6820301144502565,
      "language_loss": 0.75320792,
      "learning_rate": 1.262119376137516e-07,
      "loss": 0.7747767,
      "num_input_tokens_seen": 160006965,
      "step": 7399,
      "time_per_iteration": 2.688326120376587
    },
    {
      "auxiliary_loss_clip": 0.01140018,
      "auxiliary_loss_mlp": 0.01019359,
      "balance_loss_clip": 1.03986168,
      "balance_loss_mlp": 1.01251626,
      "epoch": 0.8897973907292731,
      "flos": 26468283110400.0,
      "grad_norm": 1.6443331574761637,
      "language_loss": 0.84982324,
      "learning_rate": 1.2593974077062707e-07,
      "loss": 0.87141699,
      "num_input_tokens_seen": 160028585,
      "step": 7400,
      "time_per_iteration": 2.6313655376434326
    },
    {
      "auxiliary_loss_clip": 0.01114415,
      "auxiliary_loss_mlp": 0.01024972,
      "balance_loss_clip": 1.04022384,
      "balance_loss_mlp": 1.01753306,
      "epoch": 0.8899176336199123,
      "flos": 26249694894720.0,
      "grad_norm": 2.007878851886122,
      "language_loss": 0.6344924,
      "learning_rate": 1.2566782822095423e-07,
      "loss": 0.65588617,
      "num_input_tokens_seen": 160048840,
      "step": 7401,
      "time_per_iteration": 2.6943907737731934
    },
    {
      "auxiliary_loss_clip": 0.01128924,
      "auxiliary_loss_mlp": 0.01023161,
      "balance_loss_clip": 1.04418802,
      "balance_loss_mlp": 1.01629734,
      "epoch": 0.8900378765105513,
      "flos": 20811742156800.0,
      "grad_norm": 2.688259023987226,
      "language_loss": 0.7130596,
      "learning_rate": 1.2539620000598162e-07,
      "loss": 0.73458046,
      "num_input_tokens_seen": 160068175,
      "step": 7402,
      "time_per_iteration": 2.7264931201934814
    },
    {
      "auxiliary_loss_clip": 0.01162421,
      "auxiliary_loss_mlp": 0.01023305,
      "balance_loss_clip": 1.04402065,
      "balance_loss_mlp": 1.01592922,
      "epoch": 0.8901581194011904,
      "flos": 16472333018880.0,
      "grad_norm": 1.8415711332903504,
      "language_loss": 0.79644769,
      "learning_rate": 1.2512485616691492e-07,
      "loss": 0.81830502,
      "num_input_tokens_seen": 160085230,
      "step": 7403,
      "time_per_iteration": 2.6056599617004395
    },
    {
      "auxiliary_loss_clip": 0.01126001,
      "auxiliary_loss_mlp": 0.01031322,
      "balance_loss_clip": 1.04224563,
      "balance_loss_mlp": 1.02393162,
      "epoch": 0.8902783622918296,
      "flos": 35155253773440.0,
      "grad_norm": 1.4241507269050806,
      "language_loss": 0.80785096,
      "learning_rate": 1.2485379674491681e-07,
      "loss": 0.8294242,
      "num_input_tokens_seen": 160111425,
      "step": 7404,
      "time_per_iteration": 2.8429160118103027
    },
    {
      "auxiliary_loss_clip": 0.0113867,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.04542136,
      "balance_loss_mlp": 1.0202992,
      "epoch": 0.8903986051824686,
      "flos": 17201068145280.0,
      "grad_norm": 2.5302309011058313,
      "language_loss": 0.79112697,
      "learning_rate": 1.2458302178110657e-07,
      "loss": 0.81279093,
      "num_input_tokens_seen": 160129790,
      "step": 7405,
      "time_per_iteration": 2.681103229522705
    },
    {
      "auxiliary_loss_clip": 0.0110937,
      "auxiliary_loss_mlp": 0.01021859,
      "balance_loss_clip": 1.03756368,
      "balance_loss_mlp": 1.01519537,
      "epoch": 0.8905188480731077,
      "flos": 25483863997440.0,
      "grad_norm": 2.4855416708025064,
      "language_loss": 0.82375866,
      "learning_rate": 1.2431253131656118e-07,
      "loss": 0.8450709,
      "num_input_tokens_seen": 160149265,
      "step": 7406,
      "time_per_iteration": 2.7609238624572754
    },
    {
      "auxiliary_loss_clip": 0.01130915,
      "auxiliary_loss_mlp": 0.01020843,
      "balance_loss_clip": 1.04296982,
      "balance_loss_mlp": 1.01394665,
      "epoch": 0.8906390909637467,
      "flos": 23365888502400.0,
      "grad_norm": 2.1644730437035204,
      "language_loss": 0.76758629,
      "learning_rate": 1.240423253923133e-07,
      "loss": 0.78910387,
      "num_input_tokens_seen": 160168870,
      "step": 7407,
      "time_per_iteration": 2.6819076538085938
    },
    {
      "auxiliary_loss_clip": 0.0115305,
      "auxiliary_loss_mlp": 0.0102611,
      "balance_loss_clip": 1.04401493,
      "balance_loss_mlp": 1.01913679,
      "epoch": 0.8907593338543859,
      "flos": 21068790860160.0,
      "grad_norm": 8.16772683438691,
      "language_loss": 0.69314778,
      "learning_rate": 1.237724040493533e-07,
      "loss": 0.71493936,
      "num_input_tokens_seen": 160187495,
      "step": 7408,
      "time_per_iteration": 2.6372807025909424
    },
    {
      "auxiliary_loss_clip": 0.01170458,
      "auxiliary_loss_mlp": 0.01027479,
      "balance_loss_clip": 1.04835653,
      "balance_loss_mlp": 1.01962876,
      "epoch": 0.8908795767450249,
      "flos": 21869562712320.0,
      "grad_norm": 2.812310645491133,
      "language_loss": 0.73079407,
      "learning_rate": 1.2350276732862773e-07,
      "loss": 0.75277352,
      "num_input_tokens_seen": 160208520,
      "step": 7409,
      "time_per_iteration": 2.6081199645996094
    },
    {
      "auxiliary_loss_clip": 0.01054872,
      "auxiliary_loss_mlp": 0.010014,
      "balance_loss_clip": 1.00972545,
      "balance_loss_mlp": 1.00033271,
      "epoch": 0.890999819635664,
      "flos": 66307869348480.0,
      "grad_norm": 0.847639537069281,
      "language_loss": 0.56649709,
      "learning_rate": 1.2323341527103993e-07,
      "loss": 0.58705974,
      "num_input_tokens_seen": 160263720,
      "step": 7410,
      "time_per_iteration": 3.202141761779785
    },
    {
      "auxiliary_loss_clip": 0.01162537,
      "auxiliary_loss_mlp": 0.01023133,
      "balance_loss_clip": 1.04449892,
      "balance_loss_mlp": 1.01674032,
      "epoch": 0.8911200625263032,
      "flos": 26869908055680.0,
      "grad_norm": 2.34894650105175,
      "language_loss": 0.85175741,
      "learning_rate": 1.2296434791745135e-07,
      "loss": 0.87361413,
      "num_input_tokens_seen": 160282170,
      "step": 7411,
      "time_per_iteration": 3.529440402984619
    },
    {
      "auxiliary_loss_clip": 0.01154273,
      "auxiliary_loss_mlp": 0.01020587,
      "balance_loss_clip": 1.04659677,
      "balance_loss_mlp": 1.01366103,
      "epoch": 0.8912403054169422,
      "flos": 20885825957760.0,
      "grad_norm": 2.073090076347989,
      "language_loss": 0.76717162,
      "learning_rate": 1.2269556530867875e-07,
      "loss": 0.78892016,
      "num_input_tokens_seen": 160300725,
      "step": 7412,
      "time_per_iteration": 2.6516244411468506
    },
    {
      "auxiliary_loss_clip": 0.01172062,
      "auxiliary_loss_mlp": 0.01030184,
      "balance_loss_clip": 1.04812431,
      "balance_loss_mlp": 1.0220691,
      "epoch": 0.8913605483075813,
      "flos": 27016567286400.0,
      "grad_norm": 2.8788792592715797,
      "language_loss": 0.81963843,
      "learning_rate": 1.2242706748549614e-07,
      "loss": 0.84166086,
      "num_input_tokens_seen": 160318720,
      "step": 7413,
      "time_per_iteration": 2.666343927383423
    },
    {
      "auxiliary_loss_clip": 0.01135986,
      "auxiliary_loss_mlp": 0.01023411,
      "balance_loss_clip": 1.03840721,
      "balance_loss_mlp": 1.01655602,
      "epoch": 0.8914807911982204,
      "flos": 23621500661760.0,
      "grad_norm": 4.793369530617937,
      "language_loss": 0.82423884,
      "learning_rate": 1.2215885448863473e-07,
      "loss": 0.84583282,
      "num_input_tokens_seen": 160339595,
      "step": 7414,
      "time_per_iteration": 3.8039047718048096
    },
    {
      "auxiliary_loss_clip": 0.01137541,
      "auxiliary_loss_mlp": 0.01024642,
      "balance_loss_clip": 1.04370093,
      "balance_loss_mlp": 1.01774871,
      "epoch": 0.8916010340888595,
      "flos": 24462277286400.0,
      "grad_norm": 5.219229518513067,
      "language_loss": 0.80634522,
      "learning_rate": 1.2189092635878152e-07,
      "loss": 0.82796705,
      "num_input_tokens_seen": 160361045,
      "step": 7415,
      "time_per_iteration": 2.712311029434204
    },
    {
      "auxiliary_loss_clip": 0.01113601,
      "auxiliary_loss_mlp": 0.01025452,
      "balance_loss_clip": 1.03950107,
      "balance_loss_mlp": 1.01832342,
      "epoch": 0.8917212769794985,
      "flos": 21215773313280.0,
      "grad_norm": 1.6380240561103458,
      "language_loss": 0.77288592,
      "learning_rate": 1.216232831365822e-07,
      "loss": 0.79427648,
      "num_input_tokens_seen": 160379990,
      "step": 7416,
      "time_per_iteration": 2.7154834270477295
    },
    {
      "auxiliary_loss_clip": 0.01142639,
      "auxiliary_loss_mlp": 0.01025782,
      "balance_loss_clip": 1.04339504,
      "balance_loss_mlp": 1.01806271,
      "epoch": 0.8918415198701377,
      "flos": 25513992529920.0,
      "grad_norm": 1.7292768916864134,
      "language_loss": 0.80725074,
      "learning_rate": 1.2135592486263678e-07,
      "loss": 0.82893491,
      "num_input_tokens_seen": 160399240,
      "step": 7417,
      "time_per_iteration": 3.607553243637085
    },
    {
      "auxiliary_loss_clip": 0.01134989,
      "auxiliary_loss_mlp": 0.0102213,
      "balance_loss_clip": 1.0419867,
      "balance_loss_mlp": 1.01518881,
      "epoch": 0.8919617627607768,
      "flos": 37853006693760.0,
      "grad_norm": 1.7062655248543732,
      "language_loss": 0.61033928,
      "learning_rate": 1.2108885157750415e-07,
      "loss": 0.63191044,
      "num_input_tokens_seen": 160421600,
      "step": 7418,
      "time_per_iteration": 3.7237048149108887
    },
    {
      "auxiliary_loss_clip": 0.01117704,
      "auxiliary_loss_mlp": 0.0076164,
      "balance_loss_clip": 1.04309857,
      "balance_loss_mlp": 1.00030851,
      "epoch": 0.8920820056514158,
      "flos": 26213676531840.0,
      "grad_norm": 1.865435452240683,
      "language_loss": 0.80500793,
      "learning_rate": 1.2082206332169897e-07,
      "loss": 0.8238014,
      "num_input_tokens_seen": 160441695,
      "step": 7419,
      "time_per_iteration": 2.8149659633636475
    },
    {
      "auxiliary_loss_clip": 0.01133172,
      "auxiliary_loss_mlp": 0.0102642,
      "balance_loss_clip": 1.04403472,
      "balance_loss_mlp": 1.01980984,
      "epoch": 0.892202248542055,
      "flos": 17383135207680.0,
      "grad_norm": 2.4215260767518427,
      "language_loss": 0.73088449,
      "learning_rate": 1.2055556013569225e-07,
      "loss": 0.75248039,
      "num_input_tokens_seen": 160457205,
      "step": 7420,
      "time_per_iteration": 2.627798318862915
    },
    {
      "auxiliary_loss_clip": 0.01138333,
      "auxiliary_loss_mlp": 0.0102217,
      "balance_loss_clip": 1.04181409,
      "balance_loss_mlp": 1.01458836,
      "epoch": 0.892322491432694,
      "flos": 21324223451520.0,
      "grad_norm": 1.7985816818642788,
      "language_loss": 0.82247329,
      "learning_rate": 1.2028934205991315e-07,
      "loss": 0.84407836,
      "num_input_tokens_seen": 160476525,
      "step": 7421,
      "time_per_iteration": 2.6140503883361816
    },
    {
      "auxiliary_loss_clip": 0.0114898,
      "auxiliary_loss_mlp": 0.01028081,
      "balance_loss_clip": 1.04147053,
      "balance_loss_mlp": 1.02134538,
      "epoch": 0.8924427343233331,
      "flos": 24029374573440.0,
      "grad_norm": 1.4390806320399865,
      "language_loss": 0.76475668,
      "learning_rate": 1.2002340913474607e-07,
      "loss": 0.78652728,
      "num_input_tokens_seen": 160500160,
      "step": 7422,
      "time_per_iteration": 2.706087589263916
    },
    {
      "auxiliary_loss_clip": 0.01168299,
      "auxiliary_loss_mlp": 0.01022978,
      "balance_loss_clip": 1.04721308,
      "balance_loss_mlp": 1.01535439,
      "epoch": 0.8925629772139723,
      "flos": 30008069631360.0,
      "grad_norm": 1.9321722516506206,
      "language_loss": 0.7388947,
      "learning_rate": 1.1975776140053317e-07,
      "loss": 0.76080751,
      "num_input_tokens_seen": 160520130,
      "step": 7423,
      "time_per_iteration": 2.659843683242798
    },
    {
      "auxiliary_loss_clip": 0.01110478,
      "auxiliary_loss_mlp": 0.01025422,
      "balance_loss_clip": 1.03987348,
      "balance_loss_mlp": 1.01771522,
      "epoch": 0.8926832201046113,
      "flos": 22601709630720.0,
      "grad_norm": 3.940942586809801,
      "language_loss": 0.73182666,
      "learning_rate": 1.194923988975729e-07,
      "loss": 0.75318563,
      "num_input_tokens_seen": 160539730,
      "step": 7424,
      "time_per_iteration": 2.7696735858917236
    },
    {
      "auxiliary_loss_clip": 0.0111957,
      "auxiliary_loss_mlp": 0.01020564,
      "balance_loss_clip": 1.04199338,
      "balance_loss_mlp": 1.01338124,
      "epoch": 0.8928034629952504,
      "flos": 13297722117120.0,
      "grad_norm": 3.144243989824138,
      "language_loss": 0.73608142,
      "learning_rate": 1.192273216661206e-07,
      "loss": 0.75748277,
      "num_input_tokens_seen": 160557820,
      "step": 7425,
      "time_per_iteration": 2.665731906890869
    },
    {
      "auxiliary_loss_clip": 0.01011678,
      "auxiliary_loss_mlp": 0.01000526,
      "balance_loss_clip": 1.01052523,
      "balance_loss_mlp": 0.99945349,
      "epoch": 0.8929237058858895,
      "flos": 54854556744960.0,
      "grad_norm": 0.763635780715979,
      "language_loss": 0.57453543,
      "learning_rate": 1.189625297463881e-07,
      "loss": 0.59465742,
      "num_input_tokens_seen": 160619510,
      "step": 7426,
      "time_per_iteration": 3.4184765815734863
    },
    {
      "auxiliary_loss_clip": 0.01091492,
      "auxiliary_loss_mlp": 0.0102441,
      "balance_loss_clip": 1.03698611,
      "balance_loss_mlp": 1.01784754,
      "epoch": 0.8930439487765286,
      "flos": 28883850785280.0,
      "grad_norm": 1.749773550830242,
      "language_loss": 0.79909915,
      "learning_rate": 1.1869802317854394e-07,
      "loss": 0.8202582,
      "num_input_tokens_seen": 160643295,
      "step": 7427,
      "time_per_iteration": 3.100395441055298
    },
    {
      "auxiliary_loss_clip": 0.0111043,
      "auxiliary_loss_mlp": 0.01023538,
      "balance_loss_clip": 1.03865194,
      "balance_loss_mlp": 1.01637614,
      "epoch": 0.8931641916671677,
      "flos": 22419283432320.0,
      "grad_norm": 2.0612789524427924,
      "language_loss": 0.72209048,
      "learning_rate": 1.1843380200271425e-07,
      "loss": 0.74343014,
      "num_input_tokens_seen": 160662495,
      "step": 7428,
      "time_per_iteration": 2.710832118988037
    },
    {
      "auxiliary_loss_clip": 0.01118938,
      "auxiliary_loss_mlp": 0.01027496,
      "balance_loss_clip": 1.04115832,
      "balance_loss_mlp": 1.01953006,
      "epoch": 0.8932844345578068,
      "flos": 25843149786240.0,
      "grad_norm": 3.385117097313365,
      "language_loss": 0.80501288,
      "learning_rate": 1.181698662589805e-07,
      "loss": 0.82647723,
      "num_input_tokens_seen": 160682080,
      "step": 7429,
      "time_per_iteration": 2.701418161392212
    },
    {
      "auxiliary_loss_clip": 0.0114798,
      "auxiliary_loss_mlp": 0.01028463,
      "balance_loss_clip": 1.04255867,
      "balance_loss_mlp": 1.02099442,
      "epoch": 0.8934046774484459,
      "flos": 22925803069440.0,
      "grad_norm": 2.6514332249491543,
      "language_loss": 0.7601729,
      "learning_rate": 1.1790621598738249e-07,
      "loss": 0.78193736,
      "num_input_tokens_seen": 160700395,
      "step": 7430,
      "time_per_iteration": 2.7431771755218506
    },
    {
      "auxiliary_loss_clip": 0.01163087,
      "auxiliary_loss_mlp": 0.010231,
      "balance_loss_clip": 1.04709959,
      "balance_loss_mlp": 1.01661468,
      "epoch": 0.8935249203390849,
      "flos": 24462097718400.0,
      "grad_norm": 2.788430681484839,
      "language_loss": 0.74693847,
      "learning_rate": 1.1764285122791461e-07,
      "loss": 0.76880032,
      "num_input_tokens_seen": 160721115,
      "step": 7431,
      "time_per_iteration": 2.679210662841797
    },
    {
      "auxiliary_loss_clip": 0.01152079,
      "auxiliary_loss_mlp": 0.01024578,
      "balance_loss_clip": 1.04295921,
      "balance_loss_mlp": 1.01802993,
      "epoch": 0.8936451632297241,
      "flos": 15742735966080.0,
      "grad_norm": 1.8504831465312919,
      "language_loss": 0.76830089,
      "learning_rate": 1.173797720205294e-07,
      "loss": 0.79006743,
      "num_input_tokens_seen": 160739150,
      "step": 7432,
      "time_per_iteration": 2.5991640090942383
    },
    {
      "auxiliary_loss_clip": 0.01155254,
      "auxiliary_loss_mlp": 0.01028444,
      "balance_loss_clip": 1.04613972,
      "balance_loss_mlp": 1.02070713,
      "epoch": 0.8937654061203631,
      "flos": 35115500396160.0,
      "grad_norm": 2.178020690475307,
      "language_loss": 0.71372533,
      "learning_rate": 1.1711697840513602e-07,
      "loss": 0.73556232,
      "num_input_tokens_seen": 160758585,
      "step": 7433,
      "time_per_iteration": 2.7459726333618164
    },
    {
      "auxiliary_loss_clip": 0.01142382,
      "auxiliary_loss_mlp": 0.01020416,
      "balance_loss_clip": 1.04116237,
      "balance_loss_mlp": 1.01390743,
      "epoch": 0.8938856490110022,
      "flos": 16107444708480.0,
      "grad_norm": 1.9325758414998935,
      "language_loss": 0.71199518,
      "learning_rate": 1.1685447042160012e-07,
      "loss": 0.73362321,
      "num_input_tokens_seen": 160776620,
      "step": 7434,
      "time_per_iteration": 2.6141843795776367
    },
    {
      "auxiliary_loss_clip": 0.01164859,
      "auxiliary_loss_mlp": 0.01029155,
      "balance_loss_clip": 1.04384351,
      "balance_loss_mlp": 1.02148962,
      "epoch": 0.8940058919016414,
      "flos": 20704189858560.0,
      "grad_norm": 1.530142484748639,
      "language_loss": 0.71317071,
      "learning_rate": 1.1659224810974367e-07,
      "loss": 0.73511088,
      "num_input_tokens_seen": 160796580,
      "step": 7435,
      "time_per_iteration": 2.5673561096191406
    },
    {
      "auxiliary_loss_clip": 0.01135345,
      "auxiliary_loss_mlp": 0.01028864,
      "balance_loss_clip": 1.04347646,
      "balance_loss_mlp": 1.0211513,
      "epoch": 0.8941261347922804,
      "flos": 25229041937280.0,
      "grad_norm": 1.5126272640937286,
      "language_loss": 0.68308073,
      "learning_rate": 1.1633031150934591e-07,
      "loss": 0.70472276,
      "num_input_tokens_seen": 160819610,
      "step": 7436,
      "time_per_iteration": 2.7280995845794678
    },
    {
      "auxiliary_loss_clip": 0.01153379,
      "auxiliary_loss_mlp": 0.01021751,
      "balance_loss_clip": 1.04577756,
      "balance_loss_mlp": 1.01421058,
      "epoch": 0.8942463776829195,
      "flos": 19537236806400.0,
      "grad_norm": 1.8953367033192885,
      "language_loss": 0.79640096,
      "learning_rate": 1.1606866066014176e-07,
      "loss": 0.81815231,
      "num_input_tokens_seen": 160838660,
      "step": 7437,
      "time_per_iteration": 3.942559242248535
    },
    {
      "auxiliary_loss_clip": 0.01120033,
      "auxiliary_loss_mlp": 0.01021318,
      "balance_loss_clip": 1.04087043,
      "balance_loss_mlp": 1.01407349,
      "epoch": 0.8943666205735585,
      "flos": 22301567585280.0,
      "grad_norm": 2.488436171341624,
      "language_loss": 0.75217509,
      "learning_rate": 1.1580729560182434e-07,
      "loss": 0.77358854,
      "num_input_tokens_seen": 160854515,
      "step": 7438,
      "time_per_iteration": 2.72472882270813
    },
    {
      "auxiliary_loss_clip": 0.01167693,
      "auxiliary_loss_mlp": 0.00761634,
      "balance_loss_clip": 1.04682326,
      "balance_loss_mlp": 1.00031435,
      "epoch": 0.8944868634641977,
      "flos": 18912893581440.0,
      "grad_norm": 1.6584847870898263,
      "language_loss": 0.71202075,
      "learning_rate": 1.1554621637404171e-07,
      "loss": 0.73131406,
      "num_input_tokens_seen": 160872605,
      "step": 7439,
      "time_per_iteration": 2.5792744159698486
    },
    {
      "auxiliary_loss_clip": 0.01148945,
      "auxiliary_loss_mlp": 0.01019307,
      "balance_loss_clip": 1.04319513,
      "balance_loss_mlp": 1.01221454,
      "epoch": 0.8946071063548368,
      "flos": 14460904241280.0,
      "grad_norm": 3.1879553021463725,
      "language_loss": 0.60706693,
      "learning_rate": 1.1528542301639999e-07,
      "loss": 0.62874949,
      "num_input_tokens_seen": 160889395,
      "step": 7440,
      "time_per_iteration": 3.5832526683807373
    },
    {
      "auxiliary_loss_clip": 0.01122519,
      "auxiliary_loss_mlp": 0.01024536,
      "balance_loss_clip": 1.0398519,
      "balance_loss_mlp": 1.01756859,
      "epoch": 0.8947273492454758,
      "flos": 20084084438400.0,
      "grad_norm": 2.330444272556525,
      "language_loss": 0.82260215,
      "learning_rate": 1.1502491556846105e-07,
      "loss": 0.8440727,
      "num_input_tokens_seen": 160907890,
      "step": 7441,
      "time_per_iteration": 2.705014228820801
    },
    {
      "auxiliary_loss_clip": 0.01135763,
      "auxiliary_loss_mlp": 0.01022387,
      "balance_loss_clip": 1.04189277,
      "balance_loss_mlp": 1.01515138,
      "epoch": 0.894847592136115,
      "flos": 18550555136640.0,
      "grad_norm": 2.747932214351184,
      "language_loss": 0.81284738,
      "learning_rate": 1.1476469406974331e-07,
      "loss": 0.83442891,
      "num_input_tokens_seen": 160923490,
      "step": 7442,
      "time_per_iteration": 3.553168535232544
    },
    {
      "auxiliary_loss_clip": 0.01162049,
      "auxiliary_loss_mlp": 0.01024867,
      "balance_loss_clip": 1.04543829,
      "balance_loss_mlp": 1.01801848,
      "epoch": 0.894967835026754,
      "flos": 23478468704640.0,
      "grad_norm": 1.7217554917059674,
      "language_loss": 0.77182031,
      "learning_rate": 1.1450475855972341e-07,
      "loss": 0.79368955,
      "num_input_tokens_seen": 160944280,
      "step": 7443,
      "time_per_iteration": 2.64827561378479
    },
    {
      "auxiliary_loss_clip": 0.01133327,
      "auxiliary_loss_mlp": 0.00762572,
      "balance_loss_clip": 1.03969169,
      "balance_loss_mlp": 1.00038099,
      "epoch": 0.8950880779173931,
      "flos": 15188310564480.0,
      "grad_norm": 1.971946134054133,
      "language_loss": 0.70522928,
      "learning_rate": 1.1424510907783158e-07,
      "loss": 0.72418833,
      "num_input_tokens_seen": 160961560,
      "step": 7444,
      "time_per_iteration": 3.6086862087249756
    },
    {
      "auxiliary_loss_clip": 0.01138953,
      "auxiliary_loss_mlp": 0.01020634,
      "balance_loss_clip": 1.04027367,
      "balance_loss_mlp": 1.01434278,
      "epoch": 0.8952083208080323,
      "flos": 22091957769600.0,
      "grad_norm": 1.71738364927884,
      "language_loss": 0.82828802,
      "learning_rate": 1.1398574566345787e-07,
      "loss": 0.84988391,
      "num_input_tokens_seen": 160982195,
      "step": 7445,
      "time_per_iteration": 2.6523241996765137
    },
    {
      "auxiliary_loss_clip": 0.01140599,
      "auxiliary_loss_mlp": 0.01022676,
      "balance_loss_clip": 1.04015625,
      "balance_loss_mlp": 1.01516294,
      "epoch": 0.8953285636986713,
      "flos": 23254026572160.0,
      "grad_norm": 2.3488273394815504,
      "language_loss": 0.82372141,
      "learning_rate": 1.1372666835594702e-07,
      "loss": 0.8453542,
      "num_input_tokens_seen": 161000520,
      "step": 7446,
      "time_per_iteration": 2.650315523147583
    },
    {
      "auxiliary_loss_clip": 0.01136933,
      "auxiliary_loss_mlp": 0.01021151,
      "balance_loss_clip": 1.04414439,
      "balance_loss_mlp": 1.013834,
      "epoch": 0.8954488065893104,
      "flos": 16362661818240.0,
      "grad_norm": 1.8575356389970492,
      "language_loss": 0.71789634,
      "learning_rate": 1.1346787719460071e-07,
      "loss": 0.73947716,
      "num_input_tokens_seen": 161019405,
      "step": 7447,
      "time_per_iteration": 2.6332733631134033
    },
    {
      "auxiliary_loss_clip": 0.01133584,
      "auxiliary_loss_mlp": 0.0102496,
      "balance_loss_clip": 1.04170942,
      "balance_loss_mlp": 1.01827204,
      "epoch": 0.8955690494799495,
      "flos": 18257883120000.0,
      "grad_norm": 1.7460261340910948,
      "language_loss": 0.71859759,
      "learning_rate": 1.1320937221867732e-07,
      "loss": 0.740183,
      "num_input_tokens_seen": 161036985,
      "step": 7448,
      "time_per_iteration": 2.6681582927703857
    },
    {
      "auxiliary_loss_clip": 0.01133794,
      "auxiliary_loss_mlp": 0.01022578,
      "balance_loss_clip": 1.04031754,
      "balance_loss_mlp": 1.01589322,
      "epoch": 0.8956892923705886,
      "flos": 25447486498560.0,
      "grad_norm": 1.8967884547298888,
      "language_loss": 0.79698253,
      "learning_rate": 1.1295115346739192e-07,
      "loss": 0.8185463,
      "num_input_tokens_seen": 161056985,
      "step": 7449,
      "time_per_iteration": 2.659618377685547
    },
    {
      "auxiliary_loss_clip": 0.0113992,
      "auxiliary_loss_mlp": 0.01026927,
      "balance_loss_clip": 1.04264605,
      "balance_loss_mlp": 1.01917219,
      "epoch": 0.8958095352612276,
      "flos": 52661883939840.0,
      "grad_norm": 2.806370654231317,
      "language_loss": 0.73193061,
      "learning_rate": 1.1269322097991629e-07,
      "loss": 0.75359905,
      "num_input_tokens_seen": 161080270,
      "step": 7450,
      "time_per_iteration": 2.930205821990967
    },
    {
      "auxiliary_loss_clip": 0.0115641,
      "auxiliary_loss_mlp": 0.0102947,
      "balance_loss_clip": 1.04591346,
      "balance_loss_mlp": 1.0220077,
      "epoch": 0.8959297781518668,
      "flos": 23186335392000.0,
      "grad_norm": 5.356569425663245,
      "language_loss": 0.6796906,
      "learning_rate": 1.1243557479537846e-07,
      "loss": 0.70154941,
      "num_input_tokens_seen": 161100160,
      "step": 7451,
      "time_per_iteration": 2.632756233215332
    },
    {
      "auxiliary_loss_clip": 0.01163021,
      "auxiliary_loss_mlp": 0.01022517,
      "balance_loss_clip": 1.04302394,
      "balance_loss_mlp": 1.01519799,
      "epoch": 0.8960500210425059,
      "flos": 20334309557760.0,
      "grad_norm": 1.98928265159776,
      "language_loss": 0.68950129,
      "learning_rate": 1.121782149528634e-07,
      "loss": 0.71135664,
      "num_input_tokens_seen": 161117260,
      "step": 7452,
      "time_per_iteration": 2.6182119846343994
    },
    {
      "auxiliary_loss_clip": 0.01139385,
      "auxiliary_loss_mlp": 0.01027904,
      "balance_loss_clip": 1.04454505,
      "balance_loss_mlp": 1.02131224,
      "epoch": 0.8961702639331449,
      "flos": 19901694153600.0,
      "grad_norm": 1.9835677328775578,
      "language_loss": 0.78941154,
      "learning_rate": 1.1192114149141208e-07,
      "loss": 0.81108445,
      "num_input_tokens_seen": 161136895,
      "step": 7453,
      "time_per_iteration": 2.6843714714050293
    },
    {
      "auxiliary_loss_clip": 0.01141766,
      "auxiliary_loss_mlp": 0.01025397,
      "balance_loss_clip": 1.0429728,
      "balance_loss_mlp": 1.01804733,
      "epoch": 0.8962905068237841,
      "flos": 12896348567040.0,
      "grad_norm": 2.4932292350177017,
      "language_loss": 0.65137076,
      "learning_rate": 1.1166435445002197e-07,
      "loss": 0.67304242,
      "num_input_tokens_seen": 161154565,
      "step": 7454,
      "time_per_iteration": 2.622429609298706
    },
    {
      "auxiliary_loss_clip": 0.01153929,
      "auxiliary_loss_mlp": 0.01026143,
      "balance_loss_clip": 1.04527688,
      "balance_loss_mlp": 1.01902902,
      "epoch": 0.8964107497144231,
      "flos": 23440331439360.0,
      "grad_norm": 2.0570353185960815,
      "language_loss": 0.68534786,
      "learning_rate": 1.1140785386764818e-07,
      "loss": 0.70714855,
      "num_input_tokens_seen": 161173265,
      "step": 7455,
      "time_per_iteration": 2.664933681488037
    },
    {
      "auxiliary_loss_clip": 0.01144371,
      "auxiliary_loss_mlp": 0.01022719,
      "balance_loss_clip": 1.04161286,
      "balance_loss_mlp": 1.01575446,
      "epoch": 0.8965309926050622,
      "flos": 19500176949120.0,
      "grad_norm": 2.107377953890634,
      "language_loss": 0.69888914,
      "learning_rate": 1.1115163978320153e-07,
      "loss": 0.72056001,
      "num_input_tokens_seen": 161191995,
      "step": 7456,
      "time_per_iteration": 2.5850014686584473
    },
    {
      "auxiliary_loss_clip": 0.01155747,
      "auxiliary_loss_mlp": 0.00762578,
      "balance_loss_clip": 1.04469156,
      "balance_loss_mlp": 1.00034785,
      "epoch": 0.8966512354957014,
      "flos": 28658008022400.0,
      "grad_norm": 2.1364327448431246,
      "language_loss": 0.82588494,
      "learning_rate": 1.1089571223554917e-07,
      "loss": 0.84506816,
      "num_input_tokens_seen": 161212880,
      "step": 7457,
      "time_per_iteration": 2.7034072875976562
    },
    {
      "auxiliary_loss_clip": 0.01151984,
      "auxiliary_loss_mlp": 0.0102524,
      "balance_loss_clip": 1.04275131,
      "balance_loss_mlp": 1.0184927,
      "epoch": 0.8967714783863404,
      "flos": 23370916406400.0,
      "grad_norm": 1.9829344362950219,
      "language_loss": 0.85416275,
      "learning_rate": 1.1064007126351537e-07,
      "loss": 0.87593508,
      "num_input_tokens_seen": 161233595,
      "step": 7458,
      "time_per_iteration": 2.656642436981201
    },
    {
      "auxiliary_loss_clip": 0.01132287,
      "auxiliary_loss_mlp": 0.01031591,
      "balance_loss_clip": 1.0435338,
      "balance_loss_mlp": 1.02449834,
      "epoch": 0.8968917212769795,
      "flos": 24535175938560.0,
      "grad_norm": 2.4169906311501608,
      "language_loss": 0.76247895,
      "learning_rate": 1.1038471690588003e-07,
      "loss": 0.7841177,
      "num_input_tokens_seen": 161252740,
      "step": 7459,
      "time_per_iteration": 2.7051680088043213
    },
    {
      "auxiliary_loss_clip": 0.01108159,
      "auxiliary_loss_mlp": 0.01025251,
      "balance_loss_clip": 1.04040778,
      "balance_loss_mlp": 1.01831615,
      "epoch": 0.8970119641676186,
      "flos": 23475416048640.0,
      "grad_norm": 1.9187004965295893,
      "language_loss": 0.80185378,
      "learning_rate": 1.1012964920138145e-07,
      "loss": 0.82318795,
      "num_input_tokens_seen": 161272325,
      "step": 7460,
      "time_per_iteration": 2.7652335166931152
    },
    {
      "auxiliary_loss_clip": 0.01131332,
      "auxiliary_loss_mlp": 0.01021168,
      "balance_loss_clip": 1.03952754,
      "balance_loss_mlp": 1.01469755,
      "epoch": 0.8971322070582577,
      "flos": 24538192680960.0,
      "grad_norm": 1.5685500694359769,
      "language_loss": 0.75856066,
      "learning_rate": 1.0987486818871205e-07,
      "loss": 0.78008568,
      "num_input_tokens_seen": 161295915,
      "step": 7461,
      "time_per_iteration": 2.7367734909057617
    },
    {
      "auxiliary_loss_clip": 0.01150157,
      "auxiliary_loss_mlp": 0.00761804,
      "balance_loss_clip": 1.04405928,
      "balance_loss_mlp": 1.00041413,
      "epoch": 0.8972524499488967,
      "flos": 21797454159360.0,
      "grad_norm": 2.4364966682686418,
      "language_loss": 0.73503733,
      "learning_rate": 1.0962037390652245e-07,
      "loss": 0.75415695,
      "num_input_tokens_seen": 161314935,
      "step": 7462,
      "time_per_iteration": 2.6520724296569824
    },
    {
      "auxiliary_loss_clip": 0.01136764,
      "auxiliary_loss_mlp": 0.01028894,
      "balance_loss_clip": 1.04324317,
      "balance_loss_mlp": 1.02171433,
      "epoch": 0.8973726928395359,
      "flos": 21726243446400.0,
      "grad_norm": 1.656673793861709,
      "language_loss": 0.71796644,
      "learning_rate": 1.0936616639341911e-07,
      "loss": 0.73962301,
      "num_input_tokens_seen": 161335225,
      "step": 7463,
      "time_per_iteration": 3.623293876647949
    },
    {
      "auxiliary_loss_clip": 0.01051061,
      "auxiliary_loss_mlp": 0.01001871,
      "balance_loss_clip": 1.01240206,
      "balance_loss_mlp": 1.00082803,
      "epoch": 0.897492935730175,
      "flos": 53837100097920.0,
      "grad_norm": 0.7363434419277795,
      "language_loss": 0.54720843,
      "learning_rate": 1.0911224568796473e-07,
      "loss": 0.56773782,
      "num_input_tokens_seen": 161393420,
      "step": 7464,
      "time_per_iteration": 3.2166359424591064
    },
    {
      "auxiliary_loss_clip": 0.01149107,
      "auxiliary_loss_mlp": 0.010238,
      "balance_loss_clip": 1.04502499,
      "balance_loss_mlp": 1.01700211,
      "epoch": 0.897613178620814,
      "flos": 18290346036480.0,
      "grad_norm": 2.4335203194975197,
      "language_loss": 0.71045792,
      "learning_rate": 1.0885861182867984e-07,
      "loss": 0.73218703,
      "num_input_tokens_seen": 161411525,
      "step": 7465,
      "time_per_iteration": 2.5866880416870117
    },
    {
      "auxiliary_loss_clip": 0.01138681,
      "auxiliary_loss_mlp": 0.01029263,
      "balance_loss_clip": 1.04242885,
      "balance_loss_mlp": 1.02240205,
      "epoch": 0.8977334215114532,
      "flos": 32993718059520.0,
      "grad_norm": 1.7809748942484074,
      "language_loss": 0.70909262,
      "learning_rate": 1.0860526485403942e-07,
      "loss": 0.73077214,
      "num_input_tokens_seen": 161432800,
      "step": 7466,
      "time_per_iteration": 3.6939661502838135
    },
    {
      "auxiliary_loss_clip": 0.01162435,
      "auxiliary_loss_mlp": 0.01026655,
      "balance_loss_clip": 1.04414499,
      "balance_loss_mlp": 1.02022338,
      "epoch": 0.8978536644020922,
      "flos": 15195636938880.0,
      "grad_norm": 1.6034380364275431,
      "language_loss": 0.76846254,
      "learning_rate": 1.0835220480247675e-07,
      "loss": 0.79035342,
      "num_input_tokens_seen": 161451295,
      "step": 7467,
      "time_per_iteration": 2.5587549209594727
    },
    {
      "auxiliary_loss_clip": 0.0113556,
      "auxiliary_loss_mlp": 0.01025083,
      "balance_loss_clip": 1.04268861,
      "balance_loss_mlp": 1.01793671,
      "epoch": 0.8979739072927313,
      "flos": 18004389863040.0,
      "grad_norm": 1.9043424507471454,
      "language_loss": 0.83555281,
      "learning_rate": 1.0809943171238067e-07,
      "loss": 0.85715926,
      "num_input_tokens_seen": 161469220,
      "step": 7468,
      "time_per_iteration": 3.5206382274627686
    },
    {
      "auxiliary_loss_clip": 0.01146144,
      "auxiliary_loss_mlp": 0.01031053,
      "balance_loss_clip": 1.04443288,
      "balance_loss_mlp": 1.02326822,
      "epoch": 0.8980941501833704,
      "flos": 22271546793600.0,
      "grad_norm": 2.0831953382323665,
      "language_loss": 0.62651372,
      "learning_rate": 1.078469456220965e-07,
      "loss": 0.64828575,
      "num_input_tokens_seen": 161489375,
      "step": 7469,
      "time_per_iteration": 2.6650235652923584
    },
    {
      "auxiliary_loss_clip": 0.0115022,
      "auxiliary_loss_mlp": 0.01025155,
      "balance_loss_clip": 1.04174829,
      "balance_loss_mlp": 1.0181663,
      "epoch": 0.8982143930740095,
      "flos": 37560729726720.0,
      "grad_norm": 1.7762213474101405,
      "language_loss": 0.6986059,
      "learning_rate": 1.0759474656992606e-07,
      "loss": 0.72035968,
      "num_input_tokens_seen": 161512145,
      "step": 7470,
      "time_per_iteration": 2.7258965969085693
    },
    {
      "auxiliary_loss_clip": 0.01141684,
      "auxiliary_loss_mlp": 0.01022982,
      "balance_loss_clip": 1.04008782,
      "balance_loss_mlp": 1.01568639,
      "epoch": 0.8983346359646486,
      "flos": 18076893465600.0,
      "grad_norm": 2.987719323287399,
      "language_loss": 0.78198719,
      "learning_rate": 1.0734283459412785e-07,
      "loss": 0.80363381,
      "num_input_tokens_seen": 161528995,
      "step": 7471,
      "time_per_iteration": 3.5803396701812744
    },
    {
      "auxiliary_loss_clip": 0.01112282,
      "auxiliary_loss_mlp": 0.01024141,
      "balance_loss_clip": 1.0388639,
      "balance_loss_mlp": 1.01633286,
      "epoch": 0.8984548788552876,
      "flos": 20558895344640.0,
      "grad_norm": 2.5897210527543235,
      "language_loss": 0.80967963,
      "learning_rate": 1.0709120973291707e-07,
      "loss": 0.83104384,
      "num_input_tokens_seen": 161548775,
      "step": 7472,
      "time_per_iteration": 2.7570033073425293
    },
    {
      "auxiliary_loss_clip": 0.01167845,
      "auxiliary_loss_mlp": 0.01029714,
      "balance_loss_clip": 1.0474149,
      "balance_loss_mlp": 1.02255213,
      "epoch": 0.8985751217459268,
      "flos": 17785442511360.0,
      "grad_norm": 2.220282513955287,
      "language_loss": 0.77384651,
      "learning_rate": 1.0683987202446475e-07,
      "loss": 0.79582208,
      "num_input_tokens_seen": 161566960,
      "step": 7473,
      "time_per_iteration": 2.6377439498901367
    },
    {
      "auxiliary_loss_clip": 0.01155272,
      "auxiliary_loss_mlp": 0.01027969,
      "balance_loss_clip": 1.04498696,
      "balance_loss_mlp": 1.02106977,
      "epoch": 0.8986953646365659,
      "flos": 21617003208960.0,
      "grad_norm": 1.931760119871698,
      "language_loss": 0.69728506,
      "learning_rate": 1.0658882150689862e-07,
      "loss": 0.71911752,
      "num_input_tokens_seen": 161585820,
      "step": 7474,
      "time_per_iteration": 2.7070140838623047
    },
    {
      "auxiliary_loss_clip": 0.01127006,
      "auxiliary_loss_mlp": 0.01026816,
      "balance_loss_clip": 1.04211164,
      "balance_loss_mlp": 1.01931822,
      "epoch": 0.8988156075272049,
      "flos": 14027355083520.0,
      "grad_norm": 4.705391490372896,
      "language_loss": 0.78556466,
      "learning_rate": 1.0633805821830288e-07,
      "loss": 0.80710292,
      "num_input_tokens_seen": 161602505,
      "step": 7475,
      "time_per_iteration": 2.7412021160125732
    },
    {
      "auxiliary_loss_clip": 0.01139145,
      "auxiliary_loss_mlp": 0.0102231,
      "balance_loss_clip": 1.04458427,
      "balance_loss_mlp": 1.01546121,
      "epoch": 0.8989358504178441,
      "flos": 29059202004480.0,
      "grad_norm": 2.3817809681620283,
      "language_loss": 0.82976413,
      "learning_rate": 1.0608758219671753e-07,
      "loss": 0.85137862,
      "num_input_tokens_seen": 161621545,
      "step": 7476,
      "time_per_iteration": 2.8094987869262695
    },
    {
      "auxiliary_loss_clip": 0.01142813,
      "auxiliary_loss_mlp": 0.01026727,
      "balance_loss_clip": 1.04343736,
      "balance_loss_mlp": 1.0200218,
      "epoch": 0.8990560933084831,
      "flos": 20230420446720.0,
      "grad_norm": 1.6013338599733564,
      "language_loss": 0.70772678,
      "learning_rate": 1.0583739348014065e-07,
      "loss": 0.72942215,
      "num_input_tokens_seen": 161642630,
      "step": 7477,
      "time_per_iteration": 2.6500439643859863
    },
    {
      "auxiliary_loss_clip": 0.01167009,
      "auxiliary_loss_mlp": 0.01021757,
      "balance_loss_clip": 1.04803276,
      "balance_loss_mlp": 1.01495564,
      "epoch": 0.8991763361991222,
      "flos": 25520672459520.0,
      "grad_norm": 3.057212763917033,
      "language_loss": 0.84628826,
      "learning_rate": 1.0558749210652518e-07,
      "loss": 0.86817586,
      "num_input_tokens_seen": 161662560,
      "step": 7478,
      "time_per_iteration": 2.633516550064087
    },
    {
      "auxiliary_loss_clip": 0.01129169,
      "auxiliary_loss_mlp": 0.01024636,
      "balance_loss_clip": 1.04147816,
      "balance_loss_mlp": 1.01797819,
      "epoch": 0.8992965790897613,
      "flos": 25119191168640.0,
      "grad_norm": 1.688268081091111,
      "language_loss": 0.85692185,
      "learning_rate": 1.053378781137808e-07,
      "loss": 0.87845987,
      "num_input_tokens_seen": 161683480,
      "step": 7479,
      "time_per_iteration": 2.7071781158447266
    },
    {
      "auxiliary_loss_clip": 0.01142559,
      "auxiliary_loss_mlp": 0.01024995,
      "balance_loss_clip": 1.043432,
      "balance_loss_mlp": 1.0173893,
      "epoch": 0.8994168219804004,
      "flos": 16070815814400.0,
      "grad_norm": 2.6824374199823984,
      "language_loss": 0.77866232,
      "learning_rate": 1.0508855153977392e-07,
      "loss": 0.80033791,
      "num_input_tokens_seen": 161699945,
      "step": 7480,
      "time_per_iteration": 2.687398672103882
    },
    {
      "auxiliary_loss_clip": 0.01150574,
      "auxiliary_loss_mlp": 0.01024877,
      "balance_loss_clip": 1.0413357,
      "balance_loss_mlp": 1.01784408,
      "epoch": 0.8995370648710395,
      "flos": 24825764966400.0,
      "grad_norm": 2.3312294621835536,
      "language_loss": 0.66562444,
      "learning_rate": 1.0483951242232669e-07,
      "loss": 0.68737894,
      "num_input_tokens_seen": 161720420,
      "step": 7481,
      "time_per_iteration": 2.6407556533813477
    },
    {
      "auxiliary_loss_clip": 0.01063808,
      "auxiliary_loss_mlp": 0.01001259,
      "balance_loss_clip": 1.00975394,
      "balance_loss_mlp": 1.00021029,
      "epoch": 0.8996573077616786,
      "flos": 63116238378240.0,
      "grad_norm": 0.9689030899997468,
      "language_loss": 0.57694322,
      "learning_rate": 1.0459076079921936e-07,
      "loss": 0.59759384,
      "num_input_tokens_seen": 161773080,
      "step": 7482,
      "time_per_iteration": 3.2190961837768555
    },
    {
      "auxiliary_loss_clip": 0.0113227,
      "auxiliary_loss_mlp": 0.01026154,
      "balance_loss_clip": 1.04362512,
      "balance_loss_mlp": 1.0189805,
      "epoch": 0.8997775506523177,
      "flos": 18219674027520.0,
      "grad_norm": 2.25796825242338,
      "language_loss": 0.84744614,
      "learning_rate": 1.0434229670818618e-07,
      "loss": 0.86903036,
      "num_input_tokens_seen": 161789755,
      "step": 7483,
      "time_per_iteration": 2.6516647338867188
    },
    {
      "auxiliary_loss_clip": 0.01129451,
      "auxiliary_loss_mlp": 0.01031402,
      "balance_loss_clip": 1.04130197,
      "balance_loss_mlp": 1.02452064,
      "epoch": 0.8998977935429567,
      "flos": 24166768095360.0,
      "grad_norm": 1.5540545793543252,
      "language_loss": 0.80122232,
      "learning_rate": 1.0409412018691944e-07,
      "loss": 0.82283086,
      "num_input_tokens_seen": 161810220,
      "step": 7484,
      "time_per_iteration": 2.711315631866455
    },
    {
      "auxiliary_loss_clip": 0.01135366,
      "auxiliary_loss_mlp": 0.01021026,
      "balance_loss_clip": 1.04350448,
      "balance_loss_mlp": 1.01328349,
      "epoch": 0.9000180364335959,
      "flos": 20773030273920.0,
      "grad_norm": 2.605849447683078,
      "language_loss": 0.74546456,
      "learning_rate": 1.0384623127306724e-07,
      "loss": 0.76702845,
      "num_input_tokens_seen": 161827565,
      "step": 7485,
      "time_per_iteration": 2.7109477519989014
    },
    {
      "auxiliary_loss_clip": 0.01119638,
      "auxiliary_loss_mlp": 0.01021802,
      "balance_loss_clip": 1.03947008,
      "balance_loss_mlp": 1.0149864,
      "epoch": 0.900138279324235,
      "flos": 19205745166080.0,
      "grad_norm": 1.7203711481362576,
      "language_loss": 0.79460794,
      "learning_rate": 1.0359863000423397e-07,
      "loss": 0.8160224,
      "num_input_tokens_seen": 161845700,
      "step": 7486,
      "time_per_iteration": 2.657818078994751
    },
    {
      "auxiliary_loss_clip": 0.01164726,
      "auxiliary_loss_mlp": 0.01028557,
      "balance_loss_clip": 1.04427338,
      "balance_loss_mlp": 1.02184832,
      "epoch": 0.900258522214874,
      "flos": 28731158069760.0,
      "grad_norm": 1.7635779941819547,
      "language_loss": 0.71668434,
      "learning_rate": 1.0335131641798112e-07,
      "loss": 0.73861718,
      "num_input_tokens_seen": 161867660,
      "step": 7487,
      "time_per_iteration": 2.6508100032806396
    },
    {
      "auxiliary_loss_clip": 0.0104216,
      "auxiliary_loss_mlp": 0.0100099,
      "balance_loss_clip": 1.00967264,
      "balance_loss_mlp": 0.99992335,
      "epoch": 0.9003787651055132,
      "flos": 58280685655680.0,
      "grad_norm": 0.8042407582268606,
      "language_loss": 0.55619001,
      "learning_rate": 1.0310429055182512e-07,
      "loss": 0.57662153,
      "num_input_tokens_seen": 161921980,
      "step": 7488,
      "time_per_iteration": 4.086040735244751
    },
    {
      "auxiliary_loss_clip": 0.0112545,
      "auxiliary_loss_mlp": 0.01025846,
      "balance_loss_clip": 1.0423882,
      "balance_loss_mlp": 1.01887226,
      "epoch": 0.9004990079961522,
      "flos": 25556475340800.0,
      "grad_norm": 1.7234547273838055,
      "language_loss": 0.74270356,
      "learning_rate": 1.0285755244324024e-07,
      "loss": 0.76421642,
      "num_input_tokens_seen": 161942725,
      "step": 7489,
      "time_per_iteration": 2.711121082305908
    },
    {
      "auxiliary_loss_clip": 0.01138869,
      "auxiliary_loss_mlp": 0.00762184,
      "balance_loss_clip": 1.03988457,
      "balance_loss_mlp": 1.00026011,
      "epoch": 0.9006192508867913,
      "flos": 23335185352320.0,
      "grad_norm": 1.411741920754481,
      "language_loss": 0.68765146,
      "learning_rate": 1.0261110212965629e-07,
      "loss": 0.70666194,
      "num_input_tokens_seen": 161964520,
      "step": 7490,
      "time_per_iteration": 2.6944119930267334
    },
    {
      "auxiliary_loss_clip": 0.01134467,
      "auxiliary_loss_mlp": 0.01023412,
      "balance_loss_clip": 1.041605,
      "balance_loss_mlp": 1.01625657,
      "epoch": 0.9007394937774305,
      "flos": 18040300485120.0,
      "grad_norm": 2.643206446593076,
      "language_loss": 0.79130399,
      "learning_rate": 1.023649396484596e-07,
      "loss": 0.81288278,
      "num_input_tokens_seen": 161983575,
      "step": 7491,
      "time_per_iteration": 2.633141040802002
    },
    {
      "auxiliary_loss_clip": 0.01164539,
      "auxiliary_loss_mlp": 0.01028868,
      "balance_loss_clip": 1.0461483,
      "balance_loss_mlp": 1.0214982,
      "epoch": 0.9008597366680695,
      "flos": 43068456633600.0,
      "grad_norm": 1.9141367663972866,
      "language_loss": 0.67315137,
      "learning_rate": 1.0211906503699275e-07,
      "loss": 0.69508547,
      "num_input_tokens_seen": 162006550,
      "step": 7492,
      "time_per_iteration": 2.7811005115509033
    },
    {
      "auxiliary_loss_clip": 0.01155748,
      "auxiliary_loss_mlp": 0.01027391,
      "balance_loss_clip": 1.04628682,
      "balance_loss_mlp": 1.02006578,
      "epoch": 0.9009799795587086,
      "flos": 14939055112320.0,
      "grad_norm": 2.434508339004845,
      "language_loss": 0.81922352,
      "learning_rate": 1.0187347833255455e-07,
      "loss": 0.84105492,
      "num_input_tokens_seen": 162022455,
      "step": 7493,
      "time_per_iteration": 3.5432701110839844
    },
    {
      "auxiliary_loss_clip": 0.01165715,
      "auxiliary_loss_mlp": 0.01031405,
      "balance_loss_clip": 1.04692388,
      "balance_loss_mlp": 1.02443433,
      "epoch": 0.9011002224493477,
      "flos": 21579584215680.0,
      "grad_norm": 1.9614395438153616,
      "language_loss": 0.79110283,
      "learning_rate": 1.0162817957240056e-07,
      "loss": 0.81307405,
      "num_input_tokens_seen": 162042350,
      "step": 7494,
      "time_per_iteration": 2.585740089416504
    },
    {
      "auxiliary_loss_clip": 0.01055133,
      "auxiliary_loss_mlp": 0.01001501,
      "balance_loss_clip": 1.01080346,
      "balance_loss_mlp": 1.0004462,
      "epoch": 0.9012204653399868,
      "flos": 71166367883520.0,
      "grad_norm": 0.8914843051871135,
      "language_loss": 0.63008344,
      "learning_rate": 1.0138316879374253e-07,
      "loss": 0.65064979,
      "num_input_tokens_seen": 162111640,
      "step": 7495,
      "time_per_iteration": 4.191587448120117
    },
    {
      "auxiliary_loss_clip": 0.01141358,
      "auxiliary_loss_mlp": 0.0102307,
      "balance_loss_clip": 1.04531932,
      "balance_loss_mlp": 1.01585197,
      "epoch": 0.9013407082306258,
      "flos": 15594963413760.0,
      "grad_norm": 3.8747119074460903,
      "language_loss": 0.74548513,
      "learning_rate": 1.0113844603374833e-07,
      "loss": 0.76712942,
      "num_input_tokens_seen": 162128165,
      "step": 7496,
      "time_per_iteration": 2.578484535217285
    },
    {
      "auxiliary_loss_clip": 0.0113561,
      "auxiliary_loss_mlp": 0.01030425,
      "balance_loss_clip": 1.04095495,
      "balance_loss_mlp": 1.02283084,
      "epoch": 0.901460951121265,
      "flos": 15049157276160.0,
      "grad_norm": 2.5960247204441798,
      "language_loss": 0.7182011,
      "learning_rate": 1.0089401132954178e-07,
      "loss": 0.73986149,
      "num_input_tokens_seen": 162146145,
      "step": 7497,
      "time_per_iteration": 3.596393585205078
    },
    {
      "auxiliary_loss_clip": 0.01138036,
      "auxiliary_loss_mlp": 0.01029905,
      "balance_loss_clip": 1.04474664,
      "balance_loss_mlp": 1.02289224,
      "epoch": 0.9015811940119041,
      "flos": 22236857233920.0,
      "grad_norm": 1.5750861749138123,
      "language_loss": 0.72389007,
      "learning_rate": 1.006498647182037e-07,
      "loss": 0.74556947,
      "num_input_tokens_seen": 162164800,
      "step": 7498,
      "time_per_iteration": 2.68811297416687
    },
    {
      "auxiliary_loss_clip": 0.01093288,
      "auxiliary_loss_mlp": 0.01021752,
      "balance_loss_clip": 1.03695416,
      "balance_loss_mlp": 1.01459908,
      "epoch": 0.9017014369025431,
      "flos": 24973824827520.0,
      "grad_norm": 2.2004999555830373,
      "language_loss": 0.71438187,
      "learning_rate": 1.004060062367713e-07,
      "loss": 0.73553222,
      "num_input_tokens_seen": 162185895,
      "step": 7499,
      "time_per_iteration": 2.783907890319824
    },
    {
      "auxiliary_loss_clip": 0.01151949,
      "auxiliary_loss_mlp": 0.01024046,
      "balance_loss_clip": 1.04301107,
      "balance_loss_mlp": 1.01713789,
      "epoch": 0.9018216797931822,
      "flos": 18114168804480.0,
      "grad_norm": 2.309144761479108,
      "language_loss": 0.69272566,
      "learning_rate": 1.0016243592223728e-07,
      "loss": 0.71448559,
      "num_input_tokens_seen": 162206295,
      "step": 7500,
      "time_per_iteration": 2.622805118560791
    },
    {
      "auxiliary_loss_clip": 0.01091262,
      "auxiliary_loss_mlp": 0.01025826,
      "balance_loss_clip": 1.03768635,
      "balance_loss_mlp": 1.01859331,
      "epoch": 0.9019419226838213,
      "flos": 37268452759680.0,
      "grad_norm": 1.757979475690111,
      "language_loss": 0.65867019,
      "learning_rate": 9.991915381155114e-08,
      "loss": 0.6798411,
      "num_input_tokens_seen": 162229275,
      "step": 7501,
      "time_per_iteration": 2.9719340801239014
    },
    {
      "auxiliary_loss_clip": 0.01155395,
      "auxiliary_loss_mlp": 0.01029521,
      "balance_loss_clip": 1.04430795,
      "balance_loss_mlp": 1.02182651,
      "epoch": 0.9020621655744604,
      "flos": 23441121538560.0,
      "grad_norm": 2.380677541815729,
      "language_loss": 0.75073987,
      "learning_rate": 9.967615994161871e-08,
      "loss": 0.77258909,
      "num_input_tokens_seen": 162248935,
      "step": 7502,
      "time_per_iteration": 2.6853790283203125
    },
    {
      "auxiliary_loss_clip": 0.01164816,
      "auxiliary_loss_mlp": 0.01027779,
      "balance_loss_clip": 1.04642856,
      "balance_loss_mlp": 1.02114463,
      "epoch": 0.9021824084650995,
      "flos": 22857465444480.0,
      "grad_norm": 1.6989298696149613,
      "language_loss": 0.78080213,
      "learning_rate": 9.943345434930161e-08,
      "loss": 0.80272812,
      "num_input_tokens_seen": 162269185,
      "step": 7503,
      "time_per_iteration": 2.5978803634643555
    },
    {
      "auxiliary_loss_clip": 0.01123101,
      "auxiliary_loss_mlp": 0.01023962,
      "balance_loss_clip": 1.04342222,
      "balance_loss_mlp": 1.01693487,
      "epoch": 0.9023026513557386,
      "flos": 22127581082880.0,
      "grad_norm": 2.342494841957817,
      "language_loss": 0.69161689,
      "learning_rate": 9.919103707141885e-08,
      "loss": 0.71308756,
      "num_input_tokens_seen": 162288065,
      "step": 7504,
      "time_per_iteration": 2.760854959487915
    },
    {
      "auxiliary_loss_clip": 0.01152228,
      "auxiliary_loss_mlp": 0.01024093,
      "balance_loss_clip": 1.0444417,
      "balance_loss_mlp": 1.01655281,
      "epoch": 0.9024228942463777,
      "flos": 24199087357440.0,
      "grad_norm": 2.954755713620168,
      "language_loss": 0.76296848,
      "learning_rate": 9.89489081447441e-08,
      "loss": 0.78473169,
      "num_input_tokens_seen": 162305265,
      "step": 7505,
      "time_per_iteration": 2.652313470840454
    },
    {
      "auxiliary_loss_clip": 0.01134345,
      "auxiliary_loss_mlp": 0.01025552,
      "balance_loss_clip": 1.0406146,
      "balance_loss_mlp": 1.01839948,
      "epoch": 0.9025431371370167,
      "flos": 25008262992000.0,
      "grad_norm": 1.989528062955611,
      "language_loss": 0.82918298,
      "learning_rate": 9.870706760600844e-08,
      "loss": 0.85078198,
      "num_input_tokens_seen": 162325215,
      "step": 7506,
      "time_per_iteration": 2.7177555561065674
    },
    {
      "auxiliary_loss_clip": 0.01117131,
      "auxiliary_loss_mlp": 0.0102982,
      "balance_loss_clip": 1.04359579,
      "balance_loss_mlp": 1.02216101,
      "epoch": 0.9026633800276559,
      "flos": 18952862440320.0,
      "grad_norm": 1.9638228327656109,
      "language_loss": 0.72993952,
      "learning_rate": 9.846551549189918e-08,
      "loss": 0.75140905,
      "num_input_tokens_seen": 162344820,
      "step": 7507,
      "time_per_iteration": 2.7286295890808105
    },
    {
      "auxiliary_loss_clip": 0.01134573,
      "auxiliary_loss_mlp": 0.01028815,
      "balance_loss_clip": 1.0433135,
      "balance_loss_mlp": 1.02170372,
      "epoch": 0.902783622918295,
      "flos": 32416059536640.0,
      "grad_norm": 2.144435465108351,
      "language_loss": 0.69234824,
      "learning_rate": 9.822425183905902e-08,
      "loss": 0.71398211,
      "num_input_tokens_seen": 162365345,
      "step": 7508,
      "time_per_iteration": 2.760657787322998
    },
    {
      "auxiliary_loss_clip": 0.01033678,
      "auxiliary_loss_mlp": 0.01001591,
      "balance_loss_clip": 1.01046145,
      "balance_loss_mlp": 1.00054193,
      "epoch": 0.902903865808934,
      "flos": 63717453244800.0,
      "grad_norm": 0.936184080714115,
      "language_loss": 0.75099272,
      "learning_rate": 9.798327668408823e-08,
      "loss": 0.77134538,
      "num_input_tokens_seen": 162426980,
      "step": 7509,
      "time_per_iteration": 3.3759632110595703
    },
    {
      "auxiliary_loss_clip": 0.01169052,
      "auxiliary_loss_mlp": 0.01027308,
      "balance_loss_clip": 1.04614687,
      "balance_loss_mlp": 1.01877236,
      "epoch": 0.9030241086995732,
      "flos": 23804034600960.0,
      "grad_norm": 1.9082347842483731,
      "language_loss": 0.68898797,
      "learning_rate": 9.774259006354158e-08,
      "loss": 0.71095157,
      "num_input_tokens_seen": 162447050,
      "step": 7510,
      "time_per_iteration": 2.610992670059204
    },
    {
      "auxiliary_loss_clip": 0.01142042,
      "auxiliary_loss_mlp": 0.01020641,
      "balance_loss_clip": 1.0420835,
      "balance_loss_mlp": 1.01375401,
      "epoch": 0.9031443515902122,
      "flos": 26395887248640.0,
      "grad_norm": 2.256454837210997,
      "language_loss": 0.76189762,
      "learning_rate": 9.750219201393184e-08,
      "loss": 0.78352445,
      "num_input_tokens_seen": 162467015,
      "step": 7511,
      "time_per_iteration": 2.7084898948669434
    },
    {
      "auxiliary_loss_clip": 0.01150178,
      "auxiliary_loss_mlp": 0.01024607,
      "balance_loss_clip": 1.04366517,
      "balance_loss_mlp": 1.01754653,
      "epoch": 0.9032645944808513,
      "flos": 24939350749440.0,
      "grad_norm": 1.758395323599528,
      "language_loss": 0.77696788,
      "learning_rate": 9.726208257172697e-08,
      "loss": 0.79871571,
      "num_input_tokens_seen": 162488710,
      "step": 7512,
      "time_per_iteration": 2.6857612133026123
    },
    {
      "auxiliary_loss_clip": 0.01164509,
      "auxiliary_loss_mlp": 0.01025852,
      "balance_loss_clip": 1.04455233,
      "balance_loss_mlp": 1.01877117,
      "epoch": 0.9033848373714904,
      "flos": 21178821196800.0,
      "grad_norm": 1.989910450203253,
      "language_loss": 0.74446535,
      "learning_rate": 9.702226177335115e-08,
      "loss": 0.76636893,
      "num_input_tokens_seen": 162507205,
      "step": 7513,
      "time_per_iteration": 2.572045087814331
    },
    {
      "auxiliary_loss_clip": 0.01136841,
      "auxiliary_loss_mlp": 0.01027452,
      "balance_loss_clip": 1.04362929,
      "balance_loss_mlp": 1.02030504,
      "epoch": 0.9035050802621295,
      "flos": 26286359702400.0,
      "grad_norm": 1.6213920918709042,
      "language_loss": 0.72510844,
      "learning_rate": 9.67827296551853e-08,
      "loss": 0.74675131,
      "num_input_tokens_seen": 162528490,
      "step": 7514,
      "time_per_iteration": 3.6757595539093018
    },
    {
      "auxiliary_loss_clip": 0.01129282,
      "auxiliary_loss_mlp": 0.00761904,
      "balance_loss_clip": 1.04045236,
      "balance_loss_mlp": 1.00036764,
      "epoch": 0.9036253231527686,
      "flos": 24204546224640.0,
      "grad_norm": 1.86239407793086,
      "language_loss": 0.68437457,
      "learning_rate": 9.65434862535659e-08,
      "loss": 0.70328647,
      "num_input_tokens_seen": 162547860,
      "step": 7515,
      "time_per_iteration": 2.685185432434082
    },
    {
      "auxiliary_loss_clip": 0.01141398,
      "auxiliary_loss_mlp": 0.01026509,
      "balance_loss_clip": 1.04261494,
      "balance_loss_mlp": 1.01887345,
      "epoch": 0.9037455660434077,
      "flos": 18072655660800.0,
      "grad_norm": 4.829171678960358,
      "language_loss": 0.64984804,
      "learning_rate": 9.630453160478635e-08,
      "loss": 0.67152709,
      "num_input_tokens_seen": 162563215,
      "step": 7516,
      "time_per_iteration": 2.685162305831909
    },
    {
      "auxiliary_loss_clip": 0.01106083,
      "auxiliary_loss_mlp": 0.01026051,
      "balance_loss_clip": 1.03752637,
      "balance_loss_mlp": 1.01883042,
      "epoch": 0.9038658089340468,
      "flos": 24060795995520.0,
      "grad_norm": 1.5650476450505546,
      "language_loss": 0.82163799,
      "learning_rate": 9.60658657450959e-08,
      "loss": 0.84295928,
      "num_input_tokens_seen": 162583515,
      "step": 7517,
      "time_per_iteration": 2.824997663497925
    },
    {
      "auxiliary_loss_clip": 0.01122066,
      "auxiliary_loss_mlp": 0.01029205,
      "balance_loss_clip": 1.03749323,
      "balance_loss_mlp": 1.02241313,
      "epoch": 0.9039860518246858,
      "flos": 21834298535040.0,
      "grad_norm": 1.6757111802027558,
      "language_loss": 0.7991631,
      "learning_rate": 9.582748871069979e-08,
      "loss": 0.82067585,
      "num_input_tokens_seen": 162602955,
      "step": 7518,
      "time_per_iteration": 3.648031234741211
    },
    {
      "auxiliary_loss_clip": 0.01139367,
      "auxiliary_loss_mlp": 0.00761844,
      "balance_loss_clip": 1.04110622,
      "balance_loss_mlp": 1.00031078,
      "epoch": 0.904106294715325,
      "flos": 26614870513920.0,
      "grad_norm": 2.633312412071103,
      "language_loss": 0.83450294,
      "learning_rate": 9.558940053775954e-08,
      "loss": 0.85351509,
      "num_input_tokens_seen": 162621595,
      "step": 7519,
      "time_per_iteration": 2.7094030380249023
    },
    {
      "auxiliary_loss_clip": 0.01149903,
      "auxiliary_loss_mlp": 0.01028495,
      "balance_loss_clip": 1.04446495,
      "balance_loss_mlp": 1.02099991,
      "epoch": 0.904226537605964,
      "flos": 17785693906560.0,
      "grad_norm": 2.0165058882222717,
      "language_loss": 0.67766631,
      "learning_rate": 9.535160126239294e-08,
      "loss": 0.69945025,
      "num_input_tokens_seen": 162638220,
      "step": 7520,
      "time_per_iteration": 3.5656442642211914
    },
    {
      "auxiliary_loss_clip": 0.01149712,
      "auxiliary_loss_mlp": 0.01028387,
      "balance_loss_clip": 1.0443275,
      "balance_loss_mlp": 1.02110016,
      "epoch": 0.9043467804966031,
      "flos": 24790428961920.0,
      "grad_norm": 1.7630175880903183,
      "language_loss": 0.70860142,
      "learning_rate": 9.511409092067424e-08,
      "loss": 0.73038238,
      "num_input_tokens_seen": 162658575,
      "step": 7521,
      "time_per_iteration": 2.648160457611084
    },
    {
      "auxiliary_loss_clip": 0.01136643,
      "auxiliary_loss_mlp": 0.01022894,
      "balance_loss_clip": 1.04390883,
      "balance_loss_mlp": 1.01629591,
      "epoch": 0.9044670233872423,
      "flos": 22632125472000.0,
      "grad_norm": 1.8558894864122448,
      "language_loss": 0.67152953,
      "learning_rate": 9.487686954863327e-08,
      "loss": 0.69312489,
      "num_input_tokens_seen": 162678295,
      "step": 7522,
      "time_per_iteration": 3.522277593612671
    },
    {
      "auxiliary_loss_clip": 0.01149524,
      "auxiliary_loss_mlp": 0.01028615,
      "balance_loss_clip": 1.04288828,
      "balance_loss_mlp": 1.02219868,
      "epoch": 0.9045872662778813,
      "flos": 23771320289280.0,
      "grad_norm": 1.8660529656847549,
      "language_loss": 0.77287996,
      "learning_rate": 9.46399371822566e-08,
      "loss": 0.79466134,
      "num_input_tokens_seen": 162698070,
      "step": 7523,
      "time_per_iteration": 2.6655097007751465
    },
    {
      "auxiliary_loss_clip": 0.01165776,
      "auxiliary_loss_mlp": 0.01026143,
      "balance_loss_clip": 1.04575968,
      "balance_loss_mlp": 1.01918077,
      "epoch": 0.9047075091685204,
      "flos": 15191039998080.0,
      "grad_norm": 2.242243090004402,
      "language_loss": 0.72257918,
      "learning_rate": 9.440329385748657e-08,
      "loss": 0.74449837,
      "num_input_tokens_seen": 162715140,
      "step": 7524,
      "time_per_iteration": 2.5464425086975098
    },
    {
      "auxiliary_loss_clip": 0.0112425,
      "auxiliary_loss_mlp": 0.01020797,
      "balance_loss_clip": 1.04379082,
      "balance_loss_mlp": 1.01435399,
      "epoch": 0.9048277520591596,
      "flos": 18003707504640.0,
      "grad_norm": 2.1241468686941642,
      "language_loss": 0.70464754,
      "learning_rate": 9.416693961022137e-08,
      "loss": 0.72609806,
      "num_input_tokens_seen": 162733390,
      "step": 7525,
      "time_per_iteration": 2.687877893447876
    },
    {
      "auxiliary_loss_clip": 0.01082078,
      "auxiliary_loss_mlp": 0.01025042,
      "balance_loss_clip": 1.03600979,
      "balance_loss_mlp": 1.01834893,
      "epoch": 0.9049479949497986,
      "flos": 21872471713920.0,
      "grad_norm": 1.8035281831187637,
      "language_loss": 0.77182603,
      "learning_rate": 9.393087447631654e-08,
      "loss": 0.79289722,
      "num_input_tokens_seen": 162751670,
      "step": 7526,
      "time_per_iteration": 2.7627320289611816
    },
    {
      "auxiliary_loss_clip": 0.01136972,
      "auxiliary_loss_mlp": 0.01016439,
      "balance_loss_clip": 1.04058695,
      "balance_loss_mlp": 1.01015353,
      "epoch": 0.9050682378404377,
      "flos": 20773928113920.0,
      "grad_norm": 1.7185747801957985,
      "language_loss": 0.72567368,
      "learning_rate": 9.36950984915823e-08,
      "loss": 0.74720776,
      "num_input_tokens_seen": 162770025,
      "step": 7527,
      "time_per_iteration": 2.6017401218414307
    },
    {
      "auxiliary_loss_clip": 0.01167518,
      "auxiliary_loss_mlp": 0.01022498,
      "balance_loss_clip": 1.04706073,
      "balance_loss_mlp": 1.01492858,
      "epoch": 0.9051884807310768,
      "flos": 21580015178880.0,
      "grad_norm": 1.8152065297017723,
      "language_loss": 0.69522399,
      "learning_rate": 9.345961169178607e-08,
      "loss": 0.7171241,
      "num_input_tokens_seen": 162789710,
      "step": 7528,
      "time_per_iteration": 2.608520030975342
    },
    {
      "auxiliary_loss_clip": 0.01110317,
      "auxiliary_loss_mlp": 0.01024589,
      "balance_loss_clip": 1.04173255,
      "balance_loss_mlp": 1.01719773,
      "epoch": 0.9053087236217159,
      "flos": 21908059113600.0,
      "grad_norm": 1.476535750691194,
      "language_loss": 0.72860318,
      "learning_rate": 9.322441411265081e-08,
      "loss": 0.74995226,
      "num_input_tokens_seen": 162810695,
      "step": 7529,
      "time_per_iteration": 2.8112106323242188
    },
    {
      "auxiliary_loss_clip": 0.01134297,
      "auxiliary_loss_mlp": 0.01035518,
      "balance_loss_clip": 1.04374099,
      "balance_loss_mlp": 1.02859223,
      "epoch": 0.9054289665123549,
      "flos": 17055809544960.0,
      "grad_norm": 2.863518121874271,
      "language_loss": 0.73420584,
      "learning_rate": 9.298950578985554e-08,
      "loss": 0.75590402,
      "num_input_tokens_seen": 162827770,
      "step": 7530,
      "time_per_iteration": 2.6814639568328857
    },
    {
      "auxiliary_loss_clip": 0.01150735,
      "auxiliary_loss_mlp": 0.00762762,
      "balance_loss_clip": 1.04590631,
      "balance_loss_mlp": 1.00031745,
      "epoch": 0.905549209402994,
      "flos": 20777268078720.0,
      "grad_norm": 1.8363708311851827,
      "language_loss": 0.71069628,
      "learning_rate": 9.275488675903665e-08,
      "loss": 0.72983122,
      "num_input_tokens_seen": 162846715,
      "step": 7531,
      "time_per_iteration": 2.737701892852783
    },
    {
      "auxiliary_loss_clip": 0.01107133,
      "auxiliary_loss_mlp": 0.01020358,
      "balance_loss_clip": 1.04077506,
      "balance_loss_mlp": 1.0136106,
      "epoch": 0.9056694522936332,
      "flos": 21686813291520.0,
      "grad_norm": 2.1537953052384182,
      "language_loss": 0.73349404,
      "learning_rate": 9.252055705578454e-08,
      "loss": 0.75476897,
      "num_input_tokens_seen": 162866215,
      "step": 7532,
      "time_per_iteration": 2.7770280838012695
    },
    {
      "auxiliary_loss_clip": 0.01148908,
      "auxiliary_loss_mlp": 0.01023459,
      "balance_loss_clip": 1.04201305,
      "balance_loss_mlp": 1.01721263,
      "epoch": 0.9057896951842722,
      "flos": 29569133433600.0,
      "grad_norm": 1.6268520958680661,
      "language_loss": 0.72196412,
      "learning_rate": 9.228651671564747e-08,
      "loss": 0.74368775,
      "num_input_tokens_seen": 162888245,
      "step": 7533,
      "time_per_iteration": 2.651372194290161
    },
    {
      "auxiliary_loss_clip": 0.01105002,
      "auxiliary_loss_mlp": 0.01028311,
      "balance_loss_clip": 1.0425005,
      "balance_loss_mlp": 1.02130795,
      "epoch": 0.9059099380749113,
      "flos": 27892248952320.0,
      "grad_norm": 1.583931326710199,
      "language_loss": 0.77862203,
      "learning_rate": 9.205276577412901e-08,
      "loss": 0.79995519,
      "num_input_tokens_seen": 162911025,
      "step": 7534,
      "time_per_iteration": 2.783057928085327
    },
    {
      "auxiliary_loss_clip": 0.01145285,
      "auxiliary_loss_mlp": 0.0076244,
      "balance_loss_clip": 1.04340339,
      "balance_loss_mlp": 1.00036061,
      "epoch": 0.9060301809655504,
      "flos": 17748993185280.0,
      "grad_norm": 2.6456162105602634,
      "language_loss": 0.76877356,
      "learning_rate": 9.181930426668905e-08,
      "loss": 0.78785074,
      "num_input_tokens_seen": 162927820,
      "step": 7535,
      "time_per_iteration": 2.650944232940674
    },
    {
      "auxiliary_loss_clip": 0.01104364,
      "auxiliary_loss_mlp": 0.01024915,
      "balance_loss_clip": 1.0391326,
      "balance_loss_mlp": 1.0184921,
      "epoch": 0.9061504238561895,
      "flos": 31759432963200.0,
      "grad_norm": 1.5260771699625746,
      "language_loss": 0.67818153,
      "learning_rate": 9.158613222874346e-08,
      "loss": 0.69947428,
      "num_input_tokens_seen": 162949445,
      "step": 7536,
      "time_per_iteration": 2.8395633697509766
    },
    {
      "auxiliary_loss_clip": 0.01134593,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 1.04063153,
      "balance_loss_mlp": 1.0172112,
      "epoch": 0.9062706667468285,
      "flos": 20048066075520.0,
      "grad_norm": 1.517487488840768,
      "language_loss": 0.8205837,
      "learning_rate": 9.135324969566394e-08,
      "loss": 0.84217185,
      "num_input_tokens_seen": 162968945,
      "step": 7537,
      "time_per_iteration": 2.652219533920288
    },
    {
      "auxiliary_loss_clip": 0.01155623,
      "auxiliary_loss_mlp": 0.01024373,
      "balance_loss_clip": 1.04561484,
      "balance_loss_mlp": 1.01751792,
      "epoch": 0.9063909096374677,
      "flos": 18437292576000.0,
      "grad_norm": 2.310433170408504,
      "language_loss": 0.75534648,
      "learning_rate": 9.112065670277913e-08,
      "loss": 0.77714646,
      "num_input_tokens_seen": 162985310,
      "step": 7538,
      "time_per_iteration": 2.579052448272705
    },
    {
      "auxiliary_loss_clip": 0.01133621,
      "auxiliary_loss_mlp": 0.01022087,
      "balance_loss_clip": 1.03995323,
      "balance_loss_mlp": 1.01603985,
      "epoch": 0.9065111525281068,
      "flos": 33547353361920.0,
      "grad_norm": 2.187899442076199,
      "language_loss": 0.73284775,
      "learning_rate": 9.088835328537303e-08,
      "loss": 0.75440478,
      "num_input_tokens_seen": 163006900,
      "step": 7539,
      "time_per_iteration": 2.790241241455078
    },
    {
      "auxiliary_loss_clip": 0.01140371,
      "auxiliary_loss_mlp": 0.01022196,
      "balance_loss_clip": 1.04412627,
      "balance_loss_mlp": 1.01494217,
      "epoch": 0.9066313954187458,
      "flos": 23367863750400.0,
      "grad_norm": 1.9993671748199193,
      "language_loss": 0.7165553,
      "learning_rate": 9.065633947868568e-08,
      "loss": 0.73818099,
      "num_input_tokens_seen": 163026505,
      "step": 7540,
      "time_per_iteration": 2.682766914367676
    },
    {
      "auxiliary_loss_clip": 0.01122534,
      "auxiliary_loss_mlp": 0.00761649,
      "balance_loss_clip": 1.04290628,
      "balance_loss_mlp": 1.00032604,
      "epoch": 0.906751638309385,
      "flos": 26249623067520.0,
      "grad_norm": 7.352428582176788,
      "language_loss": 0.79932451,
      "learning_rate": 9.042461531791379e-08,
      "loss": 0.81816638,
      "num_input_tokens_seen": 163044925,
      "step": 7541,
      "time_per_iteration": 3.7136340141296387
    },
    {
      "auxiliary_loss_clip": 0.01161554,
      "auxiliary_loss_mlp": 0.01024747,
      "balance_loss_clip": 1.0446198,
      "balance_loss_mlp": 1.01800537,
      "epoch": 0.906871881200024,
      "flos": 16544477485440.0,
      "grad_norm": 1.7677273352814147,
      "language_loss": 0.77448726,
      "learning_rate": 9.019318083820903e-08,
      "loss": 0.79635036,
      "num_input_tokens_seen": 163063505,
      "step": 7542,
      "time_per_iteration": 2.5902040004730225
    },
    {
      "auxiliary_loss_clip": 0.01150281,
      "auxiliary_loss_mlp": 0.01025784,
      "balance_loss_clip": 1.04463542,
      "balance_loss_mlp": 1.01829731,
      "epoch": 0.9069921240906631,
      "flos": 24605129675520.0,
      "grad_norm": 1.652885743282099,
      "language_loss": 0.85334629,
      "learning_rate": 8.996203607468045e-08,
      "loss": 0.87510693,
      "num_input_tokens_seen": 163082505,
      "step": 7543,
      "time_per_iteration": 2.7205371856689453
    },
    {
      "auxiliary_loss_clip": 0.011436,
      "auxiliary_loss_mlp": 0.01023398,
      "balance_loss_clip": 1.03949761,
      "balance_loss_mlp": 1.01657009,
      "epoch": 0.9071123669813023,
      "flos": 25374731500800.0,
      "grad_norm": 1.4471665193130725,
      "language_loss": 0.75612509,
      "learning_rate": 8.973118106239241e-08,
      "loss": 0.77779508,
      "num_input_tokens_seen": 163105110,
      "step": 7544,
      "time_per_iteration": 3.639089822769165
    },
    {
      "auxiliary_loss_clip": 0.01094559,
      "auxiliary_loss_mlp": 0.01024351,
      "balance_loss_clip": 1.03490102,
      "balance_loss_mlp": 1.01705849,
      "epoch": 0.9072326098719413,
      "flos": 26725798690560.0,
      "grad_norm": 1.8874982612983875,
      "language_loss": 0.94777429,
      "learning_rate": 8.95006158363656e-08,
      "loss": 0.96896338,
      "num_input_tokens_seen": 163125295,
      "step": 7545,
      "time_per_iteration": 2.8518083095550537
    },
    {
      "auxiliary_loss_clip": 0.01153177,
      "auxiliary_loss_mlp": 0.01025357,
      "balance_loss_clip": 1.0468235,
      "balance_loss_mlp": 1.0180074,
      "epoch": 0.9073528527625804,
      "flos": 23878800760320.0,
      "grad_norm": 1.7101642751131165,
      "language_loss": 0.77300781,
      "learning_rate": 8.9270340431576e-08,
      "loss": 0.79479313,
      "num_input_tokens_seen": 163144385,
      "step": 7546,
      "time_per_iteration": 3.5462546348571777
    },
    {
      "auxiliary_loss_clip": 0.01152908,
      "auxiliary_loss_mlp": 0.01026916,
      "balance_loss_clip": 1.04404378,
      "balance_loss_mlp": 1.01985002,
      "epoch": 0.9074730956532195,
      "flos": 37852144767360.0,
      "grad_norm": 2.001078106446448,
      "language_loss": 0.73783016,
      "learning_rate": 8.904035488295658e-08,
      "loss": 0.75962842,
      "num_input_tokens_seen": 163163885,
      "step": 7547,
      "time_per_iteration": 2.874358892440796
    },
    {
      "auxiliary_loss_clip": 0.01054193,
      "auxiliary_loss_mlp": 0.00753663,
      "balance_loss_clip": 1.00993085,
      "balance_loss_mlp": 1.00017929,
      "epoch": 0.9075933385438586,
      "flos": 65173307385600.0,
      "grad_norm": 0.666049887924127,
      "language_loss": 0.53281349,
      "learning_rate": 8.881065922539632e-08,
      "loss": 0.55089206,
      "num_input_tokens_seen": 163224325,
      "step": 7548,
      "time_per_iteration": 4.027567625045776
    },
    {
      "auxiliary_loss_clip": 0.01118642,
      "auxiliary_loss_mlp": 0.01020538,
      "balance_loss_clip": 1.04331565,
      "balance_loss_mlp": 1.01412141,
      "epoch": 0.9077135814344977,
      "flos": 19931571290880.0,
      "grad_norm": 1.7157179521021957,
      "language_loss": 0.73469841,
      "learning_rate": 8.85812534937389e-08,
      "loss": 0.75609016,
      "num_input_tokens_seen": 163242425,
      "step": 7549,
      "time_per_iteration": 2.7390100955963135
    },
    {
      "auxiliary_loss_clip": 0.01157313,
      "auxiliary_loss_mlp": 0.01025567,
      "balance_loss_clip": 1.04546928,
      "balance_loss_mlp": 1.01839304,
      "epoch": 0.9078338243251368,
      "flos": 17529650784000.0,
      "grad_norm": 2.3573982909341757,
      "language_loss": 0.67368519,
      "learning_rate": 8.835213772278583e-08,
      "loss": 0.69551396,
      "num_input_tokens_seen": 163259280,
      "step": 7550,
      "time_per_iteration": 2.593651056289673
    },
    {
      "auxiliary_loss_clip": 0.01112739,
      "auxiliary_loss_mlp": 0.01019512,
      "balance_loss_clip": 1.04103625,
      "balance_loss_mlp": 1.01277113,
      "epoch": 0.9079540672157759,
      "flos": 28803410277120.0,
      "grad_norm": 1.8015758823918182,
      "language_loss": 0.79183877,
      "learning_rate": 8.812331194729373e-08,
      "loss": 0.81316125,
      "num_input_tokens_seen": 163278925,
      "step": 7551,
      "time_per_iteration": 2.7943496704101562
    },
    {
      "auxiliary_loss_clip": 0.0117083,
      "auxiliary_loss_mlp": 0.01029376,
      "balance_loss_clip": 1.04875088,
      "balance_loss_mlp": 1.02173209,
      "epoch": 0.9080743101064149,
      "flos": 23513840622720.0,
      "grad_norm": 1.8476795207579242,
      "language_loss": 0.71763378,
      "learning_rate": 8.789477620197461e-08,
      "loss": 0.73963588,
      "num_input_tokens_seen": 163298450,
      "step": 7552,
      "time_per_iteration": 2.6756112575531006
    },
    {
      "auxiliary_loss_clip": 0.01135941,
      "auxiliary_loss_mlp": 0.01025155,
      "balance_loss_clip": 1.04238939,
      "balance_loss_mlp": 1.01777327,
      "epoch": 0.9081945529970541,
      "flos": 22778102344320.0,
      "grad_norm": 5.3359523065614916,
      "language_loss": 0.79339969,
      "learning_rate": 8.766653052149831e-08,
      "loss": 0.81501061,
      "num_input_tokens_seen": 163313635,
      "step": 7553,
      "time_per_iteration": 2.6896326541900635
    },
    {
      "auxiliary_loss_clip": 0.01134449,
      "auxiliary_loss_mlp": 0.01022073,
      "balance_loss_clip": 1.04266405,
      "balance_loss_mlp": 1.015064,
      "epoch": 0.9083147958876931,
      "flos": 18873714821760.0,
      "grad_norm": 2.6474896535068826,
      "language_loss": 0.74551362,
      "learning_rate": 8.743857494048823e-08,
      "loss": 0.76707882,
      "num_input_tokens_seen": 163330450,
      "step": 7554,
      "time_per_iteration": 2.662921667098999
    },
    {
      "auxiliary_loss_clip": 0.01124985,
      "auxiliary_loss_mlp": 0.01025911,
      "balance_loss_clip": 1.04311299,
      "balance_loss_mlp": 1.01915765,
      "epoch": 0.9084350387783322,
      "flos": 18909374048640.0,
      "grad_norm": 1.9264900626783867,
      "language_loss": 0.62881768,
      "learning_rate": 8.721090949352605e-08,
      "loss": 0.65032661,
      "num_input_tokens_seen": 163346690,
      "step": 7555,
      "time_per_iteration": 2.68681001663208
    },
    {
      "auxiliary_loss_clip": 0.01161269,
      "auxiliary_loss_mlp": 0.01035409,
      "balance_loss_clip": 1.04711545,
      "balance_loss_mlp": 1.02710056,
      "epoch": 0.9085552816689714,
      "flos": 20595488325120.0,
      "grad_norm": 1.8231722980924816,
      "language_loss": 0.72986078,
      "learning_rate": 8.698353421514793e-08,
      "loss": 0.7518276,
      "num_input_tokens_seen": 163365065,
      "step": 7556,
      "time_per_iteration": 2.631441354751587
    },
    {
      "auxiliary_loss_clip": 0.01148064,
      "auxiliary_loss_mlp": 0.0102762,
      "balance_loss_clip": 1.04275632,
      "balance_loss_mlp": 1.0207293,
      "epoch": 0.9086755245596104,
      "flos": 18113163223680.0,
      "grad_norm": 4.100523558211616,
      "language_loss": 0.80881,
      "learning_rate": 8.67564491398467e-08,
      "loss": 0.83056682,
      "num_input_tokens_seen": 163382070,
      "step": 7557,
      "time_per_iteration": 2.5800838470458984
    },
    {
      "auxiliary_loss_clip": 0.01152595,
      "auxiliary_loss_mlp": 0.01028256,
      "balance_loss_clip": 1.04222643,
      "balance_loss_mlp": 1.0210979,
      "epoch": 0.9087957674502495,
      "flos": 19129793857920.0,
      "grad_norm": 1.946548101986808,
      "language_loss": 0.73406601,
      "learning_rate": 8.652965430207104e-08,
      "loss": 0.75587451,
      "num_input_tokens_seen": 163399975,
      "step": 7558,
      "time_per_iteration": 2.614776611328125
    },
    {
      "auxiliary_loss_clip": 0.01154894,
      "auxiliary_loss_mlp": 0.010258,
      "balance_loss_clip": 1.04393947,
      "balance_loss_mlp": 1.01896381,
      "epoch": 0.9089160103408886,
      "flos": 18109930999680.0,
      "grad_norm": 2.173657454733986,
      "language_loss": 0.65453291,
      "learning_rate": 8.630314973622521e-08,
      "loss": 0.67633986,
      "num_input_tokens_seen": 163417520,
      "step": 7559,
      "time_per_iteration": 2.6410157680511475
    },
    {
      "auxiliary_loss_clip": 0.01147415,
      "auxiliary_loss_mlp": 0.01026703,
      "balance_loss_clip": 1.04492664,
      "balance_loss_mlp": 1.01968122,
      "epoch": 0.9090362532315277,
      "flos": 33364855336320.0,
      "grad_norm": 2.028575082106824,
      "language_loss": 0.71012747,
      "learning_rate": 8.607693547666995e-08,
      "loss": 0.73186862,
      "num_input_tokens_seen": 163440060,
      "step": 7560,
      "time_per_iteration": 2.7446165084838867
    },
    {
      "auxiliary_loss_clip": 0.01034085,
      "auxiliary_loss_mlp": 0.01001634,
      "balance_loss_clip": 1.00928926,
      "balance_loss_mlp": 1.00053084,
      "epoch": 0.9091564961221668,
      "flos": 71480585082240.0,
      "grad_norm": 0.8842872766630178,
      "language_loss": 0.57948869,
      "learning_rate": 8.585101155772201e-08,
      "loss": 0.59984583,
      "num_input_tokens_seen": 163502180,
      "step": 7561,
      "time_per_iteration": 3.400104522705078
    },
    {
      "auxiliary_loss_clip": 0.01129703,
      "auxiliary_loss_mlp": 0.01023232,
      "balance_loss_clip": 1.03858352,
      "balance_loss_mlp": 1.01591289,
      "epoch": 0.9092767390128058,
      "flos": 24712574232960.0,
      "grad_norm": 2.0658978243693973,
      "language_loss": 0.68511021,
      "learning_rate": 8.562537801365377e-08,
      "loss": 0.70663959,
      "num_input_tokens_seen": 163521915,
      "step": 7562,
      "time_per_iteration": 2.70809006690979
    },
    {
      "auxiliary_loss_clip": 0.01166627,
      "auxiliary_loss_mlp": 0.01029731,
      "balance_loss_clip": 1.04529786,
      "balance_loss_mlp": 1.02227449,
      "epoch": 0.909396981903445,
      "flos": 23586487879680.0,
      "grad_norm": 1.9504683159369043,
      "language_loss": 0.70099032,
      "learning_rate": 8.540003487869362e-08,
      "loss": 0.72295386,
      "num_input_tokens_seen": 163543585,
      "step": 7563,
      "time_per_iteration": 2.7313616275787354
    },
    {
      "auxiliary_loss_clip": 0.01112535,
      "auxiliary_loss_mlp": 0.0102352,
      "balance_loss_clip": 1.03979373,
      "balance_loss_mlp": 1.01667738,
      "epoch": 0.909517224794084,
      "flos": 23404169422080.0,
      "grad_norm": 1.9694719006866148,
      "language_loss": 0.79831016,
      "learning_rate": 8.517498218702557e-08,
      "loss": 0.81967074,
      "num_input_tokens_seen": 163561515,
      "step": 7564,
      "time_per_iteration": 2.7416539192199707
    },
    {
      "auxiliary_loss_clip": 0.01117338,
      "auxiliary_loss_mlp": 0.01021145,
      "balance_loss_clip": 1.04021287,
      "balance_loss_mlp": 1.01445746,
      "epoch": 0.9096374676847231,
      "flos": 19208618254080.0,
      "grad_norm": 1.8523011294157627,
      "language_loss": 0.69645095,
      "learning_rate": 8.49502199727905e-08,
      "loss": 0.71783578,
      "num_input_tokens_seen": 163579540,
      "step": 7565,
      "time_per_iteration": 2.7775590419769287
    },
    {
      "auxiliary_loss_clip": 0.01145045,
      "auxiliary_loss_mlp": 0.01024026,
      "balance_loss_clip": 1.04031837,
      "balance_loss_mlp": 1.01656687,
      "epoch": 0.9097577105753623,
      "flos": 33292495388160.0,
      "grad_norm": 2.5717994917422438,
      "language_loss": 0.6620667,
      "learning_rate": 8.472574827008428e-08,
      "loss": 0.68375742,
      "num_input_tokens_seen": 163600425,
      "step": 7566,
      "time_per_iteration": 3.624068021774292
    },
    {
      "auxiliary_loss_clip": 0.01151878,
      "auxiliary_loss_mlp": 0.01019224,
      "balance_loss_clip": 1.04249239,
      "balance_loss_mlp": 1.012146,
      "epoch": 0.9098779534660013,
      "flos": 21906443001600.0,
      "grad_norm": 1.899111444721732,
      "language_loss": 0.83921778,
      "learning_rate": 8.450156711295942e-08,
      "loss": 0.86092883,
      "num_input_tokens_seen": 163620595,
      "step": 7567,
      "time_per_iteration": 2.7044951915740967
    },
    {
      "auxiliary_loss_clip": 0.01135316,
      "auxiliary_loss_mlp": 0.01019544,
      "balance_loss_clip": 1.04502118,
      "balance_loss_mlp": 1.0130204,
      "epoch": 0.9099981963566404,
      "flos": 25730354102400.0,
      "grad_norm": 2.2441126838804504,
      "language_loss": 0.8674525,
      "learning_rate": 8.427767653542383e-08,
      "loss": 0.88900113,
      "num_input_tokens_seen": 163635765,
      "step": 7568,
      "time_per_iteration": 2.6741533279418945
    },
    {
      "auxiliary_loss_clip": 0.01105929,
      "auxiliary_loss_mlp": 0.01023897,
      "balance_loss_clip": 1.03892255,
      "balance_loss_mlp": 1.01729012,
      "epoch": 0.9101184392472795,
      "flos": 21069437304960.0,
      "grad_norm": 1.9411886301653993,
      "language_loss": 0.7031737,
      "learning_rate": 8.405407657144125e-08,
      "loss": 0.72447193,
      "num_input_tokens_seen": 163654925,
      "step": 7569,
      "time_per_iteration": 3.765500545501709
    },
    {
      "auxiliary_loss_clip": 0.01132462,
      "auxiliary_loss_mlp": 0.01027365,
      "balance_loss_clip": 1.04251921,
      "balance_loss_mlp": 1.0205667,
      "epoch": 0.9102386821379186,
      "flos": 24752614919040.0,
      "grad_norm": 1.8801637001973244,
      "language_loss": 0.72531021,
      "learning_rate": 8.383076725493232e-08,
      "loss": 0.74690849,
      "num_input_tokens_seen": 163672245,
      "step": 7570,
      "time_per_iteration": 2.778780460357666
    },
    {
      "auxiliary_loss_clip": 0.0115004,
      "auxiliary_loss_mlp": 0.01019948,
      "balance_loss_clip": 1.04261577,
      "balance_loss_mlp": 1.01344776,
      "epoch": 0.9103589250285576,
      "flos": 22562818179840.0,
      "grad_norm": 1.8320344769673471,
      "language_loss": 0.68250501,
      "learning_rate": 8.360774861977216e-08,
      "loss": 0.70420492,
      "num_input_tokens_seen": 163691365,
      "step": 7571,
      "time_per_iteration": 2.6000194549560547
    },
    {
      "auxiliary_loss_clip": 0.01134425,
      "auxiliary_loss_mlp": 0.01021689,
      "balance_loss_clip": 1.03881752,
      "balance_loss_mlp": 1.0152632,
      "epoch": 0.9104791679191968,
      "flos": 25373474524800.0,
      "grad_norm": 1.8384781593638426,
      "language_loss": 0.7453596,
      "learning_rate": 8.338502069979281e-08,
      "loss": 0.76692075,
      "num_input_tokens_seen": 163711675,
      "step": 7572,
      "time_per_iteration": 3.5489108562469482
    },
    {
      "auxiliary_loss_clip": 0.01152576,
      "auxiliary_loss_mlp": 0.01027007,
      "balance_loss_clip": 1.04193687,
      "balance_loss_mlp": 1.01970863,
      "epoch": 0.9105994108098359,
      "flos": 14426681558400.0,
      "grad_norm": 3.820421690487028,
      "language_loss": 0.79727364,
      "learning_rate": 8.316258352878214e-08,
      "loss": 0.81906945,
      "num_input_tokens_seen": 163728095,
      "step": 7573,
      "time_per_iteration": 3.5043232440948486
    },
    {
      "auxiliary_loss_clip": 0.01152702,
      "auxiliary_loss_mlp": 0.01024579,
      "balance_loss_clip": 1.0412842,
      "balance_loss_mlp": 1.01696098,
      "epoch": 0.9107196537004749,
      "flos": 26718292748160.0,
      "grad_norm": 1.9296479617541982,
      "language_loss": 0.71383977,
      "learning_rate": 8.294043714048338e-08,
      "loss": 0.73561251,
      "num_input_tokens_seen": 163747175,
      "step": 7574,
      "time_per_iteration": 2.672872543334961
    },
    {
      "auxiliary_loss_clip": 0.01043754,
      "auxiliary_loss_mlp": 0.01000811,
      "balance_loss_clip": 1.01006448,
      "balance_loss_mlp": 0.99973768,
      "epoch": 0.9108398965911141,
      "flos": 66532634703360.0,
      "grad_norm": 0.7483655809720475,
      "language_loss": 0.60474563,
      "learning_rate": 8.271858156859624e-08,
      "loss": 0.62519133,
      "num_input_tokens_seen": 163812545,
      "step": 7575,
      "time_per_iteration": 3.38234543800354
    },
    {
      "auxiliary_loss_clip": 0.01162679,
      "auxiliary_loss_mlp": 0.01025358,
      "balance_loss_clip": 1.04518533,
      "balance_loss_mlp": 1.01813376,
      "epoch": 0.9109601394817531,
      "flos": 25411073086080.0,
      "grad_norm": 1.731867551314336,
      "language_loss": 0.73926282,
      "learning_rate": 8.249701684677557e-08,
      "loss": 0.76114321,
      "num_input_tokens_seen": 163833870,
      "step": 7576,
      "time_per_iteration": 2.7244439125061035
    },
    {
      "auxiliary_loss_clip": 0.0115293,
      "auxiliary_loss_mlp": 0.01021797,
      "balance_loss_clip": 1.0481602,
      "balance_loss_mlp": 1.01484728,
      "epoch": 0.9110803823723922,
      "flos": 22747794243840.0,
      "grad_norm": 1.6820271712045682,
      "language_loss": 0.80802846,
      "learning_rate": 8.227574300863294e-08,
      "loss": 0.82977569,
      "num_input_tokens_seen": 163854040,
      "step": 7577,
      "time_per_iteration": 2.66678524017334
    },
    {
      "auxiliary_loss_clip": 0.01142096,
      "auxiliary_loss_mlp": 0.01024786,
      "balance_loss_clip": 1.04468119,
      "balance_loss_mlp": 1.0174247,
      "epoch": 0.9112006252630314,
      "flos": 48469924131840.0,
      "grad_norm": 2.1013751158822878,
      "language_loss": 0.694116,
      "learning_rate": 8.205476008773548e-08,
      "loss": 0.71578491,
      "num_input_tokens_seen": 163878040,
      "step": 7578,
      "time_per_iteration": 2.855280637741089
    },
    {
      "auxiliary_loss_clip": 0.01117627,
      "auxiliary_loss_mlp": 0.01022892,
      "balance_loss_clip": 1.04395509,
      "balance_loss_mlp": 1.01659751,
      "epoch": 0.9113208681536704,
      "flos": 30009649829760.0,
      "grad_norm": 2.176768261107171,
      "language_loss": 0.82399273,
      "learning_rate": 8.183406811760596e-08,
      "loss": 0.84539795,
      "num_input_tokens_seen": 163897770,
      "step": 7579,
      "time_per_iteration": 2.7840023040771484
    },
    {
      "auxiliary_loss_clip": 0.01108667,
      "auxiliary_loss_mlp": 0.01024826,
      "balance_loss_clip": 1.03823876,
      "balance_loss_mlp": 1.01735127,
      "epoch": 0.9114411110443095,
      "flos": 25594971742080.0,
      "grad_norm": 1.528770714725637,
      "language_loss": 0.74013674,
      "learning_rate": 8.161366713172313e-08,
      "loss": 0.76147175,
      "num_input_tokens_seen": 163920160,
      "step": 7580,
      "time_per_iteration": 2.7960729598999023
    },
    {
      "auxiliary_loss_clip": 0.01126363,
      "auxiliary_loss_mlp": 0.01027997,
      "balance_loss_clip": 1.04052103,
      "balance_loss_mlp": 1.02039444,
      "epoch": 0.9115613539349486,
      "flos": 18399729928320.0,
      "grad_norm": 4.1563638957372255,
      "language_loss": 0.8413617,
      "learning_rate": 8.139355716352137e-08,
      "loss": 0.86290526,
      "num_input_tokens_seen": 163935000,
      "step": 7581,
      "time_per_iteration": 2.6205849647521973
    },
    {
      "auxiliary_loss_clip": 0.01139983,
      "auxiliary_loss_mlp": 0.0102295,
      "balance_loss_clip": 1.04232132,
      "balance_loss_mlp": 1.01584554,
      "epoch": 0.9116815968255877,
      "flos": 21726171619200.0,
      "grad_norm": 1.4819125631825243,
      "language_loss": 0.7014116,
      "learning_rate": 8.117373824639196e-08,
      "loss": 0.72304094,
      "num_input_tokens_seen": 163955265,
      "step": 7582,
      "time_per_iteration": 2.6819677352905273
    },
    {
      "auxiliary_loss_clip": 0.01063293,
      "auxiliary_loss_mlp": 0.01001445,
      "balance_loss_clip": 1.00952125,
      "balance_loss_mlp": 1.00039005,
      "epoch": 0.9118018397162267,
      "flos": 65363526835200.0,
      "grad_norm": 0.7190697644534829,
      "language_loss": 0.59229374,
      "learning_rate": 8.095421041368067e-08,
      "loss": 0.61294115,
      "num_input_tokens_seen": 164014680,
      "step": 7583,
      "time_per_iteration": 3.082826614379883
    },
    {
      "auxiliary_loss_clip": 0.01133824,
      "auxiliary_loss_mlp": 0.0076235,
      "balance_loss_clip": 1.04459012,
      "balance_loss_mlp": 1.00034189,
      "epoch": 0.9119220826068659,
      "flos": 20922885815040.0,
      "grad_norm": 1.8059256973161877,
      "language_loss": 0.70735234,
      "learning_rate": 8.073497369868999e-08,
      "loss": 0.72631419,
      "num_input_tokens_seen": 164033140,
      "step": 7584,
      "time_per_iteration": 2.845285177230835
    },
    {
      "auxiliary_loss_clip": 0.01143929,
      "auxiliary_loss_mlp": 0.01024215,
      "balance_loss_clip": 1.0435096,
      "balance_loss_mlp": 1.01716042,
      "epoch": 0.912042325497505,
      "flos": 28366449327360.0,
      "grad_norm": 1.6685230526558648,
      "language_loss": 0.75698781,
      "learning_rate": 8.051602813467772e-08,
      "loss": 0.77866924,
      "num_input_tokens_seen": 164054995,
      "step": 7585,
      "time_per_iteration": 2.7530479431152344
    },
    {
      "auxiliary_loss_clip": 0.01152444,
      "auxiliary_loss_mlp": 0.01025798,
      "balance_loss_clip": 1.04331684,
      "balance_loss_mlp": 1.01835918,
      "epoch": 0.912162568388144,
      "flos": 17566782468480.0,
      "grad_norm": 2.8226419960573224,
      "language_loss": 0.70898205,
      "learning_rate": 8.029737375485756e-08,
      "loss": 0.73076451,
      "num_input_tokens_seen": 164074225,
      "step": 7586,
      "time_per_iteration": 2.6744871139526367
    },
    {
      "auxiliary_loss_clip": 0.01165059,
      "auxiliary_loss_mlp": 0.01025661,
      "balance_loss_clip": 1.04511333,
      "balance_loss_mlp": 1.01842523,
      "epoch": 0.9122828112787832,
      "flos": 19827897661440.0,
      "grad_norm": 1.9056399274799611,
      "language_loss": 0.72658217,
      "learning_rate": 8.007901059239986e-08,
      "loss": 0.74848938,
      "num_input_tokens_seen": 164093505,
      "step": 7587,
      "time_per_iteration": 2.7120113372802734
    },
    {
      "auxiliary_loss_clip": 0.01136567,
      "auxiliary_loss_mlp": 0.01022578,
      "balance_loss_clip": 1.04137635,
      "balance_loss_mlp": 1.0158788,
      "epoch": 0.9124030541694222,
      "flos": 20813789232000.0,
      "grad_norm": 1.7846286034567664,
      "language_loss": 0.80371708,
      "learning_rate": 7.986093868042964e-08,
      "loss": 0.82530856,
      "num_input_tokens_seen": 164113750,
      "step": 7588,
      "time_per_iteration": 2.748192071914673
    },
    {
      "auxiliary_loss_clip": 0.01146757,
      "auxiliary_loss_mlp": 0.01024572,
      "balance_loss_clip": 1.04205239,
      "balance_loss_mlp": 1.01846886,
      "epoch": 0.9125232970600613,
      "flos": 25192305302400.0,
      "grad_norm": 2.2550696172116003,
      "language_loss": 0.6749078,
      "learning_rate": 7.964315805202826e-08,
      "loss": 0.69662106,
      "num_input_tokens_seen": 164134330,
      "step": 7589,
      "time_per_iteration": 2.7242565155029297
    },
    {
      "auxiliary_loss_clip": 0.01137178,
      "auxiliary_loss_mlp": 0.01028781,
      "balance_loss_clip": 1.04271328,
      "balance_loss_mlp": 1.02175379,
      "epoch": 0.9126435399507005,
      "flos": 19719591177600.0,
      "grad_norm": 1.942789697417644,
      "language_loss": 0.7344113,
      "learning_rate": 7.942566874023304e-08,
      "loss": 0.75607085,
      "num_input_tokens_seen": 164153515,
      "step": 7590,
      "time_per_iteration": 2.645998477935791
    },
    {
      "auxiliary_loss_clip": 0.01134033,
      "auxiliary_loss_mlp": 0.01025708,
      "balance_loss_clip": 1.04057574,
      "balance_loss_mlp": 1.0185051,
      "epoch": 0.9127637828413395,
      "flos": 19573614305280.0,
      "grad_norm": 2.199382407726409,
      "language_loss": 0.69409108,
      "learning_rate": 7.920847077803649e-08,
      "loss": 0.71568853,
      "num_input_tokens_seen": 164171305,
      "step": 7591,
      "time_per_iteration": 2.652820110321045
    },
    {
      "auxiliary_loss_clip": 0.01094308,
      "auxiliary_loss_mlp": 0.01029537,
      "balance_loss_clip": 1.03361201,
      "balance_loss_mlp": 1.02217841,
      "epoch": 0.9128840257319786,
      "flos": 20230635928320.0,
      "grad_norm": 1.7192022156005455,
      "language_loss": 0.82237589,
      "learning_rate": 7.899156419838826e-08,
      "loss": 0.84361434,
      "num_input_tokens_seen": 164190275,
      "step": 7592,
      "time_per_iteration": 3.6023526191711426
    },
    {
      "auxiliary_loss_clip": 0.0112109,
      "auxiliary_loss_mlp": 0.01025723,
      "balance_loss_clip": 1.0417273,
      "balance_loss_mlp": 1.01811695,
      "epoch": 0.9130042686226177,
      "flos": 24858658846080.0,
      "grad_norm": 1.6966404373521042,
      "language_loss": 0.65586615,
      "learning_rate": 7.87749490341918e-08,
      "loss": 0.67733431,
      "num_input_tokens_seen": 164210550,
      "step": 7593,
      "time_per_iteration": 2.720277786254883
    },
    {
      "auxiliary_loss_clip": 0.01168778,
      "auxiliary_loss_mlp": 0.01029471,
      "balance_loss_clip": 1.0469439,
      "balance_loss_mlp": 1.02159977,
      "epoch": 0.9131245115132568,
      "flos": 23581747284480.0,
      "grad_norm": 2.152154197513166,
      "language_loss": 0.83359933,
      "learning_rate": 7.855862531830836e-08,
      "loss": 0.85558176,
      "num_input_tokens_seen": 164226660,
      "step": 7594,
      "time_per_iteration": 2.5774543285369873
    },
    {
      "auxiliary_loss_clip": 0.01147869,
      "auxiliary_loss_mlp": 0.01029524,
      "balance_loss_clip": 1.04231191,
      "balance_loss_mlp": 1.02208877,
      "epoch": 0.9132447544038959,
      "flos": 19931607204480.0,
      "grad_norm": 2.637175875139042,
      "language_loss": 0.72541451,
      "learning_rate": 7.834259308355373e-08,
      "loss": 0.74718845,
      "num_input_tokens_seen": 164245425,
      "step": 7595,
      "time_per_iteration": 3.6096365451812744
    },
    {
      "auxiliary_loss_clip": 0.01078729,
      "auxiliary_loss_mlp": 0.01022352,
      "balance_loss_clip": 1.03595901,
      "balance_loss_mlp": 1.01530683,
      "epoch": 0.9133649972945349,
      "flos": 21981747864960.0,
      "grad_norm": 2.6485707028771146,
      "language_loss": 0.75203794,
      "learning_rate": 7.812685236269989e-08,
      "loss": 0.77304876,
      "num_input_tokens_seen": 164264085,
      "step": 7596,
      "time_per_iteration": 2.821704149246216
    },
    {
      "auxiliary_loss_clip": 0.01031093,
      "auxiliary_loss_mlp": 0.01000178,
      "balance_loss_clip": 1.01224458,
      "balance_loss_mlp": 0.9991768,
      "epoch": 0.9134852401851741,
      "flos": 71240523511680.0,
      "grad_norm": 0.7901730152502962,
      "language_loss": 0.58604628,
      "learning_rate": 7.791140318847445e-08,
      "loss": 0.606359,
      "num_input_tokens_seen": 164322220,
      "step": 7597,
      "time_per_iteration": 3.279336452484131
    },
    {
      "auxiliary_loss_clip": 0.01132185,
      "auxiliary_loss_mlp": 0.01019953,
      "balance_loss_clip": 1.04364705,
      "balance_loss_mlp": 1.01363492,
      "epoch": 0.9136054830758131,
      "flos": 23626923615360.0,
      "grad_norm": 1.6581025552927458,
      "language_loss": 0.80373216,
      "learning_rate": 7.769624559356081e-08,
      "loss": 0.82525349,
      "num_input_tokens_seen": 164345615,
      "step": 7598,
      "time_per_iteration": 3.714294195175171
    },
    {
      "auxiliary_loss_clip": 0.01149703,
      "auxiliary_loss_mlp": 0.01024349,
      "balance_loss_clip": 1.04329622,
      "balance_loss_mlp": 1.01686275,
      "epoch": 0.9137257259664522,
      "flos": 23438858981760.0,
      "grad_norm": 4.517381550510644,
      "language_loss": 0.75598282,
      "learning_rate": 7.748137961059842e-08,
      "loss": 0.77772331,
      "num_input_tokens_seen": 164359595,
      "step": 7599,
      "time_per_iteration": 3.556138038635254
    },
    {
      "auxiliary_loss_clip": 0.01161965,
      "auxiliary_loss_mlp": 0.0102144,
      "balance_loss_clip": 1.04588544,
      "balance_loss_mlp": 1.01427817,
      "epoch": 0.9138459688570914,
      "flos": 19127854523520.0,
      "grad_norm": 2.3608828447659986,
      "language_loss": 0.65603423,
      "learning_rate": 7.726680527218211e-08,
      "loss": 0.67786825,
      "num_input_tokens_seen": 164376635,
      "step": 7600,
      "time_per_iteration": 2.5688884258270264
    },
    {
      "auxiliary_loss_clip": 0.01161873,
      "auxiliary_loss_mlp": 0.0102145,
      "balance_loss_clip": 1.04131842,
      "balance_loss_mlp": 1.01436055,
      "epoch": 0.9139662117477304,
      "flos": 46281240714240.0,
      "grad_norm": 1.6694982545675758,
      "language_loss": 0.75366908,
      "learning_rate": 7.70525226108627e-08,
      "loss": 0.77550232,
      "num_input_tokens_seen": 164400305,
      "step": 7601,
      "time_per_iteration": 2.781250476837158
    },
    {
      "auxiliary_loss_clip": 0.01154398,
      "auxiliary_loss_mlp": 0.01023954,
      "balance_loss_clip": 1.04784989,
      "balance_loss_mlp": 1.0166142,
      "epoch": 0.9140864546383695,
      "flos": 22273198819200.0,
      "grad_norm": 1.7618234679457931,
      "language_loss": 0.79779029,
      "learning_rate": 7.683853165914666e-08,
      "loss": 0.81957376,
      "num_input_tokens_seen": 164418075,
      "step": 7602,
      "time_per_iteration": 2.641444683074951
    },
    {
      "auxiliary_loss_clip": 0.01110234,
      "auxiliary_loss_mlp": 0.01032169,
      "balance_loss_clip": 1.03989422,
      "balance_loss_mlp": 1.02482307,
      "epoch": 0.9142066975290086,
      "flos": 17530009920000.0,
      "grad_norm": 1.784081790343506,
      "language_loss": 0.77260429,
      "learning_rate": 7.662483244949602e-08,
      "loss": 0.79402834,
      "num_input_tokens_seen": 164435335,
      "step": 7603,
      "time_per_iteration": 2.7189106941223145
    },
    {
      "auxiliary_loss_clip": 0.01114496,
      "auxiliary_loss_mlp": 0.01026806,
      "balance_loss_clip": 1.04104626,
      "balance_loss_mlp": 1.01966536,
      "epoch": 0.9143269404196477,
      "flos": 17712148809600.0,
      "grad_norm": 2.1530088616812355,
      "language_loss": 0.80265927,
      "learning_rate": 7.641142501432951e-08,
      "loss": 0.8240723,
      "num_input_tokens_seen": 164451530,
      "step": 7604,
      "time_per_iteration": 2.689401865005493
    },
    {
      "auxiliary_loss_clip": 0.01130147,
      "auxiliary_loss_mlp": 0.01021668,
      "balance_loss_clip": 1.03933859,
      "balance_loss_mlp": 1.01463437,
      "epoch": 0.9144471833102867,
      "flos": 33323414019840.0,
      "grad_norm": 1.6793178621137155,
      "language_loss": 0.73211104,
      "learning_rate": 7.619830938602013e-08,
      "loss": 0.75362921,
      "num_input_tokens_seen": 164472755,
      "step": 7605,
      "time_per_iteration": 2.7503225803375244
    },
    {
      "auxiliary_loss_clip": 0.01147906,
      "auxiliary_loss_mlp": 0.01026589,
      "balance_loss_clip": 1.04498947,
      "balance_loss_mlp": 1.01971674,
      "epoch": 0.9145674262009259,
      "flos": 21068970428160.0,
      "grad_norm": 1.900537556057236,
      "language_loss": 0.82570368,
      "learning_rate": 7.598548559689777e-08,
      "loss": 0.84744865,
      "num_input_tokens_seen": 164491155,
      "step": 7606,
      "time_per_iteration": 2.633129835128784
    },
    {
      "auxiliary_loss_clip": 0.01115656,
      "auxiliary_loss_mlp": 0.01030475,
      "balance_loss_clip": 1.03836143,
      "balance_loss_mlp": 1.02351022,
      "epoch": 0.914687669091565,
      "flos": 16800269212800.0,
      "grad_norm": 2.307077001409553,
      "language_loss": 0.81585443,
      "learning_rate": 7.577295367924751e-08,
      "loss": 0.83731574,
      "num_input_tokens_seen": 164507555,
      "step": 7607,
      "time_per_iteration": 2.7347731590270996
    },
    {
      "auxiliary_loss_clip": 0.01143713,
      "auxiliary_loss_mlp": 0.01027443,
      "balance_loss_clip": 1.04541183,
      "balance_loss_mlp": 1.01986074,
      "epoch": 0.914807911982204,
      "flos": 25773627012480.0,
      "grad_norm": 1.7406286391012669,
      "language_loss": 0.82133275,
      "learning_rate": 7.556071366531002e-08,
      "loss": 0.84304434,
      "num_input_tokens_seen": 164528525,
      "step": 7608,
      "time_per_iteration": 2.7245092391967773
    },
    {
      "auxiliary_loss_clip": 0.01152767,
      "auxiliary_loss_mlp": 0.0102783,
      "balance_loss_clip": 1.04593718,
      "balance_loss_mlp": 1.01995659,
      "epoch": 0.9149281548728432,
      "flos": 19208043636480.0,
      "grad_norm": 1.8498768516262158,
      "language_loss": 0.78999949,
      "learning_rate": 7.53487655872822e-08,
      "loss": 0.81180549,
      "num_input_tokens_seen": 164547695,
      "step": 7609,
      "time_per_iteration": 2.587688446044922
    },
    {
      "auxiliary_loss_clip": 0.01111049,
      "auxiliary_loss_mlp": 0.01022645,
      "balance_loss_clip": 1.03733718,
      "balance_loss_mlp": 1.01591003,
      "epoch": 0.9150483977634822,
      "flos": 26870554500480.0,
      "grad_norm": 1.7347481323925467,
      "language_loss": 0.73972881,
      "learning_rate": 7.513710947731656e-08,
      "loss": 0.76106572,
      "num_input_tokens_seen": 164568905,
      "step": 7610,
      "time_per_iteration": 2.7572197914123535
    },
    {
      "auxiliary_loss_clip": 0.01129691,
      "auxiliary_loss_mlp": 0.01024837,
      "balance_loss_clip": 1.04239154,
      "balance_loss_mlp": 1.01810789,
      "epoch": 0.9151686406541213,
      "flos": 21908956953600.0,
      "grad_norm": 1.7641459596383027,
      "language_loss": 0.85138571,
      "learning_rate": 7.492574536752095e-08,
      "loss": 0.872931,
      "num_input_tokens_seen": 164588895,
      "step": 7611,
      "time_per_iteration": 2.627189874649048
    },
    {
      "auxiliary_loss_clip": 0.01147264,
      "auxiliary_loss_mlp": 0.01028747,
      "balance_loss_clip": 1.044451,
      "balance_loss_mlp": 1.02182662,
      "epoch": 0.9152888835447605,
      "flos": 27308556944640.0,
      "grad_norm": 1.9513278755997778,
      "language_loss": 0.78031385,
      "learning_rate": 7.471467328995907e-08,
      "loss": 0.80207396,
      "num_input_tokens_seen": 164607705,
      "step": 7612,
      "time_per_iteration": 2.7257165908813477
    },
    {
      "auxiliary_loss_clip": 0.01075451,
      "auxiliary_loss_mlp": 0.01024737,
      "balance_loss_clip": 1.03526545,
      "balance_loss_mlp": 1.01787937,
      "epoch": 0.9154091264353995,
      "flos": 13370728510080.0,
      "grad_norm": 2.5091948169916996,
      "language_loss": 0.61080194,
      "learning_rate": 7.450389327665018e-08,
      "loss": 0.63180387,
      "num_input_tokens_seen": 164625540,
      "step": 7613,
      "time_per_iteration": 2.9022209644317627
    },
    {
      "auxiliary_loss_clip": 0.01126207,
      "auxiliary_loss_mlp": 0.0102541,
      "balance_loss_clip": 1.04572904,
      "balance_loss_mlp": 1.01812375,
      "epoch": 0.9155293693260386,
      "flos": 20193037367040.0,
      "grad_norm": 2.358564985631856,
      "language_loss": 0.67833745,
      "learning_rate": 7.429340535957029e-08,
      "loss": 0.6998536,
      "num_input_tokens_seen": 164640735,
      "step": 7614,
      "time_per_iteration": 3.121786594390869
    },
    {
      "auxiliary_loss_clip": 0.01140241,
      "auxiliary_loss_mlp": 0.01024068,
      "balance_loss_clip": 1.04324031,
      "balance_loss_mlp": 1.01650977,
      "epoch": 0.9156496122166777,
      "flos": 19354990176000.0,
      "grad_norm": 2.802766172719222,
      "language_loss": 0.70359105,
      "learning_rate": 7.40832095706494e-08,
      "loss": 0.72523415,
      "num_input_tokens_seen": 164657430,
      "step": 7615,
      "time_per_iteration": 2.663567066192627
    },
    {
      "auxiliary_loss_clip": 0.01127128,
      "auxiliary_loss_mlp": 0.01024691,
      "balance_loss_clip": 1.04125392,
      "balance_loss_mlp": 1.0177207,
      "epoch": 0.9157698551073168,
      "flos": 21107287261440.0,
      "grad_norm": 1.8852668289405823,
      "language_loss": 0.80191004,
      "learning_rate": 7.387330594177443e-08,
      "loss": 0.82342827,
      "num_input_tokens_seen": 164679505,
      "step": 7616,
      "time_per_iteration": 2.676135540008545
    },
    {
      "auxiliary_loss_clip": 0.01115238,
      "auxiliary_loss_mlp": 0.01023149,
      "balance_loss_clip": 1.03894305,
      "balance_loss_mlp": 1.01631284,
      "epoch": 0.9158900979979558,
      "flos": 25193167228800.0,
      "grad_norm": 1.728839788754277,
      "language_loss": 0.79380012,
      "learning_rate": 7.366369450478749e-08,
      "loss": 0.81518388,
      "num_input_tokens_seen": 164700615,
      "step": 7617,
      "time_per_iteration": 2.7504794597625732
    },
    {
      "auxiliary_loss_clip": 0.01116585,
      "auxiliary_loss_mlp": 0.01026199,
      "balance_loss_clip": 1.03920627,
      "balance_loss_mlp": 1.01946688,
      "epoch": 0.916010340888595,
      "flos": 30146648302080.0,
      "grad_norm": 1.5442085069660143,
      "language_loss": 0.66633689,
      "learning_rate": 7.345437529148646e-08,
      "loss": 0.68776476,
      "num_input_tokens_seen": 164719625,
      "step": 7618,
      "time_per_iteration": 3.721467971801758
    },
    {
      "auxiliary_loss_clip": 0.0112489,
      "auxiliary_loss_mlp": 0.01026657,
      "balance_loss_clip": 1.0419488,
      "balance_loss_mlp": 1.01977825,
      "epoch": 0.9161305837792341,
      "flos": 17091827907840.0,
      "grad_norm": 2.6593901856350812,
      "language_loss": 0.7274915,
      "learning_rate": 7.324534833362483e-08,
      "loss": 0.74900693,
      "num_input_tokens_seen": 164737200,
      "step": 7619,
      "time_per_iteration": 2.7362148761749268
    },
    {
      "auxiliary_loss_clip": 0.01137455,
      "auxiliary_loss_mlp": 0.01022947,
      "balance_loss_clip": 1.04205132,
      "balance_loss_mlp": 1.01657236,
      "epoch": 0.9162508266698731,
      "flos": 22893699288960.0,
      "grad_norm": 1.6415295583188405,
      "language_loss": 0.68614334,
      "learning_rate": 7.303661366291192e-08,
      "loss": 0.7077474,
      "num_input_tokens_seen": 164757870,
      "step": 7620,
      "time_per_iteration": 2.693411350250244
    },
    {
      "auxiliary_loss_clip": 0.01107475,
      "auxiliary_loss_mlp": 0.0102492,
      "balance_loss_clip": 1.03981805,
      "balance_loss_mlp": 1.01788664,
      "epoch": 0.9163710695605123,
      "flos": 19974808287360.0,
      "grad_norm": 1.7953852896391316,
      "language_loss": 0.81623489,
      "learning_rate": 7.28281713110126e-08,
      "loss": 0.83755887,
      "num_input_tokens_seen": 164775945,
      "step": 7621,
      "time_per_iteration": 3.728485107421875
    },
    {
      "auxiliary_loss_clip": 0.0113043,
      "auxiliary_loss_mlp": 0.01022789,
      "balance_loss_clip": 1.04280972,
      "balance_loss_mlp": 1.01614356,
      "epoch": 0.9164913124511513,
      "flos": 22783812606720.0,
      "grad_norm": 1.867148212658535,
      "language_loss": 0.77450442,
      "learning_rate": 7.262002130954759e-08,
      "loss": 0.7960366,
      "num_input_tokens_seen": 164794400,
      "step": 7622,
      "time_per_iteration": 2.6745216846466064
    },
    {
      "auxiliary_loss_clip": 0.01111425,
      "auxiliary_loss_mlp": 0.01026451,
      "balance_loss_clip": 1.04043388,
      "balance_loss_mlp": 1.01926887,
      "epoch": 0.9166115553417904,
      "flos": 24900854348160.0,
      "grad_norm": 2.1311247673049714,
      "language_loss": 0.78787553,
      "learning_rate": 7.241216369009296e-08,
      "loss": 0.80925429,
      "num_input_tokens_seen": 164814585,
      "step": 7623,
      "time_per_iteration": 2.73293399810791
    },
    {
      "auxiliary_loss_clip": 0.01163305,
      "auxiliary_loss_mlp": 0.01025931,
      "balance_loss_clip": 1.04226422,
      "balance_loss_mlp": 1.018924,
      "epoch": 0.9167317982324296,
      "flos": 25702919089920.0,
      "grad_norm": 1.8615897489775755,
      "language_loss": 0.66519636,
      "learning_rate": 7.220459848418037e-08,
      "loss": 0.68708873,
      "num_input_tokens_seen": 164834660,
      "step": 7624,
      "time_per_iteration": 3.549938440322876
    },
    {
      "auxiliary_loss_clip": 0.01164544,
      "auxiliary_loss_mlp": 0.01026804,
      "balance_loss_clip": 1.04658031,
      "balance_loss_mlp": 1.0202837,
      "epoch": 0.9168520411230686,
      "flos": 15632813370240.0,
      "grad_norm": 1.7799323220635896,
      "language_loss": 0.79827434,
      "learning_rate": 7.199732572329708e-08,
      "loss": 0.82018787,
      "num_input_tokens_seen": 164852560,
      "step": 7625,
      "time_per_iteration": 2.5520882606506348
    },
    {
      "auxiliary_loss_clip": 0.01123865,
      "auxiliary_loss_mlp": 0.01029197,
      "balance_loss_clip": 1.03993666,
      "balance_loss_mlp": 1.02202666,
      "epoch": 0.9169722840137077,
      "flos": 30258151096320.0,
      "grad_norm": 2.941219494880475,
      "language_loss": 0.75851977,
      "learning_rate": 7.179034543888684e-08,
      "loss": 0.7800504,
      "num_input_tokens_seen": 164872065,
      "step": 7626,
      "time_per_iteration": 3.6876707077026367
    },
    {
      "auxiliary_loss_clip": 0.01152278,
      "auxiliary_loss_mlp": 0.01025647,
      "balance_loss_clip": 1.04301643,
      "balance_loss_mlp": 1.01892972,
      "epoch": 0.9170925269043467,
      "flos": 22491643380480.0,
      "grad_norm": 2.096808553636907,
      "language_loss": 0.77309978,
      "learning_rate": 7.158365766234808e-08,
      "loss": 0.79487908,
      "num_input_tokens_seen": 164890915,
      "step": 7627,
      "time_per_iteration": 2.5866832733154297
    },
    {
      "auxiliary_loss_clip": 0.01115637,
      "auxiliary_loss_mlp": 0.01024898,
      "balance_loss_clip": 1.0377872,
      "balance_loss_mlp": 1.01720321,
      "epoch": 0.9172127697949859,
      "flos": 22893914770560.0,
      "grad_norm": 1.9842777395160318,
      "language_loss": 0.72528505,
      "learning_rate": 7.137726242503527e-08,
      "loss": 0.74669039,
      "num_input_tokens_seen": 164909835,
      "step": 7628,
      "time_per_iteration": 2.6439201831817627
    },
    {
      "auxiliary_loss_clip": 0.0115277,
      "auxiliary_loss_mlp": 0.00761873,
      "balance_loss_clip": 1.04654431,
      "balance_loss_mlp": 1.00031793,
      "epoch": 0.917333012685625,
      "flos": 17451867882240.0,
      "grad_norm": 2.447961429849324,
      "language_loss": 0.78464431,
      "learning_rate": 7.11711597582585e-08,
      "loss": 0.80379081,
      "num_input_tokens_seen": 164927195,
      "step": 7629,
      "time_per_iteration": 2.6704189777374268
    },
    {
      "auxiliary_loss_clip": 0.01122298,
      "auxiliary_loss_mlp": 0.01024904,
      "balance_loss_clip": 1.03776443,
      "balance_loss_mlp": 1.0183444,
      "epoch": 0.917453255576264,
      "flos": 14318949692160.0,
      "grad_norm": 1.6419150571552963,
      "language_loss": 0.80087495,
      "learning_rate": 7.096534969328271e-08,
      "loss": 0.82234699,
      "num_input_tokens_seen": 164944640,
      "step": 7630,
      "time_per_iteration": 2.6644692420959473
    },
    {
      "auxiliary_loss_clip": 0.01137082,
      "auxiliary_loss_mlp": 0.01030446,
      "balance_loss_clip": 1.03846693,
      "balance_loss_mlp": 1.02337658,
      "epoch": 0.9175734984669032,
      "flos": 20741177888640.0,
      "grad_norm": 2.0198863804563447,
      "language_loss": 0.83611512,
      "learning_rate": 7.075983226132987e-08,
      "loss": 0.85779035,
      "num_input_tokens_seen": 164963570,
      "step": 7631,
      "time_per_iteration": 2.680760383605957
    },
    {
      "auxiliary_loss_clip": 0.01142049,
      "auxiliary_loss_mlp": 0.00762939,
      "balance_loss_clip": 1.0424819,
      "balance_loss_mlp": 1.00035906,
      "epoch": 0.9176937413575422,
      "flos": 14830497233280.0,
      "grad_norm": 2.52335114141742,
      "language_loss": 0.79377848,
      "learning_rate": 7.055460749357656e-08,
      "loss": 0.81282842,
      "num_input_tokens_seen": 164979850,
      "step": 7632,
      "time_per_iteration": 2.6580758094787598
    },
    {
      "auxiliary_loss_clip": 0.01136496,
      "auxiliary_loss_mlp": 0.0102626,
      "balance_loss_clip": 1.04303777,
      "balance_loss_mlp": 1.01873446,
      "epoch": 0.9178139842481813,
      "flos": 18474603828480.0,
      "grad_norm": 1.869498651624272,
      "language_loss": 0.70246351,
      "learning_rate": 7.034967542115521e-08,
      "loss": 0.72409105,
      "num_input_tokens_seen": 164998115,
      "step": 7633,
      "time_per_iteration": 2.6734867095947266
    },
    {
      "auxiliary_loss_clip": 0.01140981,
      "auxiliary_loss_mlp": 0.00762484,
      "balance_loss_clip": 1.04158628,
      "balance_loss_mlp": 1.00033212,
      "epoch": 0.9179342271388204,
      "flos": 20047455544320.0,
      "grad_norm": 2.2114816819982983,
      "language_loss": 0.75524771,
      "learning_rate": 7.014503607515388e-08,
      "loss": 0.77428234,
      "num_input_tokens_seen": 165017420,
      "step": 7634,
      "time_per_iteration": 2.6444485187530518
    },
    {
      "auxiliary_loss_clip": 0.01138164,
      "auxiliary_loss_mlp": 0.01024408,
      "balance_loss_clip": 1.04567313,
      "balance_loss_mlp": 1.01735413,
      "epoch": 0.9180544700294595,
      "flos": 24676232647680.0,
      "grad_norm": 1.9738919811691342,
      "language_loss": 0.67790091,
      "learning_rate": 6.994068948661592e-08,
      "loss": 0.69952661,
      "num_input_tokens_seen": 165035575,
      "step": 7635,
      "time_per_iteration": 2.7300009727478027
    },
    {
      "auxiliary_loss_clip": 0.0115136,
      "auxiliary_loss_mlp": 0.01030427,
      "balance_loss_clip": 1.04479647,
      "balance_loss_mlp": 1.02267814,
      "epoch": 0.9181747129200986,
      "flos": 16727478301440.0,
      "grad_norm": 1.9951825633749778,
      "language_loss": 0.76715302,
      "learning_rate": 6.973663568654142e-08,
      "loss": 0.78897095,
      "num_input_tokens_seen": 165053280,
      "step": 7636,
      "time_per_iteration": 2.684580087661743
    },
    {
      "auxiliary_loss_clip": 0.01163982,
      "auxiliary_loss_mlp": 0.01024264,
      "balance_loss_clip": 1.04551339,
      "balance_loss_mlp": 1.01671219,
      "epoch": 0.9182949558107377,
      "flos": 24271626873600.0,
      "grad_norm": 2.448733952281188,
      "language_loss": 0.6515274,
      "learning_rate": 6.953287470588386e-08,
      "loss": 0.67340988,
      "num_input_tokens_seen": 165071235,
      "step": 7637,
      "time_per_iteration": 2.6215314865112305
    },
    {
      "auxiliary_loss_clip": 0.01152627,
      "auxiliary_loss_mlp": 0.0102303,
      "balance_loss_clip": 1.04227209,
      "balance_loss_mlp": 1.0157789,
      "epoch": 0.9184151987013768,
      "flos": 22082117443200.0,
      "grad_norm": 2.750667652079265,
      "language_loss": 0.85994065,
      "learning_rate": 6.932940657555452e-08,
      "loss": 0.88169718,
      "num_input_tokens_seen": 165087365,
      "step": 7638,
      "time_per_iteration": 2.656836748123169
    },
    {
      "auxiliary_loss_clip": 0.01159731,
      "auxiliary_loss_mlp": 0.01019139,
      "balance_loss_clip": 1.04389262,
      "balance_loss_mlp": 1.01264179,
      "epoch": 0.9185354415920158,
      "flos": 32166732257280.0,
      "grad_norm": 2.1639079552361027,
      "language_loss": 0.7611686,
      "learning_rate": 6.912623132641938e-08,
      "loss": 0.78295732,
      "num_input_tokens_seen": 165112455,
      "step": 7639,
      "time_per_iteration": 2.686985492706299
    },
    {
      "auxiliary_loss_clip": 0.01139448,
      "auxiliary_loss_mlp": 0.01023865,
      "balance_loss_clip": 1.04355955,
      "balance_loss_mlp": 1.01614034,
      "epoch": 0.918655684482655,
      "flos": 20997831542400.0,
      "grad_norm": 1.9807113032764863,
      "language_loss": 0.76794708,
      "learning_rate": 6.892334898929952e-08,
      "loss": 0.78958029,
      "num_input_tokens_seen": 165132700,
      "step": 7640,
      "time_per_iteration": 2.7348110675811768
    },
    {
      "auxiliary_loss_clip": 0.01145623,
      "auxiliary_loss_mlp": 0.01022254,
      "balance_loss_clip": 1.04226339,
      "balance_loss_mlp": 1.01523888,
      "epoch": 0.918775927373294,
      "flos": 15560704817280.0,
      "grad_norm": 2.0129661269443395,
      "language_loss": 0.84737813,
      "learning_rate": 6.872075959497236e-08,
      "loss": 0.86905694,
      "num_input_tokens_seen": 165151475,
      "step": 7641,
      "time_per_iteration": 2.7128875255584717
    },
    {
      "auxiliary_loss_clip": 0.01153216,
      "auxiliary_loss_mlp": 0.01025167,
      "balance_loss_clip": 1.04248774,
      "balance_loss_mlp": 1.01825881,
      "epoch": 0.9188961702639331,
      "flos": 29934057657600.0,
      "grad_norm": 2.258568265033229,
      "language_loss": 0.83064222,
      "learning_rate": 6.85184631741702e-08,
      "loss": 0.85242611,
      "num_input_tokens_seen": 165172040,
      "step": 7642,
      "time_per_iteration": 2.6453428268432617
    },
    {
      "auxiliary_loss_clip": 0.0114767,
      "auxiliary_loss_mlp": 0.0103171,
      "balance_loss_clip": 1.04225051,
      "balance_loss_mlp": 1.02467394,
      "epoch": 0.9190164131545723,
      "flos": 20701244943360.0,
      "grad_norm": 2.0796683284365662,
      "language_loss": 0.77517885,
      "learning_rate": 6.831645975758161e-08,
      "loss": 0.79697263,
      "num_input_tokens_seen": 165189980,
      "step": 7643,
      "time_per_iteration": 2.629753589630127
    },
    {
      "auxiliary_loss_clip": 0.01130956,
      "auxiliary_loss_mlp": 0.01023539,
      "balance_loss_clip": 1.04325032,
      "balance_loss_mlp": 1.01606476,
      "epoch": 0.9191366560452113,
      "flos": 25629912696960.0,
      "grad_norm": 1.9907374066992514,
      "language_loss": 0.67289507,
      "learning_rate": 6.811474937585026e-08,
      "loss": 0.69443995,
      "num_input_tokens_seen": 165209770,
      "step": 7644,
      "time_per_iteration": 2.652667760848999
    },
    {
      "auxiliary_loss_clip": 0.01119277,
      "auxiliary_loss_mlp": 0.01023546,
      "balance_loss_clip": 1.0411669,
      "balance_loss_mlp": 1.0168016,
      "epoch": 0.9192568989358504,
      "flos": 21434325615360.0,
      "grad_norm": 2.4927396848398664,
      "language_loss": 0.79362714,
      "learning_rate": 6.79133320595755e-08,
      "loss": 0.81505537,
      "num_input_tokens_seen": 165229690,
      "step": 7645,
      "time_per_iteration": 3.74513578414917
    },
    {
      "auxiliary_loss_clip": 0.01137623,
      "auxiliary_loss_mlp": 0.01021838,
      "balance_loss_clip": 1.04342985,
      "balance_loss_mlp": 1.01495349,
      "epoch": 0.9193771418264896,
      "flos": 23185078416000.0,
      "grad_norm": 1.7976703587555092,
      "language_loss": 0.75304443,
      "learning_rate": 6.771220783931198e-08,
      "loss": 0.77463901,
      "num_input_tokens_seen": 165249850,
      "step": 7646,
      "time_per_iteration": 2.66698956489563
    },
    {
      "auxiliary_loss_clip": 0.00996804,
      "auxiliary_loss_mlp": 0.0075364,
      "balance_loss_clip": 1.01898706,
      "balance_loss_mlp": 1.00017786,
      "epoch": 0.9194973847171286,
      "flos": 70582963184640.0,
      "grad_norm": 0.867275891797583,
      "language_loss": 0.64610088,
      "learning_rate": 6.751137674556994e-08,
      "loss": 0.66360533,
      "num_input_tokens_seen": 165310235,
      "step": 7647,
      "time_per_iteration": 4.630884647369385
    },
    {
      "auxiliary_loss_clip": 0.0115395,
      "auxiliary_loss_mlp": 0.01025895,
      "balance_loss_clip": 1.04309464,
      "balance_loss_mlp": 1.01827717,
      "epoch": 0.9196176276077677,
      "flos": 14720682378240.0,
      "grad_norm": 70.57632204719853,
      "language_loss": 0.77120328,
      "learning_rate": 6.731083880881572e-08,
      "loss": 0.79300171,
      "num_input_tokens_seen": 165326455,
      "step": 7648,
      "time_per_iteration": 4.192774295806885
    },
    {
      "auxiliary_loss_clip": 0.01136006,
      "auxiliary_loss_mlp": 0.01023382,
      "balance_loss_clip": 1.04259741,
      "balance_loss_mlp": 1.01659584,
      "epoch": 0.9197378704984068,
      "flos": 23294893271040.0,
      "grad_norm": 1.9800479003521174,
      "language_loss": 0.81027013,
      "learning_rate": 6.711059405947072e-08,
      "loss": 0.831864,
      "num_input_tokens_seen": 165344645,
      "step": 7649,
      "time_per_iteration": 2.6996259689331055
    },
    {
      "auxiliary_loss_clip": 0.01120303,
      "auxiliary_loss_mlp": 0.01020789,
      "balance_loss_clip": 1.04294538,
      "balance_loss_mlp": 1.01405656,
      "epoch": 0.9198581133890459,
      "flos": 20302564913280.0,
      "grad_norm": 3.3700757297459853,
      "language_loss": 0.77239776,
      "learning_rate": 6.691064252791156e-08,
      "loss": 0.79380864,
      "num_input_tokens_seen": 165364120,
      "step": 7650,
      "time_per_iteration": 3.7209854125976562
    },
    {
      "auxiliary_loss_clip": 0.01100536,
      "auxiliary_loss_mlp": 0.01022365,
      "balance_loss_clip": 1.03865767,
      "balance_loss_mlp": 1.01515007,
      "epoch": 0.9199783562796849,
      "flos": 17675663569920.0,
      "grad_norm": 1.5602623581482853,
      "language_loss": 0.7813217,
      "learning_rate": 6.67109842444713e-08,
      "loss": 0.80255079,
      "num_input_tokens_seen": 165383050,
      "step": 7651,
      "time_per_iteration": 2.925755262374878
    },
    {
      "auxiliary_loss_clip": 0.01146705,
      "auxiliary_loss_mlp": 0.00762315,
      "balance_loss_clip": 1.0441488,
      "balance_loss_mlp": 1.00032091,
      "epoch": 0.9200985991703241,
      "flos": 17676022705920.0,
      "grad_norm": 1.8653425193656061,
      "language_loss": 0.7673403,
      "learning_rate": 6.651161923943704e-08,
      "loss": 0.78643048,
      "num_input_tokens_seen": 165400955,
      "step": 7652,
      "time_per_iteration": 3.6874988079071045
    },
    {
      "auxiliary_loss_clip": 0.01146305,
      "auxiliary_loss_mlp": 0.01022659,
      "balance_loss_clip": 1.04178286,
      "balance_loss_mlp": 1.01551247,
      "epoch": 0.9202188420609632,
      "flos": 20996574566400.0,
      "grad_norm": 1.6967670848506242,
      "language_loss": 0.76773512,
      "learning_rate": 6.631254754305326e-08,
      "loss": 0.78942478,
      "num_input_tokens_seen": 165420415,
      "step": 7653,
      "time_per_iteration": 2.711350202560425
    },
    {
      "auxiliary_loss_clip": 0.01167242,
      "auxiliary_loss_mlp": 0.01023096,
      "balance_loss_clip": 1.04577065,
      "balance_loss_mlp": 1.01557946,
      "epoch": 0.9203390849516022,
      "flos": 13918222586880.0,
      "grad_norm": 2.1981564807417704,
      "language_loss": 0.78283787,
      "learning_rate": 6.611376918551848e-08,
      "loss": 0.8047412,
      "num_input_tokens_seen": 165439200,
      "step": 7654,
      "time_per_iteration": 2.568079948425293
    },
    {
      "auxiliary_loss_clip": 0.01120523,
      "auxiliary_loss_mlp": 0.00762521,
      "balance_loss_clip": 1.03998601,
      "balance_loss_mlp": 1.00033617,
      "epoch": 0.9204593278422414,
      "flos": 21175912195200.0,
      "grad_norm": 2.085154580491555,
      "language_loss": 0.79172349,
      "learning_rate": 6.591528419698744e-08,
      "loss": 0.81055391,
      "num_input_tokens_seen": 165458985,
      "step": 7655,
      "time_per_iteration": 2.7401700019836426
    },
    {
      "auxiliary_loss_clip": 0.01137792,
      "auxiliary_loss_mlp": 0.01025445,
      "balance_loss_clip": 1.04041171,
      "balance_loss_mlp": 1.01899588,
      "epoch": 0.9205795707328804,
      "flos": 14501375890560.0,
      "grad_norm": 2.190766671938305,
      "language_loss": 0.83071071,
      "learning_rate": 6.571709260756986e-08,
      "loss": 0.85234308,
      "num_input_tokens_seen": 165475630,
      "step": 7656,
      "time_per_iteration": 2.7331273555755615
    },
    {
      "auxiliary_loss_clip": 0.01156201,
      "auxiliary_loss_mlp": 0.01027043,
      "balance_loss_clip": 1.04998958,
      "balance_loss_mlp": 1.0197835,
      "epoch": 0.9206998136235195,
      "flos": 22417559579520.0,
      "grad_norm": 3.0191293233101946,
      "language_loss": 0.75988764,
      "learning_rate": 6.551919444733122e-08,
      "loss": 0.7817201,
      "num_input_tokens_seen": 165493445,
      "step": 7657,
      "time_per_iteration": 2.6642906665802
    },
    {
      "auxiliary_loss_clip": 0.01138683,
      "auxiliary_loss_mlp": 0.01025722,
      "balance_loss_clip": 1.04506397,
      "balance_loss_mlp": 1.01845634,
      "epoch": 0.9208200565141585,
      "flos": 53358407544960.0,
      "grad_norm": 1.8877926264656681,
      "language_loss": 0.65719759,
      "learning_rate": 6.53215897462931e-08,
      "loss": 0.67884165,
      "num_input_tokens_seen": 165517200,
      "step": 7658,
      "time_per_iteration": 3.002746105194092
    },
    {
      "auxiliary_loss_clip": 0.01147631,
      "auxiliary_loss_mlp": 0.01021545,
      "balance_loss_clip": 1.04217911,
      "balance_loss_mlp": 1.01453865,
      "epoch": 0.9209402994047977,
      "flos": 30589139946240.0,
      "grad_norm": 2.2884652133290686,
      "language_loss": 0.75016123,
      "learning_rate": 6.512427853443103e-08,
      "loss": 0.77185297,
      "num_input_tokens_seen": 165539280,
      "step": 7659,
      "time_per_iteration": 2.835571765899658
    },
    {
      "auxiliary_loss_clip": 0.01152354,
      "auxiliary_loss_mlp": 0.0102613,
      "balance_loss_clip": 1.04364312,
      "balance_loss_mlp": 1.01909065,
      "epoch": 0.9210605422954368,
      "flos": 29132711187840.0,
      "grad_norm": 1.5970410887814221,
      "language_loss": 0.75950086,
      "learning_rate": 6.492726084167799e-08,
      "loss": 0.78128564,
      "num_input_tokens_seen": 165561395,
      "step": 7660,
      "time_per_iteration": 2.7324769496917725
    },
    {
      "auxiliary_loss_clip": 0.01063044,
      "auxiliary_loss_mlp": 0.01000822,
      "balance_loss_clip": 1.00922656,
      "balance_loss_mlp": 0.99980301,
      "epoch": 0.9211807851860758,
      "flos": 54853838472960.0,
      "grad_norm": 0.7788976093199151,
      "language_loss": 0.57500839,
      "learning_rate": 6.473053669792072e-08,
      "loss": 0.59564704,
      "num_input_tokens_seen": 165616085,
      "step": 7661,
      "time_per_iteration": 3.1087706089019775
    },
    {
      "auxiliary_loss_clip": 0.01150209,
      "auxiliary_loss_mlp": 0.01024553,
      "balance_loss_clip": 1.04261994,
      "balance_loss_mlp": 1.016927,
      "epoch": 0.921301028076715,
      "flos": 19201974238080.0,
      "grad_norm": 2.9915358886369012,
      "language_loss": 0.73130906,
      "learning_rate": 6.453410613300248e-08,
      "loss": 0.75305665,
      "num_input_tokens_seen": 165634015,
      "step": 7662,
      "time_per_iteration": 2.6802449226379395
    },
    {
      "auxiliary_loss_clip": 0.01095627,
      "auxiliary_loss_mlp": 0.01026433,
      "balance_loss_clip": 1.03967595,
      "balance_loss_mlp": 1.01968312,
      "epoch": 0.921421270967354,
      "flos": 27526893765120.0,
      "grad_norm": 1.7618804112599578,
      "language_loss": 0.58599901,
      "learning_rate": 6.43379691767214e-08,
      "loss": 0.60721964,
      "num_input_tokens_seen": 165653220,
      "step": 7663,
      "time_per_iteration": 2.894577741622925
    },
    {
      "auxiliary_loss_clip": 0.0102345,
      "auxiliary_loss_mlp": 0.01000491,
      "balance_loss_clip": 1.00990915,
      "balance_loss_mlp": 0.99949569,
      "epoch": 0.9215415138579931,
      "flos": 70209311955840.0,
      "grad_norm": 0.7141141417373716,
      "language_loss": 0.55136305,
      "learning_rate": 6.414212585883105e-08,
      "loss": 0.57160246,
      "num_input_tokens_seen": 165715850,
      "step": 7664,
      "time_per_iteration": 3.3525962829589844
    },
    {
      "auxiliary_loss_clip": 0.01138997,
      "auxiliary_loss_mlp": 0.01023297,
      "balance_loss_clip": 1.04227829,
      "balance_loss_mlp": 1.01615047,
      "epoch": 0.9216617567486323,
      "flos": 35553107790720.0,
      "grad_norm": 1.8456535018688633,
      "language_loss": 0.69853044,
      "learning_rate": 6.394657620904143e-08,
      "loss": 0.72015333,
      "num_input_tokens_seen": 165738960,
      "step": 7665,
      "time_per_iteration": 2.802016258239746
    },
    {
      "auxiliary_loss_clip": 0.01168234,
      "auxiliary_loss_mlp": 0.01026099,
      "balance_loss_clip": 1.04581404,
      "balance_loss_mlp": 1.01884794,
      "epoch": 0.9217819996392713,
      "flos": 29533330552320.0,
      "grad_norm": 2.393104760792951,
      "language_loss": 0.7203747,
      "learning_rate": 6.375132025701657e-08,
      "loss": 0.74231803,
      "num_input_tokens_seen": 165761260,
      "step": 7666,
      "time_per_iteration": 2.674694299697876
    },
    {
      "auxiliary_loss_clip": 0.01169845,
      "auxiliary_loss_mlp": 0.01025043,
      "balance_loss_clip": 1.04806769,
      "balance_loss_mlp": 1.0174017,
      "epoch": 0.9219022425299104,
      "flos": 14574669592320.0,
      "grad_norm": 2.250622879798371,
      "language_loss": 0.69104517,
      "learning_rate": 6.355635803237724e-08,
      "loss": 0.7129941,
      "num_input_tokens_seen": 165776960,
      "step": 7667,
      "time_per_iteration": 2.6515181064605713
    },
    {
      "auxiliary_loss_clip": 0.0115057,
      "auxiliary_loss_mlp": 0.01031006,
      "balance_loss_clip": 1.04344392,
      "balance_loss_mlp": 1.02328718,
      "epoch": 0.9220224854205495,
      "flos": 18077503996800.0,
      "grad_norm": 3.2091062199766895,
      "language_loss": 0.79625744,
      "learning_rate": 6.336168956469867e-08,
      "loss": 0.81807315,
      "num_input_tokens_seen": 165795435,
      "step": 7668,
      "time_per_iteration": 2.6381819248199463
    },
    {
      "auxiliary_loss_clip": 0.01128403,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.04125762,
      "balance_loss_mlp": 1.01809072,
      "epoch": 0.9221427283111886,
      "flos": 24790464875520.0,
      "grad_norm": 1.7808514836310947,
      "language_loss": 0.71702391,
      "learning_rate": 6.316731488351168e-08,
      "loss": 0.73856336,
      "num_input_tokens_seen": 165816625,
      "step": 7669,
      "time_per_iteration": 2.751556873321533
    },
    {
      "auxiliary_loss_clip": 0.01149715,
      "auxiliary_loss_mlp": 0.01023992,
      "balance_loss_clip": 1.04318225,
      "balance_loss_mlp": 1.01687849,
      "epoch": 0.9222629712018277,
      "flos": 13845036625920.0,
      "grad_norm": 1.6897075047621537,
      "language_loss": 0.63340747,
      "learning_rate": 6.297323401830334e-08,
      "loss": 0.65514451,
      "num_input_tokens_seen": 165835410,
      "step": 7670,
      "time_per_iteration": 3.536201000213623
    },
    {
      "auxiliary_loss_clip": 0.01151627,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 1.04266334,
      "balance_loss_mlp": 1.02054918,
      "epoch": 0.9223832140924668,
      "flos": 21616177196160.0,
      "grad_norm": 2.086363842336269,
      "language_loss": 0.68748069,
      "learning_rate": 6.277944699851523e-08,
      "loss": 0.70927685,
      "num_input_tokens_seen": 165854930,
      "step": 7671,
      "time_per_iteration": 2.622403383255005
    },
    {
      "auxiliary_loss_clip": 0.01165635,
      "auxiliary_loss_mlp": 0.01025591,
      "balance_loss_clip": 1.04562235,
      "balance_loss_mlp": 1.01754999,
      "epoch": 0.9225034569831059,
      "flos": 21142084561920.0,
      "grad_norm": 1.824916280736535,
      "language_loss": 0.73132509,
      "learning_rate": 6.25859538535447e-08,
      "loss": 0.75323737,
      "num_input_tokens_seen": 165875725,
      "step": 7672,
      "time_per_iteration": 2.6278393268585205
    },
    {
      "auxiliary_loss_clip": 0.01135963,
      "auxiliary_loss_mlp": 0.01026589,
      "balance_loss_clip": 1.04331851,
      "balance_loss_mlp": 1.01968372,
      "epoch": 0.9226236998737449,
      "flos": 12495046844160.0,
      "grad_norm": 2.6617504956215927,
      "language_loss": 0.78394806,
      "learning_rate": 6.239275461274474e-08,
      "loss": 0.8055737,
      "num_input_tokens_seen": 165892100,
      "step": 7673,
      "time_per_iteration": 4.027694225311279
    },
    {
      "auxiliary_loss_clip": 0.01149761,
      "auxiliary_loss_mlp": 0.01024672,
      "balance_loss_clip": 1.04413927,
      "balance_loss_mlp": 1.01800776,
      "epoch": 0.9227439427643841,
      "flos": 26214071581440.0,
      "grad_norm": 1.644299290024939,
      "language_loss": 0.85973966,
      "learning_rate": 6.219984930542299e-08,
      "loss": 0.88148403,
      "num_input_tokens_seen": 165912840,
      "step": 7674,
      "time_per_iteration": 2.726835012435913
    },
    {
      "auxiliary_loss_clip": 0.01155129,
      "auxiliary_loss_mlp": 0.01022842,
      "balance_loss_clip": 1.04499102,
      "balance_loss_mlp": 1.01555538,
      "epoch": 0.9228641856550232,
      "flos": 17967581400960.0,
      "grad_norm": 2.0040158320986907,
      "language_loss": 0.76088351,
      "learning_rate": 6.200723796084383e-08,
      "loss": 0.78266323,
      "num_input_tokens_seen": 165930935,
      "step": 7675,
      "time_per_iteration": 2.6631922721862793
    },
    {
      "auxiliary_loss_clip": 0.01037869,
      "auxiliary_loss_mlp": 0.01000994,
      "balance_loss_clip": 1.01008344,
      "balance_loss_mlp": 1.00000465,
      "epoch": 0.9229844285456622,
      "flos": 70420609710720.0,
      "grad_norm": 0.7606740451221454,
      "language_loss": 0.62987721,
      "learning_rate": 6.181492060822546e-08,
      "loss": 0.65026581,
      "num_input_tokens_seen": 165991110,
      "step": 7676,
      "time_per_iteration": 4.32470965385437
    },
    {
      "auxiliary_loss_clip": 0.01106338,
      "auxiliary_loss_mlp": 0.01025192,
      "balance_loss_clip": 1.03828168,
      "balance_loss_mlp": 1.01772368,
      "epoch": 0.9231046714363014,
      "flos": 17967832796160.0,
      "grad_norm": 2.4555163423599895,
      "language_loss": 0.81684756,
      "learning_rate": 6.162289727674274e-08,
      "loss": 0.83816284,
      "num_input_tokens_seen": 166008790,
      "step": 7677,
      "time_per_iteration": 2.6765387058258057
    },
    {
      "auxiliary_loss_clip": 0.01122744,
      "auxiliary_loss_mlp": 0.01024339,
      "balance_loss_clip": 1.04137444,
      "balance_loss_mlp": 1.01794076,
      "epoch": 0.9232249143269404,
      "flos": 17858233422720.0,
      "grad_norm": 2.232418189839277,
      "language_loss": 0.87472391,
      "learning_rate": 6.143116799552527e-08,
      "loss": 0.8961947,
      "num_input_tokens_seen": 166025035,
      "step": 7678,
      "time_per_iteration": 3.789640426635742
    },
    {
      "auxiliary_loss_clip": 0.01155602,
      "auxiliary_loss_mlp": 0.01024751,
      "balance_loss_clip": 1.04488897,
      "balance_loss_mlp": 1.01753891,
      "epoch": 0.9233451572175795,
      "flos": 23404384903680.0,
      "grad_norm": 2.6675227806483353,
      "language_loss": 0.56304872,
      "learning_rate": 6.123973279365802e-08,
      "loss": 0.58485222,
      "num_input_tokens_seen": 166044010,
      "step": 7679,
      "time_per_iteration": 2.6596646308898926
    },
    {
      "auxiliary_loss_clip": 0.01153774,
      "auxiliary_loss_mlp": 0.01024702,
      "balance_loss_clip": 1.04497588,
      "balance_loss_mlp": 1.01772547,
      "epoch": 0.9234654001082186,
      "flos": 17999326045440.0,
      "grad_norm": 1.8551810414797425,
      "language_loss": 0.77686268,
      "learning_rate": 6.10485917001824e-08,
      "loss": 0.7986474,
      "num_input_tokens_seen": 166061865,
      "step": 7680,
      "time_per_iteration": 2.670602321624756
    },
    {
      "auxiliary_loss_clip": 0.01140925,
      "auxiliary_loss_mlp": 0.01024871,
      "balance_loss_clip": 1.04155183,
      "balance_loss_mlp": 1.01825237,
      "epoch": 0.9235856429988577,
      "flos": 24750747411840.0,
      "grad_norm": 1.6655642386366822,
      "language_loss": 0.80844843,
      "learning_rate": 6.085774474409322e-08,
      "loss": 0.83010644,
      "num_input_tokens_seen": 166082425,
      "step": 7681,
      "time_per_iteration": 2.7491488456726074
    },
    {
      "auxiliary_loss_clip": 0.011375,
      "auxiliary_loss_mlp": 0.01027086,
      "balance_loss_clip": 1.04409826,
      "balance_loss_mlp": 1.01980853,
      "epoch": 0.9237058858894968,
      "flos": 14099894599680.0,
      "grad_norm": 1.8356102126232892,
      "language_loss": 0.703978,
      "learning_rate": 6.066719195434267e-08,
      "loss": 0.72562391,
      "num_input_tokens_seen": 166100225,
      "step": 7682,
      "time_per_iteration": 2.7139244079589844
    },
    {
      "auxiliary_loss_clip": 0.01156666,
      "auxiliary_loss_mlp": 0.01033974,
      "balance_loss_clip": 1.04629803,
      "balance_loss_mlp": 1.02591884,
      "epoch": 0.9238261287801359,
      "flos": 28694529175680.0,
      "grad_norm": 2.0994964481574447,
      "language_loss": 0.66366994,
      "learning_rate": 6.047693335983717e-08,
      "loss": 0.68557632,
      "num_input_tokens_seen": 166122570,
      "step": 7683,
      "time_per_iteration": 2.8360519409179688
    },
    {
      "auxiliary_loss_clip": 0.01151223,
      "auxiliary_loss_mlp": 0.01023838,
      "balance_loss_clip": 1.04180574,
      "balance_loss_mlp": 1.01650023,
      "epoch": 0.923946371670775,
      "flos": 23111856541440.0,
      "grad_norm": 2.4617144184375834,
      "language_loss": 0.8238374,
      "learning_rate": 6.028696898943853e-08,
      "loss": 0.84558797,
      "num_input_tokens_seen": 166141630,
      "step": 7684,
      "time_per_iteration": 2.680131196975708
    },
    {
      "auxiliary_loss_clip": 0.01136026,
      "auxiliary_loss_mlp": 0.00762714,
      "balance_loss_clip": 1.0415529,
      "balance_loss_mlp": 1.00036728,
      "epoch": 0.924066614561414,
      "flos": 21867120587520.0,
      "grad_norm": 2.022335845583231,
      "language_loss": 0.70889747,
      "learning_rate": 6.00972988719648e-08,
      "loss": 0.72788489,
      "num_input_tokens_seen": 166159865,
      "step": 7685,
      "time_per_iteration": 2.7904062271118164
    },
    {
      "auxiliary_loss_clip": 0.01126611,
      "auxiliary_loss_mlp": 0.00762828,
      "balance_loss_clip": 1.04316783,
      "balance_loss_mlp": 1.00034618,
      "epoch": 0.9241868574520532,
      "flos": 28511887495680.0,
      "grad_norm": 3.230472128418823,
      "language_loss": 0.7063309,
      "learning_rate": 5.990792303618807e-08,
      "loss": 0.72522527,
      "num_input_tokens_seen": 166179445,
      "step": 7686,
      "time_per_iteration": 2.8453075885772705
    },
    {
      "auxiliary_loss_clip": 0.01120322,
      "auxiliary_loss_mlp": 0.01021851,
      "balance_loss_clip": 1.04075718,
      "balance_loss_mlp": 1.01530051,
      "epoch": 0.9243071003426923,
      "flos": 30518324282880.0,
      "grad_norm": 1.7530778305320385,
      "language_loss": 0.6972394,
      "learning_rate": 5.971884151083695e-08,
      "loss": 0.71866119,
      "num_input_tokens_seen": 166201855,
      "step": 7687,
      "time_per_iteration": 2.809601068496704
    },
    {
      "auxiliary_loss_clip": 0.0114012,
      "auxiliary_loss_mlp": 0.01023131,
      "balance_loss_clip": 1.04236031,
      "balance_loss_mlp": 1.0165211,
      "epoch": 0.9244273432333313,
      "flos": 28658331244800.0,
      "grad_norm": 1.9290235363172512,
      "language_loss": 0.7426399,
      "learning_rate": 5.9530054324595124e-08,
      "loss": 0.76427245,
      "num_input_tokens_seen": 166221970,
      "step": 7688,
      "time_per_iteration": 2.8052711486816406
    },
    {
      "auxiliary_loss_clip": 0.01048168,
      "auxiliary_loss_mlp": 0.00753568,
      "balance_loss_clip": 1.00862217,
      "balance_loss_mlp": 1.00020397,
      "epoch": 0.9245475861239704,
      "flos": 66230589237120.0,
      "grad_norm": 0.7175134977871213,
      "language_loss": 0.5755372,
      "learning_rate": 5.934156150610103e-08,
      "loss": 0.59355462,
      "num_input_tokens_seen": 166279335,
      "step": 7689,
      "time_per_iteration": 3.410245418548584
    },
    {
      "auxiliary_loss_clip": 0.01133828,
      "auxiliary_loss_mlp": 0.0103039,
      "balance_loss_clip": 1.04050279,
      "balance_loss_mlp": 1.0231303,
      "epoch": 0.9246678290146095,
      "flos": 24239918142720.0,
      "grad_norm": 2.374628247061118,
      "language_loss": 0.79123747,
      "learning_rate": 5.915336308394914e-08,
      "loss": 0.81287974,
      "num_input_tokens_seen": 166298170,
      "step": 7690,
      "time_per_iteration": 2.792848825454712
    },
    {
      "auxiliary_loss_clip": 0.0114488,
      "auxiliary_loss_mlp": 0.01021306,
      "balance_loss_clip": 1.0429188,
      "balance_loss_mlp": 1.01473486,
      "epoch": 0.9247880719052486,
      "flos": 18988808976000.0,
      "grad_norm": 1.595738349584563,
      "language_loss": 0.77083743,
      "learning_rate": 5.89654590866886e-08,
      "loss": 0.7924993,
      "num_input_tokens_seen": 166317670,
      "step": 7691,
      "time_per_iteration": 2.6710033416748047
    },
    {
      "auxiliary_loss_clip": 0.01099415,
      "auxiliary_loss_mlp": 0.01025507,
      "balance_loss_clip": 1.04200244,
      "balance_loss_mlp": 1.01795471,
      "epoch": 0.9249083147958876,
      "flos": 24024095274240.0,
      "grad_norm": 1.977395743562468,
      "language_loss": 0.88056278,
      "learning_rate": 5.877784954282483e-08,
      "loss": 0.90181202,
      "num_input_tokens_seen": 166337010,
      "step": 7692,
      "time_per_iteration": 2.828361749649048
    },
    {
      "auxiliary_loss_clip": 0.011534,
      "auxiliary_loss_mlp": 0.01027702,
      "balance_loss_clip": 1.04268837,
      "balance_loss_mlp": 1.02012002,
      "epoch": 0.9250285576865268,
      "flos": 30773972355840.0,
      "grad_norm": 2.3693818580247368,
      "language_loss": 0.72656834,
      "learning_rate": 5.8590534480817963e-08,
      "loss": 0.74837935,
      "num_input_tokens_seen": 166358735,
      "step": 7693,
      "time_per_iteration": 2.7988007068634033
    },
    {
      "auxiliary_loss_clip": 0.0116767,
      "auxiliary_loss_mlp": 0.0102939,
      "balance_loss_clip": 1.04789639,
      "balance_loss_mlp": 1.02187657,
      "epoch": 0.9251488005771659,
      "flos": 10633581348480.0,
      "grad_norm": 2.347855859426235,
      "language_loss": 0.72441661,
      "learning_rate": 5.840351392908349e-08,
      "loss": 0.74638718,
      "num_input_tokens_seen": 166374455,
      "step": 7694,
      "time_per_iteration": 2.6447720527648926
    },
    {
      "auxiliary_loss_clip": 0.01143829,
      "auxiliary_loss_mlp": 0.00762014,
      "balance_loss_clip": 1.04234385,
      "balance_loss_mlp": 1.00035703,
      "epoch": 0.9252690434678049,
      "flos": 23586416052480.0,
      "grad_norm": 6.237704858237511,
      "language_loss": 0.70988679,
      "learning_rate": 5.821678791599205e-08,
      "loss": 0.72894514,
      "num_input_tokens_seen": 166393900,
      "step": 7695,
      "time_per_iteration": 2.7154760360717773
    },
    {
      "auxiliary_loss_clip": 0.01136815,
      "auxiliary_loss_mlp": 0.01027565,
      "balance_loss_clip": 1.04388678,
      "balance_loss_mlp": 1.0205729,
      "epoch": 0.9253892863584441,
      "flos": 21469158829440.0,
      "grad_norm": 1.6283441928302789,
      "language_loss": 0.8097589,
      "learning_rate": 5.803035646986965e-08,
      "loss": 0.83140272,
      "num_input_tokens_seen": 166413235,
      "step": 7696,
      "time_per_iteration": 3.6199681758880615
    },
    {
      "auxiliary_loss_clip": 0.01165244,
      "auxiliary_loss_mlp": 0.01027533,
      "balance_loss_clip": 1.04470158,
      "balance_loss_mlp": 1.02030325,
      "epoch": 0.9255095292490831,
      "flos": 17456680304640.0,
      "grad_norm": 2.5567230482555066,
      "language_loss": 0.67423368,
      "learning_rate": 5.7844219618998766e-08,
      "loss": 0.69616139,
      "num_input_tokens_seen": 166427560,
      "step": 7697,
      "time_per_iteration": 2.623157262802124
    },
    {
      "auxiliary_loss_clip": 0.01109336,
      "auxiliary_loss_mlp": 0.01033842,
      "balance_loss_clip": 1.03674126,
      "balance_loss_mlp": 1.02708602,
      "epoch": 0.9256297721397222,
      "flos": 24750675584640.0,
      "grad_norm": 1.847020266103541,
      "language_loss": 0.71910119,
      "learning_rate": 5.765837739161505e-08,
      "loss": 0.74053299,
      "num_input_tokens_seen": 166446680,
      "step": 7698,
      "time_per_iteration": 3.7242591381073
    },
    {
      "auxiliary_loss_clip": 0.01123075,
      "auxiliary_loss_mlp": 0.01022052,
      "balance_loss_clip": 1.04039359,
      "balance_loss_mlp": 1.01491773,
      "epoch": 0.9257500150303614,
      "flos": 23112215677440.0,
      "grad_norm": 1.8150193322032848,
      "language_loss": 0.7442897,
      "learning_rate": 5.7472829815911504e-08,
      "loss": 0.76574099,
      "num_input_tokens_seen": 166465505,
      "step": 7699,
      "time_per_iteration": 2.7585813999176025
    },
    {
      "auxiliary_loss_clip": 0.01130298,
      "auxiliary_loss_mlp": 0.01028837,
      "balance_loss_clip": 1.04032016,
      "balance_loss_mlp": 1.02172577,
      "epoch": 0.9258702579210004,
      "flos": 22564685687040.0,
      "grad_norm": 1.6215126110760196,
      "language_loss": 0.81709594,
      "learning_rate": 5.7287576920035164e-08,
      "loss": 0.8386873,
      "num_input_tokens_seen": 166484520,
      "step": 7700,
      "time_per_iteration": 2.8206911087036133
    },
    {
      "auxiliary_loss_clip": 0.01118429,
      "auxiliary_loss_mlp": 0.01031752,
      "balance_loss_clip": 1.03959584,
      "balance_loss_mlp": 1.02483237,
      "epoch": 0.9259905008116395,
      "flos": 30004298703360.0,
      "grad_norm": 1.857430771510208,
      "language_loss": 0.76705909,
      "learning_rate": 5.7102618732088435e-08,
      "loss": 0.78856099,
      "num_input_tokens_seen": 166503850,
      "step": 7701,
      "time_per_iteration": 2.8854329586029053
    },
    {
      "auxiliary_loss_clip": 0.01142672,
      "auxiliary_loss_mlp": 0.01024779,
      "balance_loss_clip": 1.04337943,
      "balance_loss_mlp": 1.01835966,
      "epoch": 0.9261107437022786,
      "flos": 24572128055040.0,
      "grad_norm": 1.8672752245825892,
      "language_loss": 0.74538279,
      "learning_rate": 5.6917955280130216e-08,
      "loss": 0.76705724,
      "num_input_tokens_seen": 166525330,
      "step": 7702,
      "time_per_iteration": 3.691372871398926
    },
    {
      "auxiliary_loss_clip": 0.0114916,
      "auxiliary_loss_mlp": 0.01023412,
      "balance_loss_clip": 1.04585612,
      "balance_loss_mlp": 1.01657844,
      "epoch": 0.9262309865929177,
      "flos": 22018448586240.0,
      "grad_norm": 2.3341077036594244,
      "language_loss": 0.72752434,
      "learning_rate": 5.6733586592172755e-08,
      "loss": 0.74925005,
      "num_input_tokens_seen": 166544825,
      "step": 7703,
      "time_per_iteration": 2.6406376361846924
    },
    {
      "auxiliary_loss_clip": 0.0112942,
      "auxiliary_loss_mlp": 0.00761828,
      "balance_loss_clip": 1.03906143,
      "balance_loss_mlp": 1.00030303,
      "epoch": 0.9263512294835567,
      "flos": 20339481116160.0,
      "grad_norm": 2.4114946264566774,
      "language_loss": 0.80249923,
      "learning_rate": 5.6549512696185244e-08,
      "loss": 0.82141173,
      "num_input_tokens_seen": 166563325,
      "step": 7704,
      "time_per_iteration": 3.6386144161224365
    },
    {
      "auxiliary_loss_clip": 0.01162408,
      "auxiliary_loss_mlp": 0.01023545,
      "balance_loss_clip": 1.04532897,
      "balance_loss_mlp": 1.01630008,
      "epoch": 0.9264714723741959,
      "flos": 21215378263680.0,
      "grad_norm": 1.9149634046535378,
      "language_loss": 0.68489289,
      "learning_rate": 5.636573362009156e-08,
      "loss": 0.70675248,
      "num_input_tokens_seen": 166583385,
      "step": 7705,
      "time_per_iteration": 2.6227099895477295
    },
    {
      "auxiliary_loss_clip": 0.01167996,
      "auxiliary_loss_mlp": 0.01025802,
      "balance_loss_clip": 1.04506803,
      "balance_loss_mlp": 1.01862526,
      "epoch": 0.926591715264835,
      "flos": 18004964480640.0,
      "grad_norm": 1.9307717490346854,
      "language_loss": 0.7680614,
      "learning_rate": 5.618224939177074e-08,
      "loss": 0.78999943,
      "num_input_tokens_seen": 166601290,
      "step": 7706,
      "time_per_iteration": 2.6301047801971436
    },
    {
      "auxiliary_loss_clip": 0.01128483,
      "auxiliary_loss_mlp": 0.01020984,
      "balance_loss_clip": 1.04107785,
      "balance_loss_mlp": 1.01399827,
      "epoch": 0.926711958155474,
      "flos": 36167969825280.0,
      "grad_norm": 1.7482415298251806,
      "language_loss": 0.7038185,
      "learning_rate": 5.599906003905719e-08,
      "loss": 0.72531319,
      "num_input_tokens_seen": 166623835,
      "step": 7707,
      "time_per_iteration": 2.8109073638916016
    },
    {
      "auxiliary_loss_clip": 0.01146601,
      "auxiliary_loss_mlp": 0.01025647,
      "balance_loss_clip": 1.04575801,
      "balance_loss_mlp": 1.01844645,
      "epoch": 0.9268322010461132,
      "flos": 21032736583680.0,
      "grad_norm": 2.7586734975875222,
      "language_loss": 0.81805748,
      "learning_rate": 5.581616558974023e-08,
      "loss": 0.83977997,
      "num_input_tokens_seen": 166642400,
      "step": 7708,
      "time_per_iteration": 2.644639015197754
    },
    {
      "auxiliary_loss_clip": 0.01158762,
      "auxiliary_loss_mlp": 0.00762795,
      "balance_loss_clip": 1.04607558,
      "balance_loss_mlp": 1.00039482,
      "epoch": 0.9269524439367522,
      "flos": 22964838174720.0,
      "grad_norm": 1.8495868924937249,
      "language_loss": 0.79094899,
      "learning_rate": 5.5633566071565444e-08,
      "loss": 0.81016445,
      "num_input_tokens_seen": 166661640,
      "step": 7709,
      "time_per_iteration": 2.706026077270508
    },
    {
      "auxiliary_loss_clip": 0.01100208,
      "auxiliary_loss_mlp": 0.0102268,
      "balance_loss_clip": 1.03918409,
      "balance_loss_mlp": 1.01580787,
      "epoch": 0.9270726868273913,
      "flos": 41975551468800.0,
      "grad_norm": 2.160482751388584,
      "language_loss": 0.70487028,
      "learning_rate": 5.5451261512232896e-08,
      "loss": 0.72609913,
      "num_input_tokens_seen": 166684320,
      "step": 7710,
      "time_per_iteration": 3.10477614402771
    },
    {
      "auxiliary_loss_clip": 0.01154358,
      "auxiliary_loss_mlp": 0.01021083,
      "balance_loss_clip": 1.04213214,
      "balance_loss_mlp": 1.01404417,
      "epoch": 0.9271929297180305,
      "flos": 19791771557760.0,
      "grad_norm": 1.7845967766221538,
      "language_loss": 0.62753314,
      "learning_rate": 5.5269251939397576e-08,
      "loss": 0.64928758,
      "num_input_tokens_seen": 166703835,
      "step": 7711,
      "time_per_iteration": 2.689244508743286
    },
    {
      "auxiliary_loss_clip": 0.01122783,
      "auxiliary_loss_mlp": 0.01022362,
      "balance_loss_clip": 1.03814077,
      "balance_loss_mlp": 1.01464629,
      "epoch": 0.9273131726086695,
      "flos": 19968343839360.0,
      "grad_norm": 2.1437829452280672,
      "language_loss": 0.77060151,
      "learning_rate": 5.508753738067073e-08,
      "loss": 0.79205298,
      "num_input_tokens_seen": 166723375,
      "step": 7712,
      "time_per_iteration": 2.7311367988586426
    },
    {
      "auxiliary_loss_clip": 0.01153428,
      "auxiliary_loss_mlp": 0.01021447,
      "balance_loss_clip": 1.04258609,
      "balance_loss_mlp": 1.01460707,
      "epoch": 0.9274334154993086,
      "flos": 23258587599360.0,
      "grad_norm": 2.346389908135547,
      "language_loss": 0.79110593,
      "learning_rate": 5.4906117863617875e-08,
      "loss": 0.81285465,
      "num_input_tokens_seen": 166742760,
      "step": 7713,
      "time_per_iteration": 2.702477216720581
    },
    {
      "auxiliary_loss_clip": 0.01117625,
      "auxiliary_loss_mlp": 0.01023248,
      "balance_loss_clip": 1.03842402,
      "balance_loss_mlp": 1.01635206,
      "epoch": 0.9275536583899477,
      "flos": 31795343585280.0,
      "grad_norm": 1.7867036294819574,
      "language_loss": 0.78419089,
      "learning_rate": 5.4724993415760533e-08,
      "loss": 0.80559963,
      "num_input_tokens_seen": 166761115,
      "step": 7714,
      "time_per_iteration": 2.7863175868988037
    },
    {
      "auxiliary_loss_clip": 0.01132952,
      "auxiliary_loss_mlp": 0.007624,
      "balance_loss_clip": 1.04109061,
      "balance_loss_mlp": 1.00035238,
      "epoch": 0.9276739012805868,
      "flos": 18696998885760.0,
      "grad_norm": 2.7862575117878676,
      "language_loss": 0.7454533,
      "learning_rate": 5.454416406457496e-08,
      "loss": 0.7644068,
      "num_input_tokens_seen": 166780210,
      "step": 7715,
      "time_per_iteration": 2.8035240173339844
    },
    {
      "auxiliary_loss_clip": 0.01150929,
      "auxiliary_loss_mlp": 0.0102653,
      "balance_loss_clip": 1.04287648,
      "balance_loss_mlp": 1.02032471,
      "epoch": 0.9277941441712259,
      "flos": 13879079740800.0,
      "grad_norm": 2.4838415438350925,
      "language_loss": 0.74083537,
      "learning_rate": 5.436362983749299e-08,
      "loss": 0.76260996,
      "num_input_tokens_seen": 166795380,
      "step": 7716,
      "time_per_iteration": 2.705544948577881
    },
    {
      "auxiliary_loss_clip": 0.01115742,
      "auxiliary_loss_mlp": 0.01024231,
      "balance_loss_clip": 1.0432514,
      "balance_loss_mlp": 1.01745379,
      "epoch": 0.927914387061865,
      "flos": 23258659426560.0,
      "grad_norm": 2.020664356382373,
      "language_loss": 0.64387786,
      "learning_rate": 5.418339076190137e-08,
      "loss": 0.66527754,
      "num_input_tokens_seen": 166814890,
      "step": 7717,
      "time_per_iteration": 2.7757089138031006
    },
    {
      "auxiliary_loss_clip": 0.01132373,
      "auxiliary_loss_mlp": 0.01027146,
      "balance_loss_clip": 1.0440948,
      "balance_loss_mlp": 1.01937068,
      "epoch": 0.9280346299525041,
      "flos": 18073733068800.0,
      "grad_norm": 1.989673923929145,
      "language_loss": 0.88488573,
      "learning_rate": 5.400344686514202e-08,
      "loss": 0.90648091,
      "num_input_tokens_seen": 166832475,
      "step": 7718,
      "time_per_iteration": 2.7214879989624023
    },
    {
      "auxiliary_loss_clip": 0.01149,
      "auxiliary_loss_mlp": 0.01021924,
      "balance_loss_clip": 1.04427052,
      "balance_loss_mlp": 1.01536417,
      "epoch": 0.9281548728431431,
      "flos": 22342901160960.0,
      "grad_norm": 1.8614258488940845,
      "language_loss": 0.66426712,
      "learning_rate": 5.38237981745131e-08,
      "loss": 0.68597633,
      "num_input_tokens_seen": 166850590,
      "step": 7719,
      "time_per_iteration": 2.7828149795532227
    },
    {
      "auxiliary_loss_clip": 0.01153923,
      "auxiliary_loss_mlp": 0.00762196,
      "balance_loss_clip": 1.04462957,
      "balance_loss_mlp": 1.00030065,
      "epoch": 0.9282751157337822,
      "flos": 18843765857280.0,
      "grad_norm": 1.6077372123470461,
      "language_loss": 0.81444955,
      "learning_rate": 5.364444471726592e-08,
      "loss": 0.83361077,
      "num_input_tokens_seen": 166869795,
      "step": 7720,
      "time_per_iteration": 2.7692368030548096
    },
    {
      "auxiliary_loss_clip": 0.01149455,
      "auxiliary_loss_mlp": 0.01021736,
      "balance_loss_clip": 1.04323602,
      "balance_loss_mlp": 1.01416874,
      "epoch": 0.9283953586244214,
      "flos": 25556834476800.0,
      "grad_norm": 4.966131890377145,
      "language_loss": 0.80048931,
      "learning_rate": 5.346538652060939e-08,
      "loss": 0.82220125,
      "num_input_tokens_seen": 166891150,
      "step": 7721,
      "time_per_iteration": 2.727600336074829
    },
    {
      "auxiliary_loss_clip": 0.01134397,
      "auxiliary_loss_mlp": 0.01025266,
      "balance_loss_clip": 1.04376006,
      "balance_loss_mlp": 1.01860809,
      "epoch": 0.9285156015150604,
      "flos": 18223480869120.0,
      "grad_norm": 1.8025786567549196,
      "language_loss": 0.70239776,
      "learning_rate": 5.3286623611705994e-08,
      "loss": 0.72399437,
      "num_input_tokens_seen": 166909195,
      "step": 7722,
      "time_per_iteration": 3.6132497787475586
    },
    {
      "auxiliary_loss_clip": 0.01063176,
      "auxiliary_loss_mlp": 0.0100254,
      "balance_loss_clip": 1.00952041,
      "balance_loss_mlp": 1.00153315,
      "epoch": 0.9286358444056995,
      "flos": 66400017690240.0,
      "grad_norm": 0.8200105678183449,
      "language_loss": 0.60591877,
      "learning_rate": 5.3108156017673824e-08,
      "loss": 0.62657589,
      "num_input_tokens_seen": 166970955,
      "step": 7723,
      "time_per_iteration": 3.297715187072754
    },
    {
      "auxiliary_loss_clip": 0.01143477,
      "auxiliary_loss_mlp": 0.01027459,
      "balance_loss_clip": 1.04320085,
      "balance_loss_mlp": 1.02024138,
      "epoch": 0.9287560872963386,
      "flos": 22345630594560.0,
      "grad_norm": 1.691352115176769,
      "language_loss": 0.71634185,
      "learning_rate": 5.2929983765586775e-08,
      "loss": 0.73805124,
      "num_input_tokens_seen": 166989735,
      "step": 7724,
      "time_per_iteration": 3.694568157196045
    },
    {
      "auxiliary_loss_clip": 0.01166674,
      "auxiliary_loss_mlp": 0.01027243,
      "balance_loss_clip": 1.04615474,
      "balance_loss_mlp": 1.02043641,
      "epoch": 0.9288763301869777,
      "flos": 25700225569920.0,
      "grad_norm": 1.915673040649655,
      "language_loss": 0.62453049,
      "learning_rate": 5.275210688247278e-08,
      "loss": 0.64646965,
      "num_input_tokens_seen": 167010060,
      "step": 7725,
      "time_per_iteration": 2.6725692749023438
    },
    {
      "auxiliary_loss_clip": 0.01108398,
      "auxiliary_loss_mlp": 0.01023689,
      "balance_loss_clip": 1.04114199,
      "balance_loss_mlp": 1.01704264,
      "epoch": 0.9289965730776167,
      "flos": 12312046028160.0,
      "grad_norm": 1.9191507914748371,
      "language_loss": 0.85203362,
      "learning_rate": 5.257452539531604e-08,
      "loss": 0.87335443,
      "num_input_tokens_seen": 167027130,
      "step": 7726,
      "time_per_iteration": 2.78682017326355
    },
    {
      "auxiliary_loss_clip": 0.01151033,
      "auxiliary_loss_mlp": 0.01022542,
      "balance_loss_clip": 1.04237795,
      "balance_loss_mlp": 1.01578546,
      "epoch": 0.9291168159682559,
      "flos": 26685973486080.0,
      "grad_norm": 1.5091065190472088,
      "language_loss": 0.68430847,
      "learning_rate": 5.2397239331055445e-08,
      "loss": 0.7060442,
      "num_input_tokens_seen": 167049130,
      "step": 7727,
      "time_per_iteration": 2.804305076599121
    },
    {
      "auxiliary_loss_clip": 0.0113368,
      "auxiliary_loss_mlp": 0.01025415,
      "balance_loss_clip": 1.04384935,
      "balance_loss_mlp": 1.01819086,
      "epoch": 0.929237058858895,
      "flos": 14538256179840.0,
      "grad_norm": 2.1821373105038697,
      "language_loss": 0.80717742,
      "learning_rate": 5.2220248716585036e-08,
      "loss": 0.82876837,
      "num_input_tokens_seen": 167066810,
      "step": 7728,
      "time_per_iteration": 3.5550451278686523
    },
    {
      "auxiliary_loss_clip": 0.01143415,
      "auxiliary_loss_mlp": 0.01022517,
      "balance_loss_clip": 1.04186797,
      "balance_loss_mlp": 1.01476288,
      "epoch": 0.929357301749534,
      "flos": 23835456023040.0,
      "grad_norm": 2.52075062372884,
      "language_loss": 0.75634712,
      "learning_rate": 5.204355357875445e-08,
      "loss": 0.77800643,
      "num_input_tokens_seen": 167085155,
      "step": 7729,
      "time_per_iteration": 2.7067008018493652
    },
    {
      "auxiliary_loss_clip": 0.01136003,
      "auxiliary_loss_mlp": 0.01026526,
      "balance_loss_clip": 1.04264712,
      "balance_loss_mlp": 1.01962376,
      "epoch": 0.9294775446401732,
      "flos": 12969319046400.0,
      "grad_norm": 11.53088398761752,
      "language_loss": 0.70732695,
      "learning_rate": 5.1867153944367584e-08,
      "loss": 0.72895217,
      "num_input_tokens_seen": 167101545,
      "step": 7730,
      "time_per_iteration": 3.5283844470977783
    },
    {
      "auxiliary_loss_clip": 0.01128284,
      "auxiliary_loss_mlp": 0.01029308,
      "balance_loss_clip": 1.04191828,
      "balance_loss_mlp": 1.02215815,
      "epoch": 0.9295977875308122,
      "flos": 26211809024640.0,
      "grad_norm": 1.6003775611316275,
      "language_loss": 0.73592728,
      "learning_rate": 5.16910498401848e-08,
      "loss": 0.75750321,
      "num_input_tokens_seen": 167120995,
      "step": 7731,
      "time_per_iteration": 2.791487455368042
    },
    {
      "auxiliary_loss_clip": 0.01163227,
      "auxiliary_loss_mlp": 0.01024939,
      "balance_loss_clip": 1.04568398,
      "balance_loss_mlp": 1.01838517,
      "epoch": 0.9297180304214513,
      "flos": 16472297105280.0,
      "grad_norm": 2.149901837926539,
      "language_loss": 0.83360624,
      "learning_rate": 5.151524129292073e-08,
      "loss": 0.85548782,
      "num_input_tokens_seen": 167138890,
      "step": 7732,
      "time_per_iteration": 2.5565011501312256
    },
    {
      "auxiliary_loss_clip": 0.01151755,
      "auxiliary_loss_mlp": 0.01022984,
      "balance_loss_clip": 1.04456377,
      "balance_loss_mlp": 1.01572108,
      "epoch": 0.9298382733120905,
      "flos": 24060436859520.0,
      "grad_norm": 2.1451038748985938,
      "language_loss": 0.666682,
      "learning_rate": 5.1339728329245155e-08,
      "loss": 0.68842936,
      "num_input_tokens_seen": 167159455,
      "step": 7733,
      "time_per_iteration": 2.7086613178253174
    },
    {
      "auxiliary_loss_clip": 0.01170176,
      "auxiliary_loss_mlp": 0.01030394,
      "balance_loss_clip": 1.04722846,
      "balance_loss_mlp": 1.02274621,
      "epoch": 0.9299585162027295,
      "flos": 22127652910080.0,
      "grad_norm": 2.264131385128065,
      "language_loss": 0.79724526,
      "learning_rate": 5.116451097578367e-08,
      "loss": 0.81925094,
      "num_input_tokens_seen": 167178495,
      "step": 7734,
      "time_per_iteration": 2.5959465503692627
    },
    {
      "auxiliary_loss_clip": 0.01121119,
      "auxiliary_loss_mlp": 0.01025314,
      "balance_loss_clip": 1.04005265,
      "balance_loss_mlp": 1.01862907,
      "epoch": 0.9300787590933686,
      "flos": 21471780522240.0,
      "grad_norm": 1.587551270804688,
      "language_loss": 0.74489617,
      "learning_rate": 5.0989589259115895e-08,
      "loss": 0.76636052,
      "num_input_tokens_seen": 167199380,
      "step": 7735,
      "time_per_iteration": 2.7350351810455322
    },
    {
      "auxiliary_loss_clip": 0.01146027,
      "auxiliary_loss_mlp": 0.01028902,
      "balance_loss_clip": 1.04129112,
      "balance_loss_mlp": 1.02093267,
      "epoch": 0.9301990019840077,
      "flos": 17779588594560.0,
      "grad_norm": 1.8076409330002325,
      "language_loss": 0.71664739,
      "learning_rate": 5.081496320577816e-08,
      "loss": 0.7383967,
      "num_input_tokens_seen": 167216500,
      "step": 7736,
      "time_per_iteration": 2.5866611003875732
    },
    {
      "auxiliary_loss_clip": 0.01044613,
      "auxiliary_loss_mlp": 0.01002503,
      "balance_loss_clip": 1.01491237,
      "balance_loss_mlp": 1.00132835,
      "epoch": 0.9303192448746468,
      "flos": 58896122307840.0,
      "grad_norm": 0.930919676187039,
      "language_loss": 0.61173594,
      "learning_rate": 5.0640632842260835e-08,
      "loss": 0.6322071,
      "num_input_tokens_seen": 167276760,
      "step": 7737,
      "time_per_iteration": 3.3171186447143555
    },
    {
      "auxiliary_loss_clip": 0.01121042,
      "auxiliary_loss_mlp": 0.00761695,
      "balance_loss_clip": 1.04327965,
      "balance_loss_mlp": 1.00025773,
      "epoch": 0.9304394877652858,
      "flos": 57663522172800.0,
      "grad_norm": 1.4339064532722021,
      "language_loss": 0.72538948,
      "learning_rate": 5.0466598195009426e-08,
      "loss": 0.7442168,
      "num_input_tokens_seen": 167303630,
      "step": 7738,
      "time_per_iteration": 3.1268222332000732
    },
    {
      "auxiliary_loss_clip": 0.01123745,
      "auxiliary_loss_mlp": 0.01023508,
      "balance_loss_clip": 1.04050279,
      "balance_loss_mlp": 1.01671576,
      "epoch": 0.930559730655925,
      "flos": 20996143603200.0,
      "grad_norm": 2.1735328844863253,
      "language_loss": 0.70382792,
      "learning_rate": 5.0292859290425036e-08,
      "loss": 0.72530043,
      "num_input_tokens_seen": 167321500,
      "step": 7739,
      "time_per_iteration": 2.662423610687256
    },
    {
      "auxiliary_loss_clip": 0.01162052,
      "auxiliary_loss_mlp": 0.01019924,
      "balance_loss_clip": 1.0445565,
      "balance_loss_mlp": 1.01326358,
      "epoch": 0.9306799735465641,
      "flos": 23258264376960.0,
      "grad_norm": 1.8913820498741023,
      "language_loss": 0.77877569,
      "learning_rate": 5.011941615486348e-08,
      "loss": 0.80059552,
      "num_input_tokens_seen": 167340615,
      "step": 7740,
      "time_per_iteration": 2.6315221786499023
    },
    {
      "auxiliary_loss_clip": 0.0116484,
      "auxiliary_loss_mlp": 0.010257,
      "balance_loss_clip": 1.04542494,
      "balance_loss_mlp": 1.01879144,
      "epoch": 0.9308002164372031,
      "flos": 15231547560960.0,
      "grad_norm": 2.5088427097540165,
      "language_loss": 0.84406769,
      "learning_rate": 4.994626881463659e-08,
      "loss": 0.86597306,
      "num_input_tokens_seen": 167356870,
      "step": 7741,
      "time_per_iteration": 2.559896469116211
    },
    {
      "auxiliary_loss_clip": 0.01094358,
      "auxiliary_loss_mlp": 0.01026044,
      "balance_loss_clip": 1.03704906,
      "balance_loss_mlp": 1.01918066,
      "epoch": 0.9309204593278423,
      "flos": 30847481539200.0,
      "grad_norm": 1.6620141467546423,
      "language_loss": 0.71425784,
      "learning_rate": 4.9773417296009814e-08,
      "loss": 0.73546189,
      "num_input_tokens_seen": 167378390,
      "step": 7742,
      "time_per_iteration": 2.851039171218872
    },
    {
      "auxiliary_loss_clip": 0.01157139,
      "auxiliary_loss_mlp": 0.01027167,
      "balance_loss_clip": 1.04666686,
      "balance_loss_mlp": 1.01984191,
      "epoch": 0.9310407022184813,
      "flos": 23037269950080.0,
      "grad_norm": 2.668062255613747,
      "language_loss": 0.65280366,
      "learning_rate": 4.960086162520527e-08,
      "loss": 0.67464674,
      "num_input_tokens_seen": 167398480,
      "step": 7743,
      "time_per_iteration": 2.7241408824920654
    },
    {
      "auxiliary_loss_clip": 0.01114957,
      "auxiliary_loss_mlp": 0.01026864,
      "balance_loss_clip": 1.04033422,
      "balance_loss_mlp": 1.02000976,
      "epoch": 0.9311609451091204,
      "flos": 22127976132480.0,
      "grad_norm": 1.9338800004379093,
      "language_loss": 0.8255167,
      "learning_rate": 4.942860182839936e-08,
      "loss": 0.84693491,
      "num_input_tokens_seen": 167416825,
      "step": 7744,
      "time_per_iteration": 2.709965229034424
    },
    {
      "auxiliary_loss_clip": 0.01134409,
      "auxiliary_loss_mlp": 0.0102686,
      "balance_loss_clip": 1.04204452,
      "balance_loss_mlp": 1.01940322,
      "epoch": 0.9312811879997596,
      "flos": 21099206701440.0,
      "grad_norm": 1.805952358620401,
      "language_loss": 0.79908335,
      "learning_rate": 4.925663793172341e-08,
      "loss": 0.820696,
      "num_input_tokens_seen": 167434785,
      "step": 7745,
      "time_per_iteration": 2.632051467895508
    },
    {
      "auxiliary_loss_clip": 0.01036598,
      "auxiliary_loss_mlp": 0.00753556,
      "balance_loss_clip": 1.00777602,
      "balance_loss_mlp": 1.00021827,
      "epoch": 0.9314014308903986,
      "flos": 67148179096320.0,
      "grad_norm": 0.782367179020396,
      "language_loss": 0.56498611,
      "learning_rate": 4.908496996126477e-08,
      "loss": 0.58288765,
      "num_input_tokens_seen": 167498245,
      "step": 7746,
      "time_per_iteration": 3.25850510597229
    },
    {
      "auxiliary_loss_clip": 0.01151946,
      "auxiliary_loss_mlp": 0.01022636,
      "balance_loss_clip": 1.04906464,
      "balance_loss_mlp": 1.0152154,
      "epoch": 0.9315216737810377,
      "flos": 22565583527040.0,
      "grad_norm": 1.5039530029542891,
      "language_loss": 0.76702166,
      "learning_rate": 4.89135979430646e-08,
      "loss": 0.78876746,
      "num_input_tokens_seen": 167518290,
      "step": 7747,
      "time_per_iteration": 2.680576801300049
    },
    {
      "auxiliary_loss_clip": 0.01168277,
      "auxiliary_loss_mlp": 0.01027686,
      "balance_loss_clip": 1.04809153,
      "balance_loss_mlp": 1.02016973,
      "epoch": 0.9316419166716768,
      "flos": 23984054588160.0,
      "grad_norm": 1.832858628329843,
      "language_loss": 0.85226798,
      "learning_rate": 4.874252190312078e-08,
      "loss": 0.87422764,
      "num_input_tokens_seen": 167538675,
      "step": 7748,
      "time_per_iteration": 3.63077712059021
    },
    {
      "auxiliary_loss_clip": 0.01154086,
      "auxiliary_loss_mlp": 0.01022398,
      "balance_loss_clip": 1.04257059,
      "balance_loss_mlp": 1.01514447,
      "epoch": 0.9317621595623159,
      "flos": 30230464688640.0,
      "grad_norm": 1.5610527113041768,
      "language_loss": 0.64614213,
      "learning_rate": 4.857174186738477e-08,
      "loss": 0.667907,
      "num_input_tokens_seen": 167562025,
      "step": 7749,
      "time_per_iteration": 2.799370050430298
    },
    {
      "auxiliary_loss_clip": 0.01166357,
      "auxiliary_loss_mlp": 0.01024109,
      "balance_loss_clip": 1.04739797,
      "balance_loss_mlp": 1.01688528,
      "epoch": 0.931882402452955,
      "flos": 15742735966080.0,
      "grad_norm": 2.29370746712431,
      "language_loss": 0.73356223,
      "learning_rate": 4.840125786176408e-08,
      "loss": 0.75546694,
      "num_input_tokens_seen": 167578230,
      "step": 7750,
      "time_per_iteration": 3.6446855068206787
    },
    {
      "auxiliary_loss_clip": 0.01131892,
      "auxiliary_loss_mlp": 0.01026672,
      "balance_loss_clip": 1.04163146,
      "balance_loss_mlp": 1.01954007,
      "epoch": 0.932002645343594,
      "flos": 28366521154560.0,
      "grad_norm": 1.9268505268152,
      "language_loss": 0.77516472,
      "learning_rate": 4.823106991212067e-08,
      "loss": 0.79675037,
      "num_input_tokens_seen": 167597470,
      "step": 7751,
      "time_per_iteration": 2.7492470741271973
    },
    {
      "auxiliary_loss_clip": 0.01152329,
      "auxiliary_loss_mlp": 0.01023805,
      "balance_loss_clip": 1.04280186,
      "balance_loss_mlp": 1.01701903,
      "epoch": 0.9321228882342332,
      "flos": 15341146934400.0,
      "grad_norm": 1.9385269613145577,
      "language_loss": 0.83491218,
      "learning_rate": 4.806117804427212e-08,
      "loss": 0.85667348,
      "num_input_tokens_seen": 167615405,
      "step": 7752,
      "time_per_iteration": 2.6631391048431396
    },
    {
      "auxiliary_loss_clip": 0.01145487,
      "auxiliary_loss_mlp": 0.0102223,
      "balance_loss_clip": 1.04297662,
      "balance_loss_mlp": 1.01556325,
      "epoch": 0.9322431311248722,
      "flos": 17895365107200.0,
      "grad_norm": 1.9563940448572341,
      "language_loss": 0.64078355,
      "learning_rate": 4.7891582283990926e-08,
      "loss": 0.66246068,
      "num_input_tokens_seen": 167634130,
      "step": 7753,
      "time_per_iteration": 2.6204922199249268
    },
    {
      "auxiliary_loss_clip": 0.01121665,
      "auxiliary_loss_mlp": 0.01025445,
      "balance_loss_clip": 1.04030097,
      "balance_loss_mlp": 1.01858735,
      "epoch": 0.9323633740155113,
      "flos": 24169713010560.0,
      "grad_norm": 2.9241349804487857,
      "language_loss": 0.72781229,
      "learning_rate": 4.772228265700473e-08,
      "loss": 0.74928343,
      "num_input_tokens_seen": 167654990,
      "step": 7754,
      "time_per_iteration": 3.980368137359619
    },
    {
      "auxiliary_loss_clip": 0.01155326,
      "auxiliary_loss_mlp": 0.01026542,
      "balance_loss_clip": 1.0455538,
      "balance_loss_mlp": 1.01966703,
      "epoch": 0.9324836169061504,
      "flos": 15043482927360.0,
      "grad_norm": 2.177699058901137,
      "language_loss": 0.7613647,
      "learning_rate": 4.75532791889961e-08,
      "loss": 0.7831834,
      "num_input_tokens_seen": 167671690,
      "step": 7755,
      "time_per_iteration": 2.616466760635376
    },
    {
      "auxiliary_loss_clip": 0.01149224,
      "auxiliary_loss_mlp": 0.01023701,
      "balance_loss_clip": 1.04248905,
      "balance_loss_mlp": 1.01664972,
      "epoch": 0.9326038597967895,
      "flos": 18624890332800.0,
      "grad_norm": 1.7801060315291013,
      "language_loss": 0.65611911,
      "learning_rate": 4.738457190560252e-08,
      "loss": 0.67784846,
      "num_input_tokens_seen": 167690800,
      "step": 7756,
      "time_per_iteration": 3.533395767211914
    },
    {
      "auxiliary_loss_clip": 0.01106293,
      "auxiliary_loss_mlp": 0.01023202,
      "balance_loss_clip": 1.04133403,
      "balance_loss_mlp": 1.01617432,
      "epoch": 0.9327241026874286,
      "flos": 18952646958720.0,
      "grad_norm": 2.2219135936853975,
      "language_loss": 0.78761661,
      "learning_rate": 4.721616083241664e-08,
      "loss": 0.80891156,
      "num_input_tokens_seen": 167709055,
      "step": 7757,
      "time_per_iteration": 2.7389109134674072
    },
    {
      "auxiliary_loss_clip": 0.01147946,
      "auxiliary_loss_mlp": 0.01025436,
      "balance_loss_clip": 1.04443157,
      "balance_loss_mlp": 1.01819992,
      "epoch": 0.9328443455780677,
      "flos": 29570282668800.0,
      "grad_norm": 1.6897930209431782,
      "language_loss": 0.77603871,
      "learning_rate": 4.7048045994986684e-08,
      "loss": 0.79777253,
      "num_input_tokens_seen": 167729915,
      "step": 7758,
      "time_per_iteration": 2.6746304035186768
    },
    {
      "auxiliary_loss_clip": 0.01156848,
      "auxiliary_loss_mlp": 0.01027842,
      "balance_loss_clip": 1.04576528,
      "balance_loss_mlp": 1.01964641,
      "epoch": 0.9329645884687068,
      "flos": 30081722469120.0,
      "grad_norm": 2.090903063286058,
      "language_loss": 0.91058838,
      "learning_rate": 4.688022741881559e-08,
      "loss": 0.93243527,
      "num_input_tokens_seen": 167750440,
      "step": 7759,
      "time_per_iteration": 2.738821268081665
    },
    {
      "auxiliary_loss_clip": 0.01148507,
      "auxiliary_loss_mlp": 0.01022402,
      "balance_loss_clip": 1.04475665,
      "balance_loss_mlp": 1.01589608,
      "epoch": 0.9330848313593458,
      "flos": 21867982513920.0,
      "grad_norm": 1.809752912809867,
      "language_loss": 0.74939048,
      "learning_rate": 4.671270512936076e-08,
      "loss": 0.77109957,
      "num_input_tokens_seen": 167769600,
      "step": 7760,
      "time_per_iteration": 2.6239430904388428
    },
    {
      "auxiliary_loss_clip": 0.01116458,
      "auxiliary_loss_mlp": 0.01022059,
      "balance_loss_clip": 1.04004884,
      "balance_loss_mlp": 1.01517129,
      "epoch": 0.933205074249985,
      "flos": 22127221946880.0,
      "grad_norm": 1.6750784991334133,
      "language_loss": 0.82885724,
      "learning_rate": 4.6545479152035884e-08,
      "loss": 0.85024238,
      "num_input_tokens_seen": 167788770,
      "step": 7761,
      "time_per_iteration": 2.6968612670898438
    },
    {
      "auxiliary_loss_clip": 0.01152769,
      "auxiliary_loss_mlp": 0.01024741,
      "balance_loss_clip": 1.04517341,
      "balance_loss_mlp": 1.01770437,
      "epoch": 0.9333253171406241,
      "flos": 15341254675200.0,
      "grad_norm": 1.9334692979975494,
      "language_loss": 0.76373827,
      "learning_rate": 4.637854951220821e-08,
      "loss": 0.78551334,
      "num_input_tokens_seen": 167805555,
      "step": 7762,
      "time_per_iteration": 2.561016082763672
    },
    {
      "auxiliary_loss_clip": 0.01115818,
      "auxiliary_loss_mlp": 0.01028107,
      "balance_loss_clip": 1.03917027,
      "balance_loss_mlp": 1.02116895,
      "epoch": 0.9334455600312631,
      "flos": 15706142985600.0,
      "grad_norm": 5.452903051654911,
      "language_loss": 0.74843049,
      "learning_rate": 4.621191623520171e-08,
      "loss": 0.76986969,
      "num_input_tokens_seen": 167823985,
      "step": 7763,
      "time_per_iteration": 2.687645673751831
    },
    {
      "auxiliary_loss_clip": 0.01111234,
      "auxiliary_loss_mlp": 0.01021032,
      "balance_loss_clip": 1.04166794,
      "balance_loss_mlp": 1.01320601,
      "epoch": 0.9335658029219023,
      "flos": 22163563532160.0,
      "grad_norm": 2.8128340152730726,
      "language_loss": 0.84530079,
      "learning_rate": 4.604557934629372e-08,
      "loss": 0.8666234,
      "num_input_tokens_seen": 167843060,
      "step": 7764,
      "time_per_iteration": 2.7539713382720947
    },
    {
      "auxiliary_loss_clip": 0.01132466,
      "auxiliary_loss_mlp": 0.01022366,
      "balance_loss_clip": 1.04297495,
      "balance_loss_mlp": 1.01496315,
      "epoch": 0.9336860458125413,
      "flos": 20266833859200.0,
      "grad_norm": 1.7419667906815142,
      "language_loss": 0.80155754,
      "learning_rate": 4.587953887071805e-08,
      "loss": 0.82310593,
      "num_input_tokens_seen": 167862880,
      "step": 7765,
      "time_per_iteration": 2.6667873859405518
    },
    {
      "auxiliary_loss_clip": 0.0113126,
      "auxiliary_loss_mlp": 0.01026706,
      "balance_loss_clip": 1.03864336,
      "balance_loss_mlp": 1.01961899,
      "epoch": 0.9338062887031804,
      "flos": 20919689504640.0,
      "grad_norm": 1.8288559426579287,
      "language_loss": 0.85846633,
      "learning_rate": 4.5713794833662554e-08,
      "loss": 0.88004601,
      "num_input_tokens_seen": 167882095,
      "step": 7766,
      "time_per_iteration": 2.6537554264068604
    },
    {
      "auxiliary_loss_clip": 0.0116826,
      "auxiliary_loss_mlp": 0.01030062,
      "balance_loss_clip": 1.04691732,
      "balance_loss_mlp": 1.02254009,
      "epoch": 0.9339265315938196,
      "flos": 23221635482880.0,
      "grad_norm": 2.2121211469262976,
      "language_loss": 0.63188708,
      "learning_rate": 4.5548347260270236e-08,
      "loss": 0.65387028,
      "num_input_tokens_seen": 167901385,
      "step": 7767,
      "time_per_iteration": 2.6033313274383545
    },
    {
      "auxiliary_loss_clip": 0.01118723,
      "auxiliary_loss_mlp": 0.01022785,
      "balance_loss_clip": 1.04015851,
      "balance_loss_mlp": 1.01591289,
      "epoch": 0.9340467744844586,
      "flos": 22820261932800.0,
      "grad_norm": 7.4775368046058865,
      "language_loss": 0.69379044,
      "learning_rate": 4.538319617564012e-08,
      "loss": 0.71520555,
      "num_input_tokens_seen": 167920405,
      "step": 7768,
      "time_per_iteration": 2.7278037071228027
    },
    {
      "auxiliary_loss_clip": 0.01135328,
      "auxiliary_loss_mlp": 0.01028303,
      "balance_loss_clip": 1.04071164,
      "balance_loss_mlp": 1.02065587,
      "epoch": 0.9341670173750977,
      "flos": 23660428026240.0,
      "grad_norm": 1.9981844610082031,
      "language_loss": 0.7448408,
      "learning_rate": 4.521834160482485e-08,
      "loss": 0.76647711,
      "num_input_tokens_seen": 167939145,
      "step": 7769,
      "time_per_iteration": 2.6390607357025146
    },
    {
      "auxiliary_loss_clip": 0.01155716,
      "auxiliary_loss_mlp": 0.01028053,
      "balance_loss_clip": 1.04551351,
      "balance_loss_mlp": 1.02087629,
      "epoch": 0.9342872602657368,
      "flos": 24824256595200.0,
      "grad_norm": 1.533330388009997,
      "language_loss": 0.82041311,
      "learning_rate": 4.5053783572832846e-08,
      "loss": 0.84225082,
      "num_input_tokens_seen": 167959325,
      "step": 7770,
      "time_per_iteration": 2.7753350734710693
    },
    {
      "auxiliary_loss_clip": 0.01152004,
      "auxiliary_loss_mlp": 0.01022096,
      "balance_loss_clip": 1.04470313,
      "balance_loss_mlp": 1.01491332,
      "epoch": 0.9344075031563759,
      "flos": 25771831332480.0,
      "grad_norm": 2.500735363065873,
      "language_loss": 0.76397711,
      "learning_rate": 4.488952210462771e-08,
      "loss": 0.78571808,
      "num_input_tokens_seen": 167979530,
      "step": 7771,
      "time_per_iteration": 2.6625311374664307
    },
    {
      "auxiliary_loss_clip": 0.0116486,
      "auxiliary_loss_mlp": 0.01022368,
      "balance_loss_clip": 1.04604387,
      "balance_loss_mlp": 1.01539159,
      "epoch": 0.9345277460470149,
      "flos": 25551303782400.0,
      "grad_norm": 1.8983991586277111,
      "language_loss": 0.85561496,
      "learning_rate": 4.4725557225127495e-08,
      "loss": 0.87748724,
      "num_input_tokens_seen": 167997870,
      "step": 7772,
      "time_per_iteration": 2.695010185241699
    },
    {
      "auxiliary_loss_clip": 0.01151221,
      "auxiliary_loss_mlp": 0.01023607,
      "balance_loss_clip": 1.04290676,
      "balance_loss_mlp": 1.01694608,
      "epoch": 0.9346479889376541,
      "flos": 34313112432000.0,
      "grad_norm": 1.4965285580524506,
      "language_loss": 0.79083616,
      "learning_rate": 4.456188895920565e-08,
      "loss": 0.8125844,
      "num_input_tokens_seen": 168019625,
      "step": 7773,
      "time_per_iteration": 2.7694411277770996
    },
    {
      "auxiliary_loss_clip": 0.01166064,
      "auxiliary_loss_mlp": 0.01023267,
      "balance_loss_clip": 1.04618096,
      "balance_loss_mlp": 1.01612091,
      "epoch": 0.9347682318282932,
      "flos": 19093739581440.0,
      "grad_norm": 2.8396476457440483,
      "language_loss": 0.85374111,
      "learning_rate": 4.439851733169031e-08,
      "loss": 0.87563443,
      "num_input_tokens_seen": 168037415,
      "step": 7774,
      "time_per_iteration": 3.5083396434783936
    },
    {
      "auxiliary_loss_clip": 0.01121783,
      "auxiliary_loss_mlp": 0.01024472,
      "balance_loss_clip": 1.04093623,
      "balance_loss_mlp": 1.01775742,
      "epoch": 0.9348884747189322,
      "flos": 26249587153920.0,
      "grad_norm": 2.364509390695934,
      "language_loss": 0.69065124,
      "learning_rate": 4.4235442367365204e-08,
      "loss": 0.71211374,
      "num_input_tokens_seen": 168057725,
      "step": 7775,
      "time_per_iteration": 2.759178400039673
    },
    {
      "auxiliary_loss_clip": 0.01132988,
      "auxiliary_loss_mlp": 0.01024454,
      "balance_loss_clip": 1.03937697,
      "balance_loss_mlp": 1.01741219,
      "epoch": 0.9350087176095714,
      "flos": 18333080242560.0,
      "grad_norm": 2.2053460004327596,
      "language_loss": 0.79253012,
      "learning_rate": 4.4072664090968545e-08,
      "loss": 0.81410456,
      "num_input_tokens_seen": 168076110,
      "step": 7776,
      "time_per_iteration": 2.6150264739990234
    },
    {
      "auxiliary_loss_clip": 0.01135152,
      "auxiliary_loss_mlp": 0.01024349,
      "balance_loss_clip": 1.0391717,
      "balance_loss_mlp": 1.01712787,
      "epoch": 0.9351289605002104,
      "flos": 19318253541120.0,
      "grad_norm": 1.8309431712071516,
      "language_loss": 0.85097766,
      "learning_rate": 4.391018252719347e-08,
      "loss": 0.8725726,
      "num_input_tokens_seen": 168095905,
      "step": 7777,
      "time_per_iteration": 3.6331489086151123
    },
    {
      "auxiliary_loss_clip": 0.01138516,
      "auxiliary_loss_mlp": 0.01027324,
      "balance_loss_clip": 1.04096234,
      "balance_loss_mlp": 1.02017117,
      "epoch": 0.9352492033908495,
      "flos": 18799990156800.0,
      "grad_norm": 1.7702820366395635,
      "language_loss": 0.69219184,
      "learning_rate": 4.374799770068849e-08,
      "loss": 0.71385032,
      "num_input_tokens_seen": 168112580,
      "step": 7778,
      "time_per_iteration": 2.6266047954559326
    },
    {
      "auxiliary_loss_clip": 0.01149859,
      "auxiliary_loss_mlp": 0.0102741,
      "balance_loss_clip": 1.04532099,
      "balance_loss_mlp": 1.02028692,
      "epoch": 0.9353694462814887,
      "flos": 29530134241920.0,
      "grad_norm": 1.9318164640826072,
      "language_loss": 0.74492389,
      "learning_rate": 4.358610963605658e-08,
      "loss": 0.76669657,
      "num_input_tokens_seen": 168133030,
      "step": 7779,
      "time_per_iteration": 2.7109174728393555
    },
    {
      "auxiliary_loss_clip": 0.01169552,
      "auxiliary_loss_mlp": 0.0103064,
      "balance_loss_clip": 1.04768348,
      "balance_loss_mlp": 1.02355313,
      "epoch": 0.9354896891721277,
      "flos": 30665450390400.0,
      "grad_norm": 2.714088119774056,
      "language_loss": 0.68317294,
      "learning_rate": 4.342451835785677e-08,
      "loss": 0.7051748,
      "num_input_tokens_seen": 168153940,
      "step": 7780,
      "time_per_iteration": 3.647111177444458
    },
    {
      "auxiliary_loss_clip": 0.01133293,
      "auxiliary_loss_mlp": 0.01025748,
      "balance_loss_clip": 1.04078865,
      "balance_loss_mlp": 1.0191735,
      "epoch": 0.9356099320627668,
      "flos": 19463907191040.0,
      "grad_norm": 1.6910760774660425,
      "language_loss": 0.75248623,
      "learning_rate": 4.3263223890601665e-08,
      "loss": 0.77407658,
      "num_input_tokens_seen": 168172650,
      "step": 7781,
      "time_per_iteration": 2.7274491786956787
    },
    {
      "auxiliary_loss_clip": 0.01148989,
      "auxiliary_loss_mlp": 0.00761866,
      "balance_loss_clip": 1.04651976,
      "balance_loss_mlp": 1.00029325,
      "epoch": 0.9357301749534058,
      "flos": 19098156954240.0,
      "grad_norm": 1.741962188276256,
      "language_loss": 0.79114497,
      "learning_rate": 4.31022262587597e-08,
      "loss": 0.8102535,
      "num_input_tokens_seen": 168191325,
      "step": 7782,
      "time_per_iteration": 3.582162857055664
    },
    {
      "auxiliary_loss_clip": 0.01152345,
      "auxiliary_loss_mlp": 0.01030366,
      "balance_loss_clip": 1.04665971,
      "balance_loss_mlp": 1.02256417,
      "epoch": 0.935850417844045,
      "flos": 23550361776000.0,
      "grad_norm": 1.633882386229903,
      "language_loss": 0.65966439,
      "learning_rate": 4.2941525486754225e-08,
      "loss": 0.68149155,
      "num_input_tokens_seen": 168211645,
      "step": 7783,
      "time_per_iteration": 2.608125925064087
    },
    {
      "auxiliary_loss_clip": 0.01119233,
      "auxiliary_loss_mlp": 0.01025027,
      "balance_loss_clip": 1.04155445,
      "balance_loss_mlp": 1.01852417,
      "epoch": 0.935970660734684,
      "flos": 18588333265920.0,
      "grad_norm": 1.9019528100488723,
      "language_loss": 0.79561734,
      "learning_rate": 4.278112159896286e-08,
      "loss": 0.81705999,
      "num_input_tokens_seen": 168229485,
      "step": 7784,
      "time_per_iteration": 2.7679340839385986
    },
    {
      "auxiliary_loss_clip": 0.01129382,
      "auxiliary_loss_mlp": 0.01026507,
      "balance_loss_clip": 1.03858399,
      "balance_loss_mlp": 1.01934838,
      "epoch": 0.9360909036253231,
      "flos": 20631255292800.0,
      "grad_norm": 1.7431953111983045,
      "language_loss": 0.67480201,
      "learning_rate": 4.2621014619719896e-08,
      "loss": 0.69636083,
      "num_input_tokens_seen": 168247250,
      "step": 7785,
      "time_per_iteration": 2.64290714263916
    },
    {
      "auxiliary_loss_clip": 0.01040279,
      "auxiliary_loss_mlp": 0.01000824,
      "balance_loss_clip": 1.00775337,
      "balance_loss_mlp": 0.99976295,
      "epoch": 0.9362111465159623,
      "flos": 61791421052160.0,
      "grad_norm": 0.7195337756704083,
      "language_loss": 0.58561528,
      "learning_rate": 4.246120457331215e-08,
      "loss": 0.60602629,
      "num_input_tokens_seen": 168309425,
      "step": 7786,
      "time_per_iteration": 3.2914459705352783
    },
    {
      "auxiliary_loss_clip": 0.01130456,
      "auxiliary_loss_mlp": 0.01028864,
      "balance_loss_clip": 1.04350138,
      "balance_loss_mlp": 1.02170515,
      "epoch": 0.9363313894066013,
      "flos": 24170395368960.0,
      "grad_norm": 1.8530347124504223,
      "language_loss": 0.71831381,
      "learning_rate": 4.2301691483983325e-08,
      "loss": 0.73990703,
      "num_input_tokens_seen": 168329545,
      "step": 7787,
      "time_per_iteration": 2.690408706665039
    },
    {
      "auxiliary_loss_clip": 0.01154029,
      "auxiliary_loss_mlp": 0.01022588,
      "balance_loss_clip": 1.04498672,
      "balance_loss_mlp": 1.01565552,
      "epoch": 0.9364516322972404,
      "flos": 20120354196480.0,
      "grad_norm": 2.4423990863173,
      "language_loss": 0.7590729,
      "learning_rate": 4.214247537593163e-08,
      "loss": 0.78083909,
      "num_input_tokens_seen": 168348795,
      "step": 7788,
      "time_per_iteration": 2.5867996215820312
    },
    {
      "auxiliary_loss_clip": 0.01137444,
      "auxiliary_loss_mlp": 0.01027126,
      "balance_loss_clip": 1.04124641,
      "balance_loss_mlp": 1.0192765,
      "epoch": 0.9365718751878795,
      "flos": 20703758895360.0,
      "grad_norm": 2.0202133381820166,
      "language_loss": 0.80476916,
      "learning_rate": 4.1983556273309293e-08,
      "loss": 0.82641482,
      "num_input_tokens_seen": 168367545,
      "step": 7789,
      "time_per_iteration": 2.690735340118408
    },
    {
      "auxiliary_loss_clip": 0.01166917,
      "auxiliary_loss_mlp": 0.01024035,
      "balance_loss_clip": 1.04546022,
      "balance_loss_mlp": 1.01683474,
      "epoch": 0.9366921180785186,
      "flos": 18655270260480.0,
      "grad_norm": 4.870883529232793,
      "language_loss": 0.69127661,
      "learning_rate": 4.182493420022526e-08,
      "loss": 0.71318614,
      "num_input_tokens_seen": 168383215,
      "step": 7790,
      "time_per_iteration": 2.5158374309539795
    },
    {
      "auxiliary_loss_clip": 0.01123037,
      "auxiliary_loss_mlp": 0.01020668,
      "balance_loss_clip": 1.04070163,
      "balance_loss_mlp": 1.01406384,
      "epoch": 0.9368123609691577,
      "flos": 25774955815680.0,
      "grad_norm": 1.6624163416382007,
      "language_loss": 0.78532231,
      "learning_rate": 4.166660918074139e-08,
      "loss": 0.80675936,
      "num_input_tokens_seen": 168403120,
      "step": 7791,
      "time_per_iteration": 2.736804723739624
    },
    {
      "auxiliary_loss_clip": 0.01117817,
      "auxiliary_loss_mlp": 0.01023826,
      "balance_loss_clip": 1.03934169,
      "balance_loss_mlp": 1.01693583,
      "epoch": 0.9369326038597968,
      "flos": 25553386771200.0,
      "grad_norm": 1.5186264075625067,
      "language_loss": 0.73554409,
      "learning_rate": 4.15085812388758e-08,
      "loss": 0.75696045,
      "num_input_tokens_seen": 168425340,
      "step": 7792,
      "time_per_iteration": 2.78934645652771
    },
    {
      "auxiliary_loss_clip": 0.01136756,
      "auxiliary_loss_mlp": 0.0102366,
      "balance_loss_clip": 1.04402173,
      "balance_loss_mlp": 1.01626945,
      "epoch": 0.9370528467504359,
      "flos": 23220019370880.0,
      "grad_norm": 1.7701714076730637,
      "language_loss": 0.78361326,
      "learning_rate": 4.135085039860153e-08,
      "loss": 0.80521739,
      "num_input_tokens_seen": 168444740,
      "step": 7793,
      "time_per_iteration": 2.685053586959839
    },
    {
      "auxiliary_loss_clip": 0.01134728,
      "auxiliary_loss_mlp": 0.01023936,
      "balance_loss_clip": 1.04399765,
      "balance_loss_mlp": 1.01699209,
      "epoch": 0.9371730896410749,
      "flos": 24967468120320.0,
      "grad_norm": 20.48790429867453,
      "language_loss": 0.78621477,
      "learning_rate": 4.1193416683845906e-08,
      "loss": 0.80780143,
      "num_input_tokens_seen": 168463670,
      "step": 7794,
      "time_per_iteration": 2.615208864212036
    },
    {
      "auxiliary_loss_clip": 0.01125767,
      "auxiliary_loss_mlp": 0.01025731,
      "balance_loss_clip": 1.04191279,
      "balance_loss_mlp": 1.01907027,
      "epoch": 0.9372933325317141,
      "flos": 15553091134080.0,
      "grad_norm": 2.446661428820972,
      "language_loss": 0.83747149,
      "learning_rate": 4.103628011849136e-08,
      "loss": 0.85898644,
      "num_input_tokens_seen": 168479030,
      "step": 7795,
      "time_per_iteration": 2.6791586875915527
    },
    {
      "auxiliary_loss_clip": 0.01139967,
      "auxiliary_loss_mlp": 0.01025379,
      "balance_loss_clip": 1.04453039,
      "balance_loss_mlp": 1.01789284,
      "epoch": 0.9374135754223532,
      "flos": 21871861182720.0,
      "grad_norm": 1.771053699392945,
      "language_loss": 0.75902802,
      "learning_rate": 4.0879440726375506e-08,
      "loss": 0.78068143,
      "num_input_tokens_seen": 168496815,
      "step": 7796,
      "time_per_iteration": 2.6533925533294678
    },
    {
      "auxiliary_loss_clip": 0.01133583,
      "auxiliary_loss_mlp": 0.01026272,
      "balance_loss_clip": 1.03917289,
      "balance_loss_mlp": 1.01938152,
      "epoch": 0.9375338183129922,
      "flos": 22631048064000.0,
      "grad_norm": 2.6667170106470235,
      "language_loss": 0.5634867,
      "learning_rate": 4.0722898531291074e-08,
      "loss": 0.58508527,
      "num_input_tokens_seen": 168514055,
      "step": 7797,
      "time_per_iteration": 2.6774649620056152
    },
    {
      "auxiliary_loss_clip": 0.01143065,
      "auxiliary_loss_mlp": 0.01026556,
      "balance_loss_clip": 1.04404187,
      "balance_loss_mlp": 1.01937985,
      "epoch": 0.9376540612036314,
      "flos": 26104292640000.0,
      "grad_norm": 1.9259575952504808,
      "language_loss": 0.76942801,
      "learning_rate": 4.0566653556985295e-08,
      "loss": 0.79112422,
      "num_input_tokens_seen": 168534600,
      "step": 7798,
      "time_per_iteration": 2.7116200923919678
    },
    {
      "auxiliary_loss_clip": 0.01084322,
      "auxiliary_loss_mlp": 0.01024093,
      "balance_loss_clip": 1.03758717,
      "balance_loss_mlp": 1.0161891,
      "epoch": 0.9377743040942704,
      "flos": 19717580016000.0,
      "grad_norm": 2.561184629203548,
      "language_loss": 0.81600177,
      "learning_rate": 4.0410705827159886e-08,
      "loss": 0.83708596,
      "num_input_tokens_seen": 168551895,
      "step": 7799,
      "time_per_iteration": 2.8884031772613525
    },
    {
      "auxiliary_loss_clip": 0.01131485,
      "auxiliary_loss_mlp": 0.01023422,
      "balance_loss_clip": 1.03810787,
      "balance_loss_mlp": 1.01643658,
      "epoch": 0.9378945469849095,
      "flos": 15267530010240.0,
      "grad_norm": 2.2688016485441915,
      "language_loss": 0.70970428,
      "learning_rate": 4.0255055365472356e-08,
      "loss": 0.73125327,
      "num_input_tokens_seen": 168569990,
      "step": 7800,
      "time_per_iteration": 3.8120007514953613
    },
    {
      "auxiliary_loss_clip": 0.01090724,
      "auxiliary_loss_mlp": 0.0102946,
      "balance_loss_clip": 1.03390908,
      "balance_loss_mlp": 1.0221343,
      "epoch": 0.9380147898755486,
      "flos": 20591394174720.0,
      "grad_norm": 2.271287666719792,
      "language_loss": 0.74925536,
      "learning_rate": 4.009970219553471e-08,
      "loss": 0.77045721,
      "num_input_tokens_seen": 168586940,
      "step": 7801,
      "time_per_iteration": 2.7986373901367188
    },
    {
      "auxiliary_loss_clip": 0.01156314,
      "auxiliary_loss_mlp": 0.01025245,
      "balance_loss_clip": 1.0458318,
      "balance_loss_mlp": 1.01774073,
      "epoch": 0.9381350327661877,
      "flos": 26281116316800.0,
      "grad_norm": 2.5765481757130564,
      "language_loss": 0.76237684,
      "learning_rate": 3.99446463409141e-08,
      "loss": 0.78419244,
      "num_input_tokens_seen": 168604795,
      "step": 7802,
      "time_per_iteration": 2.742312431335449
    },
    {
      "auxiliary_loss_clip": 0.01154061,
      "auxiliary_loss_mlp": 0.01024369,
      "balance_loss_clip": 1.04154646,
      "balance_loss_mlp": 1.01740384,
      "epoch": 0.9382552756568268,
      "flos": 23586344225280.0,
      "grad_norm": 3.142901626774641,
      "language_loss": 0.68963683,
      "learning_rate": 3.978988782513215e-08,
      "loss": 0.71142113,
      "num_input_tokens_seen": 168622290,
      "step": 7803,
      "time_per_iteration": 3.6902055740356445
    },
    {
      "auxiliary_loss_clip": 0.01157093,
      "auxiliary_loss_mlp": 0.01028194,
      "balance_loss_clip": 1.04598856,
      "balance_loss_mlp": 1.02119088,
      "epoch": 0.9383755185474659,
      "flos": 28438809275520.0,
      "grad_norm": 1.7402425655515612,
      "language_loss": 0.76083016,
      "learning_rate": 3.963542667166586e-08,
      "loss": 0.78268301,
      "num_input_tokens_seen": 168642395,
      "step": 7804,
      "time_per_iteration": 2.6923837661743164
    },
    {
      "auxiliary_loss_clip": 0.01124488,
      "auxiliary_loss_mlp": 0.01028152,
      "balance_loss_clip": 1.0446806,
      "balance_loss_mlp": 1.02119923,
      "epoch": 0.938495761438105,
      "flos": 20449583280000.0,
      "grad_norm": 1.967341395502087,
      "language_loss": 0.68190867,
      "learning_rate": 3.9481262903946486e-08,
      "loss": 0.70343506,
      "num_input_tokens_seen": 168661840,
      "step": 7805,
      "time_per_iteration": 3.65708589553833
    },
    {
      "auxiliary_loss_clip": 0.01026833,
      "auxiliary_loss_mlp": 0.01003319,
      "balance_loss_clip": 1.00980484,
      "balance_loss_mlp": 1.00233006,
      "epoch": 0.938616004328744,
      "flos": 69302711658240.0,
      "grad_norm": 0.7714951263161394,
      "language_loss": 0.54477876,
      "learning_rate": 3.932739654536066e-08,
      "loss": 0.56508029,
      "num_input_tokens_seen": 168724540,
      "step": 7806,
      "time_per_iteration": 3.383449077606201
    },
    {
      "auxiliary_loss_clip": 0.01151863,
      "auxiliary_loss_mlp": 0.01023167,
      "balance_loss_clip": 1.04569507,
      "balance_loss_mlp": 1.0166018,
      "epoch": 0.9387362472193832,
      "flos": 18911636605440.0,
      "grad_norm": 2.400259369260182,
      "language_loss": 0.74690765,
      "learning_rate": 3.917382761925014e-08,
      "loss": 0.76865798,
      "num_input_tokens_seen": 168740375,
      "step": 7807,
      "time_per_iteration": 2.860156774520874
    },
    {
      "auxiliary_loss_clip": 0.01148057,
      "auxiliary_loss_mlp": 0.01027253,
      "balance_loss_clip": 1.0432663,
      "balance_loss_mlp": 1.02004981,
      "epoch": 0.9388564901100223,
      "flos": 26501967089280.0,
      "grad_norm": 1.8035089473895256,
      "language_loss": 0.79574811,
      "learning_rate": 3.9020556148910754e-08,
      "loss": 0.81750119,
      "num_input_tokens_seen": 168759730,
      "step": 7808,
      "time_per_iteration": 3.625833511352539
    },
    {
      "auxiliary_loss_clip": 0.01045755,
      "auxiliary_loss_mlp": 0.01000829,
      "balance_loss_clip": 1.00971889,
      "balance_loss_mlp": 0.99978602,
      "epoch": 0.9389767330006613,
      "flos": 58941083157120.0,
      "grad_norm": 0.7108238846986127,
      "language_loss": 0.56660968,
      "learning_rate": 3.8867582157593895e-08,
      "loss": 0.58707553,
      "num_input_tokens_seen": 168813935,
      "step": 7809,
      "time_per_iteration": 3.0906152725219727
    },
    {
      "auxiliary_loss_clip": 0.01149629,
      "auxiliary_loss_mlp": 0.01020577,
      "balance_loss_clip": 1.04541779,
      "balance_loss_mlp": 1.01384795,
      "epoch": 0.9390969758913005,
      "flos": 31102554994560.0,
      "grad_norm": 1.7404458595231282,
      "language_loss": 0.76556194,
      "learning_rate": 3.871490566850544e-08,
      "loss": 0.78726399,
      "num_input_tokens_seen": 168838145,
      "step": 7810,
      "time_per_iteration": 2.656121253967285
    },
    {
      "auxiliary_loss_clip": 0.01130824,
      "auxiliary_loss_mlp": 0.01023461,
      "balance_loss_clip": 1.04127574,
      "balance_loss_mlp": 1.01631474,
      "epoch": 0.9392172187819395,
      "flos": 22419391173120.0,
      "grad_norm": 2.442010802219366,
      "language_loss": 0.70644844,
      "learning_rate": 3.856252670480642e-08,
      "loss": 0.72799128,
      "num_input_tokens_seen": 168856805,
      "step": 7811,
      "time_per_iteration": 2.6810100078582764
    },
    {
      "auxiliary_loss_clip": 0.01133396,
      "auxiliary_loss_mlp": 0.0102619,
      "balance_loss_clip": 1.03973544,
      "balance_loss_mlp": 1.01943326,
      "epoch": 0.9393374616725786,
      "flos": 19719483436800.0,
      "grad_norm": 1.8189852017729227,
      "language_loss": 0.8147217,
      "learning_rate": 3.841044528961279e-08,
      "loss": 0.83631754,
      "num_input_tokens_seen": 168874600,
      "step": 7812,
      "time_per_iteration": 2.6546671390533447
    },
    {
      "auxiliary_loss_clip": 0.01165001,
      "auxiliary_loss_mlp": 0.01021159,
      "balance_loss_clip": 1.0434916,
      "balance_loss_mlp": 1.01389611,
      "epoch": 0.9394577045632178,
      "flos": 24170215800960.0,
      "grad_norm": 1.7828374881191116,
      "language_loss": 0.7874496,
      "learning_rate": 3.825866144599477e-08,
      "loss": 0.80931127,
      "num_input_tokens_seen": 168893655,
      "step": 7813,
      "time_per_iteration": 2.726494789123535
    },
    {
      "auxiliary_loss_clip": 0.0113462,
      "auxiliary_loss_mlp": 0.01024893,
      "balance_loss_clip": 1.04017222,
      "balance_loss_mlp": 1.01784742,
      "epoch": 0.9395779474538568,
      "flos": 19023929498880.0,
      "grad_norm": 2.0483207388230444,
      "language_loss": 0.75470519,
      "learning_rate": 3.8107175196978145e-08,
      "loss": 0.77630031,
      "num_input_tokens_seen": 168909960,
      "step": 7814,
      "time_per_iteration": 2.709256410598755
    },
    {
      "auxiliary_loss_clip": 0.01120402,
      "auxiliary_loss_mlp": 0.01022511,
      "balance_loss_clip": 1.04256237,
      "balance_loss_mlp": 1.01590109,
      "epoch": 0.9396981903444959,
      "flos": 14319129260160.0,
      "grad_norm": 1.869081885256342,
      "language_loss": 0.76781714,
      "learning_rate": 3.7955986565542996e-08,
      "loss": 0.78924626,
      "num_input_tokens_seen": 168928040,
      "step": 7815,
      "time_per_iteration": 2.6663522720336914
    },
    {
      "auxiliary_loss_clip": 0.01120164,
      "auxiliary_loss_mlp": 0.01023402,
      "balance_loss_clip": 1.03835082,
      "balance_loss_mlp": 1.01639283,
      "epoch": 0.9398184332351349,
      "flos": 34787564202240.0,
      "grad_norm": 2.077646123830158,
      "language_loss": 0.68294501,
      "learning_rate": 3.780509557462497e-08,
      "loss": 0.70438075,
      "num_input_tokens_seen": 168948240,
      "step": 7816,
      "time_per_iteration": 2.771965503692627
    },
    {
      "auxiliary_loss_clip": 0.01135611,
      "auxiliary_loss_mlp": 0.01022982,
      "balance_loss_clip": 1.04116869,
      "balance_loss_mlp": 1.01520324,
      "epoch": 0.9399386761257741,
      "flos": 25372253462400.0,
      "grad_norm": 1.612556315156751,
      "language_loss": 0.75450557,
      "learning_rate": 3.765450224711375e-08,
      "loss": 0.77609152,
      "num_input_tokens_seen": 168968745,
      "step": 7817,
      "time_per_iteration": 2.6917316913604736
    },
    {
      "auxiliary_loss_clip": 0.01131812,
      "auxiliary_loss_mlp": 0.01022213,
      "balance_loss_clip": 1.04269266,
      "balance_loss_mlp": 1.01534677,
      "epoch": 0.9400589190164131,
      "flos": 27304965584640.0,
      "grad_norm": 1.8364389864468247,
      "language_loss": 0.80174243,
      "learning_rate": 3.750420660585396e-08,
      "loss": 0.82328266,
      "num_input_tokens_seen": 168990685,
      "step": 7818,
      "time_per_iteration": 2.724085569381714
    },
    {
      "auxiliary_loss_clip": 0.01163054,
      "auxiliary_loss_mlp": 0.010272,
      "balance_loss_clip": 1.0448823,
      "balance_loss_mlp": 1.02010739,
      "epoch": 0.9401791619070522,
      "flos": 23399859790080.0,
      "grad_norm": 1.8569306649151234,
      "language_loss": 0.79631847,
      "learning_rate": 3.735420867364603e-08,
      "loss": 0.81822103,
      "num_input_tokens_seen": 169011665,
      "step": 7819,
      "time_per_iteration": 2.619718551635742
    },
    {
      "auxiliary_loss_clip": 0.01087713,
      "auxiliary_loss_mlp": 0.0102155,
      "balance_loss_clip": 1.03609824,
      "balance_loss_mlp": 1.01499057,
      "epoch": 0.9402994047976914,
      "flos": 35881403120640.0,
      "grad_norm": 1.7316062657255817,
      "language_loss": 0.61794841,
      "learning_rate": 3.7204508473244186e-08,
      "loss": 0.63904107,
      "num_input_tokens_seen": 169035290,
      "step": 7820,
      "time_per_iteration": 2.9421424865722656
    },
    {
      "auxiliary_loss_clip": 0.01078699,
      "auxiliary_loss_mlp": 0.01018704,
      "balance_loss_clip": 1.03927076,
      "balance_loss_mlp": 1.01212907,
      "epoch": 0.9404196476883304,
      "flos": 22236821320320.0,
      "grad_norm": 1.8232975286026603,
      "language_loss": 0.69584012,
      "learning_rate": 3.7055106027357395e-08,
      "loss": 0.7168141,
      "num_input_tokens_seen": 169055155,
      "step": 7821,
      "time_per_iteration": 2.7583436965942383
    },
    {
      "auxiliary_loss_clip": 0.0114537,
      "auxiliary_loss_mlp": 0.01024259,
      "balance_loss_clip": 1.0417819,
      "balance_loss_mlp": 1.01749992,
      "epoch": 0.9405398905789695,
      "flos": 18915802583040.0,
      "grad_norm": 2.3155985717272904,
      "language_loss": 0.71862769,
      "learning_rate": 3.690600135865063e-08,
      "loss": 0.74032402,
      "num_input_tokens_seen": 169072080,
      "step": 7822,
      "time_per_iteration": 2.631145715713501
    },
    {
      "auxiliary_loss_clip": 0.01022954,
      "auxiliary_loss_mlp": 0.01001952,
      "balance_loss_clip": 1.00952864,
      "balance_loss_mlp": 1.00095069,
      "epoch": 0.9406601334696086,
      "flos": 70274130048000.0,
      "grad_norm": 0.8215150943984585,
      "language_loss": 0.58096558,
      "learning_rate": 3.675719448974246e-08,
      "loss": 0.60121465,
      "num_input_tokens_seen": 169137170,
      "step": 7823,
      "time_per_iteration": 3.3526577949523926
    },
    {
      "auxiliary_loss_clip": 0.0110268,
      "auxiliary_loss_mlp": 0.00761865,
      "balance_loss_clip": 1.03864288,
      "balance_loss_mlp": 1.00026739,
      "epoch": 0.9407803763602477,
      "flos": 22165071903360.0,
      "grad_norm": 1.77364851509595,
      "language_loss": 0.59678477,
      "learning_rate": 3.6608685443207054e-08,
      "loss": 0.61543024,
      "num_input_tokens_seen": 169156320,
      "step": 7824,
      "time_per_iteration": 2.8513879776000977
    },
    {
      "auxiliary_loss_clip": 0.01122537,
      "auxiliary_loss_mlp": 0.01025614,
      "balance_loss_clip": 1.04032099,
      "balance_loss_mlp": 1.01895654,
      "epoch": 0.9409006192508867,
      "flos": 18879496911360.0,
      "grad_norm": 2.4005595114706284,
      "language_loss": 0.66831344,
      "learning_rate": 3.646047424157306e-08,
      "loss": 0.6897949,
      "num_input_tokens_seen": 169173295,
      "step": 7825,
      "time_per_iteration": 2.7306032180786133
    },
    {
      "auxiliary_loss_clip": 0.01137653,
      "auxiliary_loss_mlp": 0.01025671,
      "balance_loss_clip": 1.04222393,
      "balance_loss_mlp": 1.01790428,
      "epoch": 0.9410208621415259,
      "flos": 23368258800000.0,
      "grad_norm": 2.7889298689697233,
      "language_loss": 0.68924773,
      "learning_rate": 3.631256090732382e-08,
      "loss": 0.71088099,
      "num_input_tokens_seen": 169193755,
      "step": 7826,
      "time_per_iteration": 3.498997926712036
    },
    {
      "auxiliary_loss_clip": 0.01122911,
      "auxiliary_loss_mlp": 0.01023781,
      "balance_loss_clip": 1.04238105,
      "balance_loss_mlp": 1.01730776,
      "epoch": 0.941141105032165,
      "flos": 22742227635840.0,
      "grad_norm": 3.09677341447509,
      "language_loss": 0.82878339,
      "learning_rate": 3.6164945462897833e-08,
      "loss": 0.85025036,
      "num_input_tokens_seen": 169213045,
      "step": 7827,
      "time_per_iteration": 2.622340202331543
    },
    {
      "auxiliary_loss_clip": 0.01148543,
      "auxiliary_loss_mlp": 0.00761518,
      "balance_loss_clip": 1.04440129,
      "balance_loss_mlp": 1.00027084,
      "epoch": 0.941261347922804,
      "flos": 20704908130560.0,
      "grad_norm": 8.244878995842363,
      "language_loss": 0.7593354,
      "learning_rate": 3.6017627930687856e-08,
      "loss": 0.77843601,
      "num_input_tokens_seen": 169232870,
      "step": 7828,
      "time_per_iteration": 2.5137951374053955
    },
    {
      "auxiliary_loss_clip": 0.01101382,
      "auxiliary_loss_mlp": 0.01022187,
      "balance_loss_clip": 1.03564835,
      "balance_loss_mlp": 1.01470613,
      "epoch": 0.9413815908134432,
      "flos": 19421998997760.0,
      "grad_norm": 2.0917621894769067,
      "language_loss": 0.77102,
      "learning_rate": 3.587060833304267e-08,
      "loss": 0.7922557,
      "num_input_tokens_seen": 169251060,
      "step": 7829,
      "time_per_iteration": 3.5460567474365234
    },
    {
      "auxiliary_loss_clip": 0.01154413,
      "auxiliary_loss_mlp": 0.01024684,
      "balance_loss_clip": 1.04535949,
      "balance_loss_mlp": 1.01772511,
      "epoch": 0.9415018337040822,
      "flos": 17493452853120.0,
      "grad_norm": 1.9196464412862122,
      "language_loss": 0.64030641,
      "learning_rate": 3.5723886692264225e-08,
      "loss": 0.66209739,
      "num_input_tokens_seen": 169268600,
      "step": 7830,
      "time_per_iteration": 2.5743672847747803
    },
    {
      "auxiliary_loss_clip": 0.01133686,
      "auxiliary_loss_mlp": 0.01022951,
      "balance_loss_clip": 1.03984666,
      "balance_loss_mlp": 1.01594198,
      "epoch": 0.9416220765947213,
      "flos": 31831613343360.0,
      "grad_norm": 1.9359667889517151,
      "language_loss": 0.62073886,
      "learning_rate": 3.557746303061071e-08,
      "loss": 0.64230525,
      "num_input_tokens_seen": 169290355,
      "step": 7831,
      "time_per_iteration": 3.6985971927642822
    },
    {
      "auxiliary_loss_clip": 0.01135372,
      "auxiliary_loss_mlp": 0.01023854,
      "balance_loss_clip": 1.04197621,
      "balance_loss_mlp": 1.01646328,
      "epoch": 0.9417423194853605,
      "flos": 23511973115520.0,
      "grad_norm": 2.1234334634262058,
      "language_loss": 0.721129,
      "learning_rate": 3.543133737029391e-08,
      "loss": 0.74272126,
      "num_input_tokens_seen": 169310865,
      "step": 7832,
      "time_per_iteration": 2.702070951461792
    },
    {
      "auxiliary_loss_clip": 0.0115324,
      "auxiliary_loss_mlp": 0.01024108,
      "balance_loss_clip": 1.04359698,
      "balance_loss_mlp": 1.01715255,
      "epoch": 0.9418625623759995,
      "flos": 23915106432000.0,
      "grad_norm": 2.0110225017521888,
      "language_loss": 0.69028032,
      "learning_rate": 3.5285509733481214e-08,
      "loss": 0.71205384,
      "num_input_tokens_seen": 169330590,
      "step": 7833,
      "time_per_iteration": 2.592555522918701
    },
    {
      "auxiliary_loss_clip": 0.01145108,
      "auxiliary_loss_mlp": 0.01033032,
      "balance_loss_clip": 1.04166389,
      "balance_loss_mlp": 1.02597451,
      "epoch": 0.9419828052666386,
      "flos": 18076965292800.0,
      "grad_norm": 1.723215340504124,
      "language_loss": 0.76681441,
      "learning_rate": 3.513998014229469e-08,
      "loss": 0.78859586,
      "num_input_tokens_seen": 169349540,
      "step": 7834,
      "time_per_iteration": 3.540327787399292
    },
    {
      "auxiliary_loss_clip": 0.01138805,
      "auxiliary_loss_mlp": 0.01027925,
      "balance_loss_clip": 1.04408598,
      "balance_loss_mlp": 1.0207516,
      "epoch": 0.9421030481572777,
      "flos": 17712328377600.0,
      "grad_norm": 2.153471502770205,
      "language_loss": 0.86333424,
      "learning_rate": 3.499474861881069e-08,
      "loss": 0.88500148,
      "num_input_tokens_seen": 169366765,
      "step": 7835,
      "time_per_iteration": 2.670884132385254
    },
    {
      "auxiliary_loss_clip": 0.01096102,
      "auxiliary_loss_mlp": 0.01023628,
      "balance_loss_clip": 1.03959942,
      "balance_loss_mlp": 1.01667261,
      "epoch": 0.9422232910479168,
      "flos": 20194114775040.0,
      "grad_norm": 1.9154409575773295,
      "language_loss": 0.68223763,
      "learning_rate": 3.4849815185061136e-08,
      "loss": 0.70343494,
      "num_input_tokens_seen": 169386655,
      "step": 7836,
      "time_per_iteration": 2.7571349143981934
    },
    {
      "auxiliary_loss_clip": 0.01149502,
      "auxiliary_loss_mlp": 0.01022835,
      "balance_loss_clip": 1.04236627,
      "balance_loss_mlp": 1.01653171,
      "epoch": 0.9423435339385559,
      "flos": 18442571875200.0,
      "grad_norm": 1.8623034627055841,
      "language_loss": 0.7586689,
      "learning_rate": 3.470517986303223e-08,
      "loss": 0.78039229,
      "num_input_tokens_seen": 169405640,
      "step": 7837,
      "time_per_iteration": 2.5983200073242188
    },
    {
      "auxiliary_loss_clip": 0.0112291,
      "auxiliary_loss_mlp": 0.01027807,
      "balance_loss_clip": 1.04316759,
      "balance_loss_mlp": 1.0206815,
      "epoch": 0.942463776829195,
      "flos": 20080636732800.0,
      "grad_norm": 1.7681533829632001,
      "language_loss": 0.79160464,
      "learning_rate": 3.4560842674664856e-08,
      "loss": 0.81311178,
      "num_input_tokens_seen": 169424155,
      "step": 7838,
      "time_per_iteration": 2.6564249992370605
    },
    {
      "auxiliary_loss_clip": 0.01150405,
      "auxiliary_loss_mlp": 0.01026228,
      "balance_loss_clip": 1.04107678,
      "balance_loss_mlp": 1.0194273,
      "epoch": 0.9425840197198341,
      "flos": 22636255536000.0,
      "grad_norm": 2.1226708734079507,
      "language_loss": 0.75447607,
      "learning_rate": 3.441680364185506e-08,
      "loss": 0.77624243,
      "num_input_tokens_seen": 169444025,
      "step": 7839,
      "time_per_iteration": 2.620164155960083
    },
    {
      "auxiliary_loss_clip": 0.01140058,
      "auxiliary_loss_mlp": 0.01028359,
      "balance_loss_clip": 1.04436398,
      "balance_loss_mlp": 1.02050877,
      "epoch": 0.9427042626104731,
      "flos": 19937892084480.0,
      "grad_norm": 2.5904364599257823,
      "language_loss": 0.74555492,
      "learning_rate": 3.427306278645314e-08,
      "loss": 0.76723909,
      "num_input_tokens_seen": 169462480,
      "step": 7840,
      "time_per_iteration": 2.5902369022369385
    },
    {
      "auxiliary_loss_clip": 0.01107641,
      "auxiliary_loss_mlp": 0.01021272,
      "balance_loss_clip": 1.04004622,
      "balance_loss_mlp": 1.01489723,
      "epoch": 0.9428245055011123,
      "flos": 22856998567680.0,
      "grad_norm": 1.7829422666990453,
      "language_loss": 0.72964311,
      "learning_rate": 3.4129620130264767e-08,
      "loss": 0.75093222,
      "num_input_tokens_seen": 169480840,
      "step": 7841,
      "time_per_iteration": 2.7474796772003174
    },
    {
      "auxiliary_loss_clip": 0.0113879,
      "auxiliary_loss_mlp": 0.0076188,
      "balance_loss_clip": 1.04290938,
      "balance_loss_mlp": 1.00031424,
      "epoch": 0.9429447483917514,
      "flos": 20951757371520.0,
      "grad_norm": 2.481527226478723,
      "language_loss": 0.77835202,
      "learning_rate": 3.398647569505009e-08,
      "loss": 0.79735869,
      "num_input_tokens_seen": 169498265,
      "step": 7842,
      "time_per_iteration": 2.6336257457733154
    },
    {
      "auxiliary_loss_clip": 0.01127239,
      "auxiliary_loss_mlp": 0.01026547,
      "balance_loss_clip": 1.0415436,
      "balance_loss_mlp": 1.0196898,
      "epoch": 0.9430649912823904,
      "flos": 18843658116480.0,
      "grad_norm": 2.6690813613923834,
      "language_loss": 0.74677694,
      "learning_rate": 3.384362950252373e-08,
      "loss": 0.76831472,
      "num_input_tokens_seen": 169515235,
      "step": 7843,
      "time_per_iteration": 2.6857542991638184
    },
    {
      "auxiliary_loss_clip": 0.01136668,
      "auxiliary_loss_mlp": 0.01025235,
      "balance_loss_clip": 1.04246593,
      "balance_loss_mlp": 1.01821399,
      "epoch": 0.9431852341730296,
      "flos": 32556038837760.0,
      "grad_norm": 2.891247341447621,
      "language_loss": 0.5697732,
      "learning_rate": 3.3701081574355473e-08,
      "loss": 0.59139228,
      "num_input_tokens_seen": 169537195,
      "step": 7844,
      "time_per_iteration": 2.7464168071746826
    },
    {
      "auxiliary_loss_clip": 0.01045078,
      "auxiliary_loss_mlp": 0.01002146,
      "balance_loss_clip": 1.00953269,
      "balance_loss_mlp": 1.00111485,
      "epoch": 0.9433054770636686,
      "flos": 66904490252160.0,
      "grad_norm": 0.6402865208740424,
      "language_loss": 0.51644576,
      "learning_rate": 3.3558831932169796e-08,
      "loss": 0.53691792,
      "num_input_tokens_seen": 169605865,
      "step": 7845,
      "time_per_iteration": 3.294221878051758
    },
    {
      "auxiliary_loss_clip": 0.01149919,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 1.04484558,
      "balance_loss_mlp": 1.02077186,
      "epoch": 0.9434257199543077,
      "flos": 26140346916480.0,
      "grad_norm": 1.8589253929989527,
      "language_loss": 0.88326597,
      "learning_rate": 3.341688059754588e-08,
      "loss": 0.90504926,
      "num_input_tokens_seen": 169621520,
      "step": 7846,
      "time_per_iteration": 2.676137924194336
    },
    {
      "auxiliary_loss_clip": 0.01129967,
      "auxiliary_loss_mlp": 0.00762015,
      "balance_loss_clip": 1.04069257,
      "balance_loss_mlp": 1.00038695,
      "epoch": 0.9435459628449467,
      "flos": 25003486483200.0,
      "grad_norm": 2.0544964017410092,
      "language_loss": 0.7801609,
      "learning_rate": 3.327522759201762e-08,
      "loss": 0.79908073,
      "num_input_tokens_seen": 169641390,
      "step": 7847,
      "time_per_iteration": 2.7536017894744873
    },
    {
      "auxiliary_loss_clip": 0.01119163,
      "auxiliary_loss_mlp": 0.01023955,
      "balance_loss_clip": 1.03967416,
      "balance_loss_mlp": 1.01675463,
      "epoch": 0.9436662057355859,
      "flos": 22163240309760.0,
      "grad_norm": 2.150555544668834,
      "language_loss": 0.66450167,
      "learning_rate": 3.313387293707359e-08,
      "loss": 0.68593282,
      "num_input_tokens_seen": 169660095,
      "step": 7848,
      "time_per_iteration": 2.6711976528167725
    },
    {
      "auxiliary_loss_clip": 0.01118845,
      "auxiliary_loss_mlp": 0.01024193,
      "balance_loss_clip": 1.04276872,
      "balance_loss_mlp": 1.01646233,
      "epoch": 0.943786448626225,
      "flos": 20118522602880.0,
      "grad_norm": 1.9224505622188945,
      "language_loss": 0.68586612,
      "learning_rate": 3.29928166541571e-08,
      "loss": 0.70729649,
      "num_input_tokens_seen": 169679050,
      "step": 7849,
      "time_per_iteration": 2.6881844997406006
    },
    {
      "auxiliary_loss_clip": 0.01129392,
      "auxiliary_loss_mlp": 0.01023147,
      "balance_loss_clip": 1.04203224,
      "balance_loss_mlp": 1.01616168,
      "epoch": 0.943906691516864,
      "flos": 22090808534400.0,
      "grad_norm": 2.012574914533562,
      "language_loss": 0.80155396,
      "learning_rate": 3.2852058764666346e-08,
      "loss": 0.82307935,
      "num_input_tokens_seen": 169698150,
      "step": 7850,
      "time_per_iteration": 2.6562323570251465
    },
    {
      "auxiliary_loss_clip": 0.0111406,
      "auxiliary_loss_mlp": 0.0102548,
      "balance_loss_clip": 1.0438509,
      "balance_loss_mlp": 1.0189923,
      "epoch": 0.9440269344075032,
      "flos": 35298501212160.0,
      "grad_norm": 2.1482244338987764,
      "language_loss": 0.68547869,
      "learning_rate": 3.2711599289954264e-08,
      "loss": 0.70687401,
      "num_input_tokens_seen": 169722185,
      "step": 7851,
      "time_per_iteration": 2.8055436611175537
    },
    {
      "auxiliary_loss_clip": 0.01095691,
      "auxiliary_loss_mlp": 0.01024342,
      "balance_loss_clip": 1.03925538,
      "balance_loss_mlp": 1.01726139,
      "epoch": 0.9441471772981422,
      "flos": 19238136255360.0,
      "grad_norm": 1.7796143752308953,
      "language_loss": 0.77619755,
      "learning_rate": 3.257143825132847e-08,
      "loss": 0.79739785,
      "num_input_tokens_seen": 169740355,
      "step": 7852,
      "time_per_iteration": 3.7091825008392334
    },
    {
      "auxiliary_loss_clip": 0.0113509,
      "auxiliary_loss_mlp": 0.01023498,
      "balance_loss_clip": 1.04092598,
      "balance_loss_mlp": 1.0174396,
      "epoch": 0.9442674201887813,
      "flos": 25739799379200.0,
      "grad_norm": 1.808145066900263,
      "language_loss": 0.76273876,
      "learning_rate": 3.243157567005106e-08,
      "loss": 0.78432465,
      "num_input_tokens_seen": 169758535,
      "step": 7853,
      "time_per_iteration": 2.670938730239868
    },
    {
      "auxiliary_loss_clip": 0.01171353,
      "auxiliary_loss_mlp": 0.01025994,
      "balance_loss_clip": 1.05066454,
      "balance_loss_mlp": 1.01875854,
      "epoch": 0.9443876630794205,
      "flos": 15523321737600.0,
      "grad_norm": 3.676424373135769,
      "language_loss": 0.6425153,
      "learning_rate": 3.2292011567339296e-08,
      "loss": 0.66448879,
      "num_input_tokens_seen": 169776340,
      "step": 7854,
      "time_per_iteration": 2.539104700088501
    },
    {
      "auxiliary_loss_clip": 0.01151789,
      "auxiliary_loss_mlp": 0.00762125,
      "balance_loss_clip": 1.0429945,
      "balance_loss_mlp": 1.00032401,
      "epoch": 0.9445079059700595,
      "flos": 13400821128960.0,
      "grad_norm": 2.8967376786745187,
      "language_loss": 0.55890584,
      "learning_rate": 3.21527459643649e-08,
      "loss": 0.57804489,
      "num_input_tokens_seen": 169793225,
      "step": 7855,
      "time_per_iteration": 3.5756161212921143
    },
    {
      "auxiliary_loss_clip": 0.01152973,
      "auxiliary_loss_mlp": 0.01028058,
      "balance_loss_clip": 1.0445838,
      "balance_loss_mlp": 1.02121186,
      "epoch": 0.9446281488606986,
      "flos": 23659242877440.0,
      "grad_norm": 1.9345118044436524,
      "language_loss": 0.7420795,
      "learning_rate": 3.2013778882254536e-08,
      "loss": 0.76388985,
      "num_input_tokens_seen": 169812020,
      "step": 7856,
      "time_per_iteration": 3.5079874992370605
    },
    {
      "auxiliary_loss_clip": 0.01143321,
      "auxiliary_loss_mlp": 0.01026345,
      "balance_loss_clip": 1.04243731,
      "balance_loss_mlp": 1.01918662,
      "epoch": 0.9447483917513377,
      "flos": 25557337267200.0,
      "grad_norm": 1.8519482588984217,
      "language_loss": 0.75521028,
      "learning_rate": 3.1875110342088676e-08,
      "loss": 0.77690691,
      "num_input_tokens_seen": 169833470,
      "step": 7857,
      "time_per_iteration": 2.6735873222351074
    },
    {
      "auxiliary_loss_clip": 0.01130334,
      "auxiliary_loss_mlp": 0.01023101,
      "balance_loss_clip": 1.04193366,
      "balance_loss_mlp": 1.01570988,
      "epoch": 0.9448686346419768,
      "flos": 24535463247360.0,
      "grad_norm": 1.6165695169769632,
      "language_loss": 0.65628213,
      "learning_rate": 3.1736740364904035e-08,
      "loss": 0.67781651,
      "num_input_tokens_seen": 169854000,
      "step": 7858,
      "time_per_iteration": 2.7221643924713135
    },
    {
      "auxiliary_loss_clip": 0.01110037,
      "auxiliary_loss_mlp": 0.00762294,
      "balance_loss_clip": 1.04152942,
      "balance_loss_mlp": 1.00028765,
      "epoch": 0.9449888775326158,
      "flos": 14721256995840.0,
      "grad_norm": 2.1637079930197016,
      "language_loss": 0.77118349,
      "learning_rate": 3.159866897169094e-08,
      "loss": 0.7899068,
      "num_input_tokens_seen": 169872200,
      "step": 7859,
      "time_per_iteration": 2.68589186668396
    },
    {
      "auxiliary_loss_clip": 0.0113053,
      "auxiliary_loss_mlp": 0.01021013,
      "balance_loss_clip": 1.04301298,
      "balance_loss_mlp": 1.01411998,
      "epoch": 0.945109120423255,
      "flos": 15447873219840.0,
      "grad_norm": 1.9078750596760758,
      "language_loss": 0.75681674,
      "learning_rate": 3.146089618339487e-08,
      "loss": 0.77833217,
      "num_input_tokens_seen": 169889055,
      "step": 7860,
      "time_per_iteration": 3.549943447113037
    },
    {
      "auxiliary_loss_clip": 0.01122639,
      "auxiliary_loss_mlp": 0.01023042,
      "balance_loss_clip": 1.04011655,
      "balance_loss_mlp": 1.01607132,
      "epoch": 0.9452293633138941,
      "flos": 25448097029760.0,
      "grad_norm": 2.170510975199028,
      "language_loss": 0.6794796,
      "learning_rate": 3.132342202091554e-08,
      "loss": 0.70093644,
      "num_input_tokens_seen": 169909280,
      "step": 7861,
      "time_per_iteration": 2.712127923965454
    },
    {
      "auxiliary_loss_clip": 0.01165858,
      "auxiliary_loss_mlp": 0.01027352,
      "balance_loss_clip": 1.04472113,
      "balance_loss_mlp": 1.02011585,
      "epoch": 0.9453496062045331,
      "flos": 21215342350080.0,
      "grad_norm": 2.4412224811186056,
      "language_loss": 0.68627465,
      "learning_rate": 3.1186246505107595e-08,
      "loss": 0.70820677,
      "num_input_tokens_seen": 169928420,
      "step": 7862,
      "time_per_iteration": 2.6397109031677246
    },
    {
      "auxiliary_loss_clip": 0.01152053,
      "auxiliary_loss_mlp": 0.01027138,
      "balance_loss_clip": 1.04599452,
      "balance_loss_mlp": 1.02014327,
      "epoch": 0.9454698490951723,
      "flos": 20010898477440.0,
      "grad_norm": 1.7577200805806033,
      "language_loss": 0.83635545,
      "learning_rate": 3.104936965678084e-08,
      "loss": 0.85814738,
      "num_input_tokens_seen": 169946750,
      "step": 7863,
      "time_per_iteration": 2.603236198425293
    },
    {
      "auxiliary_loss_clip": 0.01152274,
      "auxiliary_loss_mlp": 0.01022159,
      "balance_loss_clip": 1.04385805,
      "balance_loss_mlp": 1.01509571,
      "epoch": 0.9455900919858113,
      "flos": 21069652786560.0,
      "grad_norm": 1.884431625539557,
      "language_loss": 0.82112896,
      "learning_rate": 3.091279149669956e-08,
      "loss": 0.84287333,
      "num_input_tokens_seen": 169965540,
      "step": 7864,
      "time_per_iteration": 2.655489444732666
    },
    {
      "auxiliary_loss_clip": 0.01151738,
      "auxiliary_loss_mlp": 0.00761958,
      "balance_loss_clip": 1.04512286,
      "balance_loss_mlp": 1.0002619,
      "epoch": 0.9457103348764504,
      "flos": 20740854666240.0,
      "grad_norm": 2.1082932355105193,
      "language_loss": 0.73691326,
      "learning_rate": 3.0776512045581624e-08,
      "loss": 0.75605023,
      "num_input_tokens_seen": 169984330,
      "step": 7865,
      "time_per_iteration": 2.6085777282714844
    },
    {
      "auxiliary_loss_clip": 0.01130716,
      "auxiliary_loss_mlp": 0.01027216,
      "balance_loss_clip": 1.04219747,
      "balance_loss_mlp": 1.01943731,
      "epoch": 0.9458305777670896,
      "flos": 21428363957760.0,
      "grad_norm": 1.7868353545588598,
      "language_loss": 0.77854562,
      "learning_rate": 3.0640531324101384e-08,
      "loss": 0.80012488,
      "num_input_tokens_seen": 170002095,
      "step": 7866,
      "time_per_iteration": 2.6421401500701904
    },
    {
      "auxiliary_loss_clip": 0.01155521,
      "auxiliary_loss_mlp": 0.01027199,
      "balance_loss_clip": 1.04809785,
      "balance_loss_mlp": 1.0193485,
      "epoch": 0.9459508206577286,
      "flos": 20011185786240.0,
      "grad_norm": 2.9441508092602344,
      "language_loss": 0.7557767,
      "learning_rate": 3.0504849352886554e-08,
      "loss": 0.77760386,
      "num_input_tokens_seen": 170020240,
      "step": 7867,
      "time_per_iteration": 2.565398693084717
    },
    {
      "auxiliary_loss_clip": 0.01150582,
      "auxiliary_loss_mlp": 0.01029576,
      "balance_loss_clip": 1.04554343,
      "balance_loss_mlp": 1.02182436,
      "epoch": 0.9460710635483677,
      "flos": 12166428291840.0,
      "grad_norm": 4.0364149872221935,
      "language_loss": 0.71309149,
      "learning_rate": 3.036946615252023e-08,
      "loss": 0.73489308,
      "num_input_tokens_seen": 170035770,
      "step": 7868,
      "time_per_iteration": 2.5777430534362793
    },
    {
      "auxiliary_loss_clip": 0.01141932,
      "auxiliary_loss_mlp": 0.01032129,
      "balance_loss_clip": 1.04285955,
      "balance_loss_mlp": 1.02522683,
      "epoch": 0.9461913064390068,
      "flos": 34276196229120.0,
      "grad_norm": 2.616990957507741,
      "language_loss": 0.67152512,
      "learning_rate": 3.0234381743539984e-08,
      "loss": 0.6932658,
      "num_input_tokens_seen": 170053385,
      "step": 7869,
      "time_per_iteration": 2.727832078933716
    },
    {
      "auxiliary_loss_clip": 0.01142097,
      "auxiliary_loss_mlp": 0.01023509,
      "balance_loss_clip": 1.04248309,
      "balance_loss_mlp": 1.01636815,
      "epoch": 0.9463115493296459,
      "flos": 19463763536640.0,
      "grad_norm": 2.0082512835847774,
      "language_loss": 0.79931319,
      "learning_rate": 3.0099596146437863e-08,
      "loss": 0.82096922,
      "num_input_tokens_seen": 170070490,
      "step": 7870,
      "time_per_iteration": 2.6754496097564697
    },
    {
      "auxiliary_loss_clip": 0.01062676,
      "auxiliary_loss_mlp": 0.01002641,
      "balance_loss_clip": 1.00913978,
      "balance_loss_mlp": 1.00160968,
      "epoch": 0.946431792220285,
      "flos": 70570824387840.0,
      "grad_norm": 0.7785239177397602,
      "language_loss": 0.60063326,
      "learning_rate": 2.996510938166086e-08,
      "loss": 0.62128645,
      "num_input_tokens_seen": 170133465,
      "step": 7871,
      "time_per_iteration": 3.27728271484375
    },
    {
      "auxiliary_loss_clip": 0.01148775,
      "auxiliary_loss_mlp": 0.01021976,
      "balance_loss_clip": 1.04512846,
      "balance_loss_mlp": 1.01571786,
      "epoch": 0.9465520351109241,
      "flos": 18947906363520.0,
      "grad_norm": 1.9345491631176246,
      "language_loss": 0.73896456,
      "learning_rate": 2.983092146960997e-08,
      "loss": 0.76067203,
      "num_input_tokens_seen": 170150810,
      "step": 7872,
      "time_per_iteration": 2.546621084213257
    },
    {
      "auxiliary_loss_clip": 0.01136145,
      "auxiliary_loss_mlp": 0.01024107,
      "balance_loss_clip": 1.03973544,
      "balance_loss_mlp": 1.01683271,
      "epoch": 0.9466722780015632,
      "flos": 19135647774720.0,
      "grad_norm": 2.199176005347075,
      "language_loss": 0.80182904,
      "learning_rate": 2.9697032430642256e-08,
      "loss": 0.82343155,
      "num_input_tokens_seen": 170169025,
      "step": 7873,
      "time_per_iteration": 2.6317009925842285
    },
    {
      "auxiliary_loss_clip": 0.01160958,
      "auxiliary_loss_mlp": 0.01021575,
      "balance_loss_clip": 1.04365051,
      "balance_loss_mlp": 1.01498246,
      "epoch": 0.9467925208922022,
      "flos": 17237912520960.0,
      "grad_norm": 2.1080195748630786,
      "language_loss": 0.73515105,
      "learning_rate": 2.9563442285067906e-08,
      "loss": 0.75697637,
      "num_input_tokens_seen": 170186070,
      "step": 7874,
      "time_per_iteration": 2.721874713897705
    },
    {
      "auxiliary_loss_clip": 0.01151361,
      "auxiliary_loss_mlp": 0.01023575,
      "balance_loss_clip": 1.04467607,
      "balance_loss_mlp": 1.01646996,
      "epoch": 0.9469127637828414,
      "flos": 29169016859520.0,
      "grad_norm": 1.9559064397152521,
      "language_loss": 0.79632425,
      "learning_rate": 2.943015105315294e-08,
      "loss": 0.81807363,
      "num_input_tokens_seen": 170206265,
      "step": 7875,
      "time_per_iteration": 2.700072765350342
    },
    {
      "auxiliary_loss_clip": 0.01111641,
      "auxiliary_loss_mlp": 0.01029234,
      "balance_loss_clip": 1.03938687,
      "balance_loss_mlp": 1.02158713,
      "epoch": 0.9470330066734804,
      "flos": 26030460234240.0,
      "grad_norm": 2.205715260630357,
      "language_loss": 0.6670866,
      "learning_rate": 2.929715875511718e-08,
      "loss": 0.68849534,
      "num_input_tokens_seen": 170225300,
      "step": 7876,
      "time_per_iteration": 2.7851738929748535
    },
    {
      "auxiliary_loss_clip": 0.01150744,
      "auxiliary_loss_mlp": 0.01027319,
      "balance_loss_clip": 1.041857,
      "balance_loss_mlp": 1.02025604,
      "epoch": 0.9471532495641195,
      "flos": 23440906056960.0,
      "grad_norm": 1.966223756285267,
      "language_loss": 0.70222205,
      "learning_rate": 2.9164465411135375e-08,
      "loss": 0.72400272,
      "num_input_tokens_seen": 170245070,
      "step": 7877,
      "time_per_iteration": 2.662336826324463
    },
    {
      "auxiliary_loss_clip": 0.01151483,
      "auxiliary_loss_mlp": 0.0102707,
      "balance_loss_clip": 1.04553914,
      "balance_loss_mlp": 1.01974511,
      "epoch": 0.9472734924547586,
      "flos": 15815850099840.0,
      "grad_norm": 1.8709866653497271,
      "language_loss": 0.80934906,
      "learning_rate": 2.9032071041337426e-08,
      "loss": 0.83113456,
      "num_input_tokens_seen": 170263305,
      "step": 7878,
      "time_per_iteration": 3.541469097137451
    },
    {
      "auxiliary_loss_clip": 0.01127551,
      "auxiliary_loss_mlp": 0.0102188,
      "balance_loss_clip": 1.04075789,
      "balance_loss_mlp": 1.01515019,
      "epoch": 0.9473937353453977,
      "flos": 11181793697280.0,
      "grad_norm": 1.6019417625161148,
      "language_loss": 0.72558171,
      "learning_rate": 2.889997566580704e-08,
      "loss": 0.74707603,
      "num_input_tokens_seen": 170281460,
      "step": 7879,
      "time_per_iteration": 2.6238603591918945
    },
    {
      "auxiliary_loss_clip": 0.0116741,
      "auxiliary_loss_mlp": 0.01024799,
      "balance_loss_clip": 1.04538226,
      "balance_loss_mlp": 1.01756024,
      "epoch": 0.9475139782360368,
      "flos": 25775530433280.0,
      "grad_norm": 1.690794623925786,
      "language_loss": 0.70086074,
      "learning_rate": 2.8768179304583086e-08,
      "loss": 0.72278285,
      "num_input_tokens_seen": 170303515,
      "step": 7880,
      "time_per_iteration": 3.6273372173309326
    },
    {
      "auxiliary_loss_clip": 0.01124576,
      "auxiliary_loss_mlp": 0.0102822,
      "balance_loss_clip": 1.04413033,
      "balance_loss_mlp": 1.0216043,
      "epoch": 0.9476342211266758,
      "flos": 22820046451200.0,
      "grad_norm": 1.5601543818308745,
      "language_loss": 0.73447388,
      "learning_rate": 2.8636681977659117e-08,
      "loss": 0.75600183,
      "num_input_tokens_seen": 170323165,
      "step": 7881,
      "time_per_iteration": 2.7378628253936768
    },
    {
      "auxiliary_loss_clip": 0.01105915,
      "auxiliary_loss_mlp": 0.01024855,
      "balance_loss_clip": 1.04171038,
      "balance_loss_mlp": 1.01701641,
      "epoch": 0.947754464017315,
      "flos": 20193611984640.0,
      "grad_norm": 2.317071342537622,
      "language_loss": 0.77832526,
      "learning_rate": 2.850548370498318e-08,
      "loss": 0.79963291,
      "num_input_tokens_seen": 170341005,
      "step": 7882,
      "time_per_iteration": 2.722477912902832
    },
    {
      "auxiliary_loss_clip": 0.01150128,
      "auxiliary_loss_mlp": 0.01023743,
      "balance_loss_clip": 1.04214072,
      "balance_loss_mlp": 1.01713896,
      "epoch": 0.9478747069079541,
      "flos": 24717925359360.0,
      "grad_norm": 1.5809888693799885,
      "language_loss": 0.71323019,
      "learning_rate": 2.8374584506457798e-08,
      "loss": 0.7349689,
      "num_input_tokens_seen": 170362280,
      "step": 7883,
      "time_per_iteration": 3.609607696533203
    },
    {
      "auxiliary_loss_clip": 0.01134516,
      "auxiliary_loss_mlp": 0.01024114,
      "balance_loss_clip": 1.04232812,
      "balance_loss_mlp": 1.01710498,
      "epoch": 0.9479949497985931,
      "flos": 21361355136000.0,
      "grad_norm": 2.6493193055394446,
      "language_loss": 0.67379683,
      "learning_rate": 2.824398440193998e-08,
      "loss": 0.69538319,
      "num_input_tokens_seen": 170381080,
      "step": 7884,
      "time_per_iteration": 2.718432903289795
    },
    {
      "auxiliary_loss_clip": 0.01100297,
      "auxiliary_loss_mlp": 0.01023556,
      "balance_loss_clip": 1.03878093,
      "balance_loss_mlp": 1.01630223,
      "epoch": 0.9481151926892323,
      "flos": 18148606968960.0,
      "grad_norm": 2.319459285527836,
      "language_loss": 0.71401632,
      "learning_rate": 2.811368341124232e-08,
      "loss": 0.73525488,
      "num_input_tokens_seen": 170400150,
      "step": 7885,
      "time_per_iteration": 2.749744415283203
    },
    {
      "auxiliary_loss_clip": 0.01137346,
      "auxiliary_loss_mlp": 0.01023478,
      "balance_loss_clip": 1.04294777,
      "balance_loss_mlp": 1.01685047,
      "epoch": 0.9482354355798713,
      "flos": 22128012046080.0,
      "grad_norm": 3.1312871845594543,
      "language_loss": 0.68416899,
      "learning_rate": 2.7983681554131222e-08,
      "loss": 0.70577729,
      "num_input_tokens_seen": 170420410,
      "step": 7886,
      "time_per_iteration": 3.5633254051208496
    },
    {
      "auxiliary_loss_clip": 0.01133988,
      "auxiliary_loss_mlp": 0.01025216,
      "balance_loss_clip": 1.04110086,
      "balance_loss_mlp": 1.017223,
      "epoch": 0.9483556784705104,
      "flos": 19063072344960.0,
      "grad_norm": 2.097583928177926,
      "language_loss": 0.70713681,
      "learning_rate": 2.7853978850327365e-08,
      "loss": 0.72872877,
      "num_input_tokens_seen": 170439580,
      "step": 7887,
      "time_per_iteration": 2.709789276123047
    },
    {
      "auxiliary_loss_clip": 0.01122646,
      "auxiliary_loss_mlp": 0.01026401,
      "balance_loss_clip": 1.04458261,
      "balance_loss_mlp": 1.01999664,
      "epoch": 0.9484759213611496,
      "flos": 25777110631680.0,
      "grad_norm": 1.8507677180515818,
      "language_loss": 0.8731817,
      "learning_rate": 2.7724575319507225e-08,
      "loss": 0.89467216,
      "num_input_tokens_seen": 170459290,
      "step": 7888,
      "time_per_iteration": 2.7674224376678467
    },
    {
      "auxiliary_loss_clip": 0.0114784,
      "auxiliary_loss_mlp": 0.01025059,
      "balance_loss_clip": 1.04115129,
      "balance_loss_mlp": 1.01882386,
      "epoch": 0.9485961642517886,
      "flos": 20667740532480.0,
      "grad_norm": 3.8850557952153957,
      "language_loss": 0.7714653,
      "learning_rate": 2.759547098130044e-08,
      "loss": 0.79319429,
      "num_input_tokens_seen": 170478020,
      "step": 7889,
      "time_per_iteration": 2.6014273166656494
    },
    {
      "auxiliary_loss_clip": 0.0116178,
      "auxiliary_loss_mlp": 0.01027663,
      "balance_loss_clip": 1.0443747,
      "balance_loss_mlp": 1.02063549,
      "epoch": 0.9487164071424277,
      "flos": 22674069578880.0,
      "grad_norm": 1.94859433417905,
      "language_loss": 0.77006108,
      "learning_rate": 2.746666585529267e-08,
      "loss": 0.79195547,
      "num_input_tokens_seen": 170498295,
      "step": 7890,
      "time_per_iteration": 2.614652395248413
    },
    {
      "auxiliary_loss_clip": 0.01144473,
      "auxiliary_loss_mlp": 0.01020323,
      "balance_loss_clip": 1.04322267,
      "balance_loss_mlp": 1.01322412,
      "epoch": 0.9488366500330668,
      "flos": 38726461716480.0,
      "grad_norm": 2.343402868644441,
      "language_loss": 0.73985994,
      "learning_rate": 2.73381599610234e-08,
      "loss": 0.76150787,
      "num_input_tokens_seen": 170518695,
      "step": 7891,
      "time_per_iteration": 2.799907684326172
    },
    {
      "auxiliary_loss_clip": 0.01144175,
      "auxiliary_loss_mlp": 0.01022154,
      "balance_loss_clip": 1.04072726,
      "balance_loss_mlp": 1.01500714,
      "epoch": 0.9489568929237059,
      "flos": 27890920149120.0,
      "grad_norm": 1.9990590770487782,
      "language_loss": 0.71405971,
      "learning_rate": 2.7209953317987033e-08,
      "loss": 0.73572296,
      "num_input_tokens_seen": 170539735,
      "step": 7892,
      "time_per_iteration": 2.709376811981201
    },
    {
      "auxiliary_loss_clip": 0.01150324,
      "auxiliary_loss_mlp": 0.01025327,
      "balance_loss_clip": 1.04421163,
      "balance_loss_mlp": 1.01834464,
      "epoch": 0.9490771358143449,
      "flos": 33580642291200.0,
      "grad_norm": 2.4083439569196923,
      "language_loss": 0.78563654,
      "learning_rate": 2.7082045945631793e-08,
      "loss": 0.80739307,
      "num_input_tokens_seen": 170561950,
      "step": 7893,
      "time_per_iteration": 2.7012529373168945
    },
    {
      "auxiliary_loss_clip": 0.01113068,
      "auxiliary_loss_mlp": 0.01021313,
      "balance_loss_clip": 1.03908563,
      "balance_loss_mlp": 1.01458621,
      "epoch": 0.9491973787049841,
      "flos": 14793796512000.0,
      "grad_norm": 2.123326056242042,
      "language_loss": 0.69360036,
      "learning_rate": 2.6954437863361712e-08,
      "loss": 0.71494412,
      "num_input_tokens_seen": 170579865,
      "step": 7894,
      "time_per_iteration": 2.7140274047851562
    },
    {
      "auxiliary_loss_clip": 0.01094293,
      "auxiliary_loss_mlp": 0.01020354,
      "balance_loss_clip": 1.03717887,
      "balance_loss_mlp": 1.01390457,
      "epoch": 0.9493176215956232,
      "flos": 25332535998720.0,
      "grad_norm": 2.172716658965047,
      "language_loss": 0.70698136,
      "learning_rate": 2.6827129090534862e-08,
      "loss": 0.72812784,
      "num_input_tokens_seen": 170600165,
      "step": 7895,
      "time_per_iteration": 2.7593164443969727
    },
    {
      "auxiliary_loss_clip": 0.01136634,
      "auxiliary_loss_mlp": 0.01028504,
      "balance_loss_clip": 1.04415321,
      "balance_loss_mlp": 1.02105975,
      "epoch": 0.9494378644862622,
      "flos": 21029971236480.0,
      "grad_norm": 1.7848795393214096,
      "language_loss": 0.77538872,
      "learning_rate": 2.670011964646335e-08,
      "loss": 0.7970401,
      "num_input_tokens_seen": 170618845,
      "step": 7896,
      "time_per_iteration": 2.6858904361724854
    },
    {
      "auxiliary_loss_clip": 0.01085643,
      "auxiliary_loss_mlp": 0.01026104,
      "balance_loss_clip": 1.03352451,
      "balance_loss_mlp": 1.01907039,
      "epoch": 0.9495581073769014,
      "flos": 15195134148480.0,
      "grad_norm": 2.3784811447099417,
      "language_loss": 0.6815356,
      "learning_rate": 2.657340955041487e-08,
      "loss": 0.70265305,
      "num_input_tokens_seen": 170637620,
      "step": 7897,
      "time_per_iteration": 2.780773401260376
    },
    {
      "auxiliary_loss_clip": 0.01136218,
      "auxiliary_loss_mlp": 0.0102643,
      "balance_loss_clip": 1.04599667,
      "balance_loss_mlp": 1.01881003,
      "epoch": 0.9496783502675404,
      "flos": 28616566705920.0,
      "grad_norm": 2.792765804563931,
      "language_loss": 0.71661323,
      "learning_rate": 2.6446998821611167e-08,
      "loss": 0.73823977,
      "num_input_tokens_seen": 170657815,
      "step": 7898,
      "time_per_iteration": 2.793844223022461
    },
    {
      "auxiliary_loss_clip": 0.01110983,
      "auxiliary_loss_mlp": 0.01025136,
      "balance_loss_clip": 1.03972173,
      "balance_loss_mlp": 1.01774228,
      "epoch": 0.9497985931581795,
      "flos": 14866874732160.0,
      "grad_norm": 2.9851947587891945,
      "language_loss": 0.71620119,
      "learning_rate": 2.6320887479228228e-08,
      "loss": 0.73756242,
      "num_input_tokens_seen": 170674415,
      "step": 7899,
      "time_per_iteration": 2.7338685989379883
    },
    {
      "auxiliary_loss_clip": 0.01140004,
      "auxiliary_loss_mlp": 0.01026625,
      "balance_loss_clip": 1.04300427,
      "balance_loss_mlp": 1.01972246,
      "epoch": 0.9499188360488187,
      "flos": 27193319136000.0,
      "grad_norm": 2.209504673183118,
      "language_loss": 0.72498256,
      "learning_rate": 2.619507554239786e-08,
      "loss": 0.74664885,
      "num_input_tokens_seen": 170692975,
      "step": 7900,
      "time_per_iteration": 2.699279546737671
    },
    {
      "auxiliary_loss_clip": 0.01133394,
      "auxiliary_loss_mlp": 0.01024472,
      "balance_loss_clip": 1.04210353,
      "balance_loss_mlp": 1.01715291,
      "epoch": 0.9500390789394577,
      "flos": 24316479982080.0,
      "grad_norm": 1.6838114410893363,
      "language_loss": 0.69775701,
      "learning_rate": 2.606956303020502e-08,
      "loss": 0.71933562,
      "num_input_tokens_seen": 170713780,
      "step": 7901,
      "time_per_iteration": 2.692856788635254
    },
    {
      "auxiliary_loss_clip": 0.01150499,
      "auxiliary_loss_mlp": 0.01027082,
      "balance_loss_clip": 1.04479337,
      "balance_loss_mlp": 1.01991487,
      "epoch": 0.9501593218300968,
      "flos": 14354752573440.0,
      "grad_norm": 1.7072490692838345,
      "language_loss": 0.84120607,
      "learning_rate": 2.5944349961690036e-08,
      "loss": 0.86298192,
      "num_input_tokens_seen": 170730800,
      "step": 7902,
      "time_per_iteration": 2.7365963459014893
    },
    {
      "auxiliary_loss_clip": 0.01122037,
      "auxiliary_loss_mlp": 0.0102219,
      "balance_loss_clip": 1.04157209,
      "balance_loss_mlp": 1.01559186,
      "epoch": 0.9502795647207359,
      "flos": 38728113742080.0,
      "grad_norm": 1.7914652830427467,
      "language_loss": 0.73148125,
      "learning_rate": 2.581943635584749e-08,
      "loss": 0.75292355,
      "num_input_tokens_seen": 170753630,
      "step": 7903,
      "time_per_iteration": 2.8693509101867676
    },
    {
      "auxiliary_loss_clip": 0.01128155,
      "auxiliary_loss_mlp": 0.01025875,
      "balance_loss_clip": 1.0406301,
      "balance_loss_mlp": 1.01892173,
      "epoch": 0.950399807611375,
      "flos": 40808023799040.0,
      "grad_norm": 1.5866928507319307,
      "language_loss": 0.65905428,
      "learning_rate": 2.569482223162689e-08,
      "loss": 0.68059456,
      "num_input_tokens_seen": 170777605,
      "step": 7904,
      "time_per_iteration": 3.752171754837036
    },
    {
      "auxiliary_loss_clip": 0.01150346,
      "auxiliary_loss_mlp": 0.01028041,
      "balance_loss_clip": 1.04280281,
      "balance_loss_mlp": 1.0207963,
      "epoch": 0.950520050502014,
      "flos": 23440403266560.0,
      "grad_norm": 1.7790623976542161,
      "language_loss": 0.72540545,
      "learning_rate": 2.5570507607932e-08,
      "loss": 0.7471894,
      "num_input_tokens_seen": 170797520,
      "step": 7905,
      "time_per_iteration": 2.7800979614257812
    },
    {
      "auxiliary_loss_clip": 0.01155307,
      "auxiliary_loss_mlp": 0.01022687,
      "balance_loss_clip": 1.04462159,
      "balance_loss_mlp": 1.01543283,
      "epoch": 0.9506402933926532,
      "flos": 17783718658560.0,
      "grad_norm": 3.518803437021101,
      "language_loss": 0.64251161,
      "learning_rate": 2.54464925036213e-08,
      "loss": 0.6642915,
      "num_input_tokens_seen": 170814810,
      "step": 7906,
      "time_per_iteration": 2.6155145168304443
    },
    {
      "auxiliary_loss_clip": 0.01151781,
      "auxiliary_loss_mlp": 0.01022901,
      "balance_loss_clip": 1.04515624,
      "balance_loss_mlp": 1.01632929,
      "epoch": 0.9507605362832923,
      "flos": 32561928668160.0,
      "grad_norm": 1.856450609561302,
      "language_loss": 0.60919785,
      "learning_rate": 2.532277693750773e-08,
      "loss": 0.63094467,
      "num_input_tokens_seen": 170835735,
      "step": 7907,
      "time_per_iteration": 3.6344966888427734
    },
    {
      "auxiliary_loss_clip": 0.01106928,
      "auxiliary_loss_mlp": 0.01024902,
      "balance_loss_clip": 1.04285598,
      "balance_loss_mlp": 1.01791644,
      "epoch": 0.9508807791739313,
      "flos": 19602054898560.0,
      "grad_norm": 6.082223907026243,
      "language_loss": 0.75835758,
      "learning_rate": 2.5199360928358948e-08,
      "loss": 0.77967584,
      "num_input_tokens_seen": 170852970,
      "step": 7908,
      "time_per_iteration": 3.580583333969116
    },
    {
      "auxiliary_loss_clip": 0.01139307,
      "auxiliary_loss_mlp": 0.00761198,
      "balance_loss_clip": 1.04000032,
      "balance_loss_mlp": 1.00028348,
      "epoch": 0.9510010220645704,
      "flos": 21471852349440.0,
      "grad_norm": 1.6789100981354363,
      "language_loss": 0.8711741,
      "learning_rate": 2.507624449489665e-08,
      "loss": 0.89017916,
      "num_input_tokens_seen": 170871600,
      "step": 7909,
      "time_per_iteration": 2.667489767074585
    },
    {
      "auxiliary_loss_clip": 0.01138484,
      "auxiliary_loss_mlp": 0.01026925,
      "balance_loss_clip": 1.04481065,
      "balance_loss_mlp": 1.01923347,
      "epoch": 0.9511212649552095,
      "flos": 18879999701760.0,
      "grad_norm": 2.301005170683377,
      "language_loss": 0.64959335,
      "learning_rate": 2.495342765579811e-08,
      "loss": 0.67124748,
      "num_input_tokens_seen": 170890260,
      "step": 7910,
      "time_per_iteration": 2.6745145320892334
    },
    {
      "auxiliary_loss_clip": 0.01107438,
      "auxiliary_loss_mlp": 0.01023191,
      "balance_loss_clip": 1.04191732,
      "balance_loss_mlp": 1.01676261,
      "epoch": 0.9512415078458486,
      "flos": 20810521094400.0,
      "grad_norm": 1.969515219392795,
      "language_loss": 0.71234626,
      "learning_rate": 2.4830910429693984e-08,
      "loss": 0.73365253,
      "num_input_tokens_seen": 170910220,
      "step": 7911,
      "time_per_iteration": 2.770380735397339
    },
    {
      "auxiliary_loss_clip": 0.01165229,
      "auxiliary_loss_mlp": 0.01026976,
      "balance_loss_clip": 1.04555464,
      "balance_loss_mlp": 1.02050257,
      "epoch": 0.9513617507364877,
      "flos": 18369565482240.0,
      "grad_norm": 1.8911599146420979,
      "language_loss": 0.79833037,
      "learning_rate": 2.470869283517052e-08,
      "loss": 0.82025242,
      "num_input_tokens_seen": 170928255,
      "step": 7912,
      "time_per_iteration": 3.5448007583618164
    },
    {
      "auxiliary_loss_clip": 0.01143412,
      "auxiliary_loss_mlp": 0.01026992,
      "balance_loss_clip": 1.04186416,
      "balance_loss_mlp": 1.02042639,
      "epoch": 0.9514819936271268,
      "flos": 25010166412800.0,
      "grad_norm": 1.6034321280480226,
      "language_loss": 0.7715615,
      "learning_rate": 2.458677489076777e-08,
      "loss": 0.79326546,
      "num_input_tokens_seen": 170949265,
      "step": 7913,
      "time_per_iteration": 2.727651357650757
    },
    {
      "auxiliary_loss_clip": 0.01142093,
      "auxiliary_loss_mlp": 0.0101969,
      "balance_loss_clip": 1.04161525,
      "balance_loss_mlp": 1.01307058,
      "epoch": 0.9516022365177659,
      "flos": 18662129758080.0,
      "grad_norm": 1.58896196670677,
      "language_loss": 0.82859874,
      "learning_rate": 2.446515661498072e-08,
      "loss": 0.85021651,
      "num_input_tokens_seen": 170968595,
      "step": 7914,
      "time_per_iteration": 2.6539628505706787
    },
    {
      "auxiliary_loss_clip": 0.01094074,
      "auxiliary_loss_mlp": 0.01020702,
      "balance_loss_clip": 1.0390749,
      "balance_loss_mlp": 1.01394629,
      "epoch": 0.9517224794084049,
      "flos": 25372109808000.0,
      "grad_norm": 2.006193083445214,
      "language_loss": 0.7441982,
      "learning_rate": 2.434383802625861e-08,
      "loss": 0.76534605,
      "num_input_tokens_seen": 170987550,
      "step": 7915,
      "time_per_iteration": 2.801605224609375
    },
    {
      "auxiliary_loss_clip": 0.0112335,
      "auxiliary_loss_mlp": 0.01020669,
      "balance_loss_clip": 1.03876233,
      "balance_loss_mlp": 1.01370442,
      "epoch": 0.9518427222990441,
      "flos": 21470918595840.0,
      "grad_norm": 1.757371130323707,
      "language_loss": 0.7396239,
      "learning_rate": 2.4222819143005168e-08,
      "loss": 0.76106411,
      "num_input_tokens_seen": 171007145,
      "step": 7916,
      "time_per_iteration": 2.8161075115203857
    },
    {
      "auxiliary_loss_clip": 0.01165002,
      "auxiliary_loss_mlp": 0.01022501,
      "balance_loss_clip": 1.04724371,
      "balance_loss_mlp": 1.01546168,
      "epoch": 0.9519629651896832,
      "flos": 21033634423680.0,
      "grad_norm": 1.820261508054258,
      "language_loss": 0.80657184,
      "learning_rate": 2.4102099983579706e-08,
      "loss": 0.82844687,
      "num_input_tokens_seen": 171026295,
      "step": 7917,
      "time_per_iteration": 2.6008799076080322
    },
    {
      "auxiliary_loss_clip": 0.01151619,
      "auxiliary_loss_mlp": 0.01024942,
      "balance_loss_clip": 1.04360652,
      "balance_loss_mlp": 1.01729441,
      "epoch": 0.9520832080803222,
      "flos": 21689219502720.0,
      "grad_norm": 1.6703885841444936,
      "language_loss": 0.77064753,
      "learning_rate": 2.3981680566294236e-08,
      "loss": 0.79241312,
      "num_input_tokens_seen": 171045895,
      "step": 7918,
      "time_per_iteration": 2.674933910369873
    },
    {
      "auxiliary_loss_clip": 0.01164153,
      "auxiliary_loss_mlp": 0.01022867,
      "balance_loss_clip": 1.04727054,
      "balance_loss_mlp": 1.0160985,
      "epoch": 0.9522034509709614,
      "flos": 23145289125120.0,
      "grad_norm": 1.8000072815995982,
      "language_loss": 0.73369431,
      "learning_rate": 2.3861560909416822e-08,
      "loss": 0.75556451,
      "num_input_tokens_seen": 171065445,
      "step": 7919,
      "time_per_iteration": 2.6103403568267822
    },
    {
      "auxiliary_loss_clip": 0.01108243,
      "auxiliary_loss_mlp": 0.01023626,
      "balance_loss_clip": 1.04037511,
      "balance_loss_mlp": 1.0165267,
      "epoch": 0.9523236938616004,
      "flos": 24679428958080.0,
      "grad_norm": 1.6763994099814123,
      "language_loss": 0.82955658,
      "learning_rate": 2.3741741031169325e-08,
      "loss": 0.85087526,
      "num_input_tokens_seen": 171085015,
      "step": 7920,
      "time_per_iteration": 2.760673761367798
    },
    {
      "auxiliary_loss_clip": 0.01101841,
      "auxiliary_loss_mlp": 0.01026838,
      "balance_loss_clip": 1.03765452,
      "balance_loss_mlp": 1.01948225,
      "epoch": 0.9524439367522395,
      "flos": 22672309812480.0,
      "grad_norm": 1.887409950985664,
      "language_loss": 0.71557426,
      "learning_rate": 2.3622220949728544e-08,
      "loss": 0.73686105,
      "num_input_tokens_seen": 171103900,
      "step": 7921,
      "time_per_iteration": 2.7461600303649902
    },
    {
      "auxiliary_loss_clip": 0.01143892,
      "auxiliary_loss_mlp": 0.0102299,
      "balance_loss_clip": 1.04239392,
      "balance_loss_mlp": 1.01568234,
      "epoch": 0.9525641796428787,
      "flos": 34055525024640.0,
      "grad_norm": 2.5585080989216555,
      "language_loss": 0.61436707,
      "learning_rate": 2.3503000683225526e-08,
      "loss": 0.63603592,
      "num_input_tokens_seen": 171121615,
      "step": 7922,
      "time_per_iteration": 2.7127275466918945
    },
    {
      "auxiliary_loss_clip": 0.01165478,
      "auxiliary_loss_mlp": 0.0102164,
      "balance_loss_clip": 1.0453366,
      "balance_loss_mlp": 1.01454103,
      "epoch": 0.9526844225335177,
      "flos": 16727083251840.0,
      "grad_norm": 2.0937550500766013,
      "language_loss": 0.8447994,
      "learning_rate": 2.3384080249745585e-08,
      "loss": 0.86667055,
      "num_input_tokens_seen": 171139505,
      "step": 7923,
      "time_per_iteration": 2.533846855163574
    },
    {
      "auxiliary_loss_clip": 0.01113594,
      "auxiliary_loss_mlp": 0.01024785,
      "balance_loss_clip": 1.04166079,
      "balance_loss_mlp": 1.01789188,
      "epoch": 0.9528046654241568,
      "flos": 36939367330560.0,
      "grad_norm": 2.492034667803386,
      "language_loss": 0.83010638,
      "learning_rate": 2.3265459667329178e-08,
      "loss": 0.85149014,
      "num_input_tokens_seen": 171158995,
      "step": 7924,
      "time_per_iteration": 2.8950395584106445
    },
    {
      "auxiliary_loss_clip": 0.01137456,
      "auxiliary_loss_mlp": 0.01022921,
      "balance_loss_clip": 1.04255676,
      "balance_loss_mlp": 1.01592898,
      "epoch": 0.9529249083147959,
      "flos": 18255010032000.0,
      "grad_norm": 2.3043803981460287,
      "language_loss": 0.86678237,
      "learning_rate": 2.31471389539708e-08,
      "loss": 0.88838619,
      "num_input_tokens_seen": 171176120,
      "step": 7925,
      "time_per_iteration": 2.6174850463867188
    },
    {
      "auxiliary_loss_clip": 0.01152606,
      "auxiliary_loss_mlp": 0.00761996,
      "balance_loss_clip": 1.04526424,
      "balance_loss_mlp": 1.00028992,
      "epoch": 0.953045151205435,
      "flos": 28658438985600.0,
      "grad_norm": 2.3200702662722876,
      "language_loss": 0.72603798,
      "learning_rate": 2.3029118127619872e-08,
      "loss": 0.74518406,
      "num_input_tokens_seen": 171195835,
      "step": 7926,
      "time_per_iteration": 2.714331865310669
    },
    {
      "auxiliary_loss_clip": 0.01128609,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.04150188,
      "balance_loss_mlp": 1.02176309,
      "epoch": 0.953165394096074,
      "flos": 21835232288640.0,
      "grad_norm": 2.0538696713358653,
      "language_loss": 0.87076724,
      "learning_rate": 2.2911397206179628e-08,
      "loss": 0.89234149,
      "num_input_tokens_seen": 171212585,
      "step": 7927,
      "time_per_iteration": 2.653094530105591
    },
    {
      "auxiliary_loss_clip": 0.01162579,
      "auxiliary_loss_mlp": 0.01023597,
      "balance_loss_clip": 1.04502308,
      "balance_loss_mlp": 1.01662934,
      "epoch": 0.9532856369867132,
      "flos": 19975059682560.0,
      "grad_norm": 1.8762704649059068,
      "language_loss": 0.62663239,
      "learning_rate": 2.279397620750845e-08,
      "loss": 0.64849412,
      "num_input_tokens_seen": 171231630,
      "step": 7928,
      "time_per_iteration": 2.59883975982666
    },
    {
      "auxiliary_loss_clip": 0.01131378,
      "auxiliary_loss_mlp": 0.01022971,
      "balance_loss_clip": 1.03955197,
      "balance_loss_mlp": 1.01611936,
      "epoch": 0.9534058798773523,
      "flos": 15049588239360.0,
      "grad_norm": 1.8945873038632957,
      "language_loss": 0.78603816,
      "learning_rate": 2.2676855149419195e-08,
      "loss": 0.80758166,
      "num_input_tokens_seen": 171248800,
      "step": 7929,
      "time_per_iteration": 2.6418237686157227
    },
    {
      "auxiliary_loss_clip": 0.01133361,
      "auxiliary_loss_mlp": 0.01022976,
      "balance_loss_clip": 1.04684055,
      "balance_loss_mlp": 1.01637745,
      "epoch": 0.9535261227679913,
      "flos": 17602800831360.0,
      "grad_norm": 2.4995334419944126,
      "language_loss": 0.75563341,
      "learning_rate": 2.2560034049678988e-08,
      "loss": 0.77719676,
      "num_input_tokens_seen": 171263150,
      "step": 7930,
      "time_per_iteration": 3.5574288368225098
    },
    {
      "auxiliary_loss_clip": 0.01170618,
      "auxiliary_loss_mlp": 0.01023407,
      "balance_loss_clip": 1.04869223,
      "balance_loss_mlp": 1.01597452,
      "epoch": 0.9536463656586305,
      "flos": 23142954741120.0,
      "grad_norm": 1.6583857749583228,
      "language_loss": 0.75529075,
      "learning_rate": 2.2443512926008988e-08,
      "loss": 0.77723098,
      "num_input_tokens_seen": 171282480,
      "step": 7931,
      "time_per_iteration": 2.5710456371307373
    },
    {
      "auxiliary_loss_clip": 0.01124106,
      "auxiliary_loss_mlp": 0.01025517,
      "balance_loss_clip": 1.03970385,
      "balance_loss_mlp": 1.01855779,
      "epoch": 0.9537666085492695,
      "flos": 18625033987200.0,
      "grad_norm": 2.258827753687845,
      "language_loss": 0.69759595,
      "learning_rate": 2.2327291796085946e-08,
      "loss": 0.71909219,
      "num_input_tokens_seen": 171300840,
      "step": 7932,
      "time_per_iteration": 2.645685911178589
    },
    {
      "auxiliary_loss_clip": 0.01164962,
      "auxiliary_loss_mlp": 0.01025305,
      "balance_loss_clip": 1.04439521,
      "balance_loss_mlp": 1.01808715,
      "epoch": 0.9538868514399086,
      "flos": 18989347680000.0,
      "grad_norm": 2.8837468984512205,
      "language_loss": 0.77756137,
      "learning_rate": 2.2211370677540197e-08,
      "loss": 0.79946405,
      "num_input_tokens_seen": 171317365,
      "step": 7933,
      "time_per_iteration": 3.5404863357543945
    },
    {
      "auxiliary_loss_clip": 0.01166636,
      "auxiliary_loss_mlp": 0.01027581,
      "balance_loss_clip": 1.04581881,
      "balance_loss_mlp": 1.0206902,
      "epoch": 0.9540070943305478,
      "flos": 16800556521600.0,
      "grad_norm": 3.156107640579223,
      "language_loss": 0.77946281,
      "learning_rate": 2.2095749587957012e-08,
      "loss": 0.80140501,
      "num_input_tokens_seen": 171335270,
      "step": 7934,
      "time_per_iteration": 3.50911545753479
    },
    {
      "auxiliary_loss_clip": 0.01133819,
      "auxiliary_loss_mlp": 0.01024697,
      "balance_loss_clip": 1.03966522,
      "balance_loss_mlp": 1.0174427,
      "epoch": 0.9541273372211868,
      "flos": 20156911263360.0,
      "grad_norm": 2.0470806587604344,
      "language_loss": 0.69336867,
      "learning_rate": 2.1980428544876138e-08,
      "loss": 0.7149539,
      "num_input_tokens_seen": 171353910,
      "step": 7935,
      "time_per_iteration": 2.682501792907715
    },
    {
      "auxiliary_loss_clip": 0.01100803,
      "auxiliary_loss_mlp": 0.01028209,
      "balance_loss_clip": 1.03462195,
      "balance_loss_mlp": 1.02115726,
      "epoch": 0.9542475801118259,
      "flos": 26725511381760.0,
      "grad_norm": 1.7201135275053703,
      "language_loss": 0.74331832,
      "learning_rate": 2.1865407565791584e-08,
      "loss": 0.76460844,
      "num_input_tokens_seen": 171375480,
      "step": 7936,
      "time_per_iteration": 2.755011558532715
    },
    {
      "auxiliary_loss_clip": 0.01135769,
      "auxiliary_loss_mlp": 0.01024413,
      "balance_loss_clip": 1.04053283,
      "balance_loss_mlp": 1.01682258,
      "epoch": 0.954367823002465,
      "flos": 23330911633920.0,
      "grad_norm": 2.777596555837622,
      "language_loss": 0.77361274,
      "learning_rate": 2.175068666815183e-08,
      "loss": 0.79521459,
      "num_input_tokens_seen": 171396320,
      "step": 7937,
      "time_per_iteration": 2.640911340713501
    },
    {
      "auxiliary_loss_clip": 0.01123746,
      "auxiliary_loss_mlp": 0.010245,
      "balance_loss_clip": 1.04210246,
      "balance_loss_mlp": 1.01693654,
      "epoch": 0.9544880658931041,
      "flos": 14902713527040.0,
      "grad_norm": 2.1849710114076806,
      "language_loss": 0.78855419,
      "learning_rate": 2.163626586935985e-08,
      "loss": 0.81003666,
      "num_input_tokens_seen": 171412860,
      "step": 7938,
      "time_per_iteration": 3.619279146194458
    },
    {
      "auxiliary_loss_clip": 0.01146062,
      "auxiliary_loss_mlp": 0.01026425,
      "balance_loss_clip": 1.04113889,
      "balance_loss_mlp": 1.01906657,
      "epoch": 0.9546083087837431,
      "flos": 29095902725760.0,
      "grad_norm": 1.8192039694937527,
      "language_loss": 0.63126552,
      "learning_rate": 2.1522145186773755e-08,
      "loss": 0.6529904,
      "num_input_tokens_seen": 171431780,
      "step": 7939,
      "time_per_iteration": 2.688520908355713
    },
    {
      "auxiliary_loss_clip": 0.01134759,
      "auxiliary_loss_mlp": 0.01022675,
      "balance_loss_clip": 1.04232836,
      "balance_loss_mlp": 1.01653922,
      "epoch": 0.9547285516743822,
      "flos": 21142335957120.0,
      "grad_norm": 2.554043643452691,
      "language_loss": 0.85471243,
      "learning_rate": 2.140832463770481e-08,
      "loss": 0.87628675,
      "num_input_tokens_seen": 171450975,
      "step": 7940,
      "time_per_iteration": 2.6702940464019775
    },
    {
      "auxiliary_loss_clip": 0.01139684,
      "auxiliary_loss_mlp": 0.01027219,
      "balance_loss_clip": 1.04179168,
      "balance_loss_mlp": 1.01990891,
      "epoch": 0.9548487945650214,
      "flos": 27490157130240.0,
      "grad_norm": 2.3593776291491713,
      "language_loss": 0.76135528,
      "learning_rate": 2.129480423941987e-08,
      "loss": 0.78302431,
      "num_input_tokens_seen": 171467645,
      "step": 7941,
      "time_per_iteration": 2.7260656356811523
    },
    {
      "auxiliary_loss_clip": 0.01142977,
      "auxiliary_loss_mlp": 0.0102137,
      "balance_loss_clip": 1.0451771,
      "balance_loss_mlp": 1.01454544,
      "epoch": 0.9549690374556604,
      "flos": 22273198819200.0,
      "grad_norm": 1.6271166189774162,
      "language_loss": 0.80155623,
      "learning_rate": 2.1181584009140052e-08,
      "loss": 0.82319975,
      "num_input_tokens_seen": 171487185,
      "step": 7942,
      "time_per_iteration": 2.6485323905944824
    },
    {
      "auxiliary_loss_clip": 0.01130478,
      "auxiliary_loss_mlp": 0.01023288,
      "balance_loss_clip": 1.04138196,
      "balance_loss_mlp": 1.01640391,
      "epoch": 0.9550892803462995,
      "flos": 17595294888960.0,
      "grad_norm": 1.893950385903907,
      "language_loss": 0.83715129,
      "learning_rate": 2.10686639640405e-08,
      "loss": 0.85868895,
      "num_input_tokens_seen": 171501275,
      "step": 7943,
      "time_per_iteration": 2.6878015995025635
    },
    {
      "auxiliary_loss_clip": 0.01155556,
      "auxiliary_loss_mlp": 0.01023434,
      "balance_loss_clip": 1.04551613,
      "balance_loss_mlp": 1.01671696,
      "epoch": 0.9552095232369386,
      "flos": 24353144789760.0,
      "grad_norm": 1.7115391463137746,
      "language_loss": 0.81409615,
      "learning_rate": 2.0956044121251294e-08,
      "loss": 0.83588606,
      "num_input_tokens_seen": 171520060,
      "step": 7944,
      "time_per_iteration": 2.6297781467437744
    },
    {
      "auxiliary_loss_clip": 0.01121167,
      "auxiliary_loss_mlp": 0.01029629,
      "balance_loss_clip": 1.04309309,
      "balance_loss_mlp": 1.02198458,
      "epoch": 0.9553297661275777,
      "flos": 22746860490240.0,
      "grad_norm": 2.116911083868702,
      "language_loss": 0.81081831,
      "learning_rate": 2.084372449785654e-08,
      "loss": 0.83232629,
      "num_input_tokens_seen": 171539895,
      "step": 7945,
      "time_per_iteration": 2.7179758548736572
    },
    {
      "auxiliary_loss_clip": 0.01134215,
      "auxiliary_loss_mlp": 0.01024025,
      "balance_loss_clip": 1.04244399,
      "balance_loss_mlp": 1.01715302,
      "epoch": 0.9554500090182168,
      "flos": 15413866018560.0,
      "grad_norm": 1.6105062475312109,
      "language_loss": 0.68888021,
      "learning_rate": 2.0731705110895282e-08,
      "loss": 0.71046257,
      "num_input_tokens_seen": 171557385,
      "step": 7946,
      "time_per_iteration": 2.6266591548919678
    },
    {
      "auxiliary_loss_clip": 0.01154655,
      "auxiliary_loss_mlp": 0.01026036,
      "balance_loss_clip": 1.0462513,
      "balance_loss_mlp": 1.01883292,
      "epoch": 0.9555702519088559,
      "flos": 23513517400320.0,
      "grad_norm": 2.1137727344909845,
      "language_loss": 0.86586177,
      "learning_rate": 2.0619985977360587e-08,
      "loss": 0.88766867,
      "num_input_tokens_seen": 171575705,
      "step": 7947,
      "time_per_iteration": 2.646944761276245
    },
    {
      "auxiliary_loss_clip": 0.01108077,
      "auxiliary_loss_mlp": 0.010238,
      "balance_loss_clip": 1.03644514,
      "balance_loss_mlp": 1.01701081,
      "epoch": 0.955690494799495,
      "flos": 22962072827520.0,
      "grad_norm": 1.7858859910691471,
      "language_loss": 0.76840699,
      "learning_rate": 2.0508567114200237e-08,
      "loss": 0.78972578,
      "num_input_tokens_seen": 171595620,
      "step": 7948,
      "time_per_iteration": 2.7115275859832764
    },
    {
      "auxiliary_loss_clip": 0.01139489,
      "auxiliary_loss_mlp": 0.01023561,
      "balance_loss_clip": 1.04204535,
      "balance_loss_mlp": 1.017079,
      "epoch": 0.955810737690134,
      "flos": 26031250333440.0,
      "grad_norm": 2.5869060511988726,
      "language_loss": 0.78903323,
      "learning_rate": 2.0397448538316485e-08,
      "loss": 0.8106637,
      "num_input_tokens_seen": 171616660,
      "step": 7949,
      "time_per_iteration": 2.682366371154785
    },
    {
      "auxiliary_loss_clip": 0.01117272,
      "auxiliary_loss_mlp": 0.0102672,
      "balance_loss_clip": 1.04017138,
      "balance_loss_mlp": 1.01943016,
      "epoch": 0.9559309805807732,
      "flos": 20849951249280.0,
      "grad_norm": 2.570137211363715,
      "language_loss": 0.67008114,
      "learning_rate": 2.028663026656563e-08,
      "loss": 0.69152105,
      "num_input_tokens_seen": 171635515,
      "step": 7950,
      "time_per_iteration": 2.77748441696167
    },
    {
      "auxiliary_loss_clip": 0.01161469,
      "auxiliary_loss_mlp": 0.00762199,
      "balance_loss_clip": 1.04451871,
      "balance_loss_mlp": 1.00032401,
      "epoch": 0.9560512234714122,
      "flos": 21578219498880.0,
      "grad_norm": 1.967345902086176,
      "language_loss": 0.72083783,
      "learning_rate": 2.0176112315758885e-08,
      "loss": 0.74007446,
      "num_input_tokens_seen": 171653305,
      "step": 7951,
      "time_per_iteration": 2.5791947841644287
    },
    {
      "auxiliary_loss_clip": 0.01116054,
      "auxiliary_loss_mlp": 0.01023091,
      "balance_loss_clip": 1.04031146,
      "balance_loss_mlp": 1.01629853,
      "epoch": 0.9561714663620513,
      "flos": 17450144029440.0,
      "grad_norm": 2.576858704771275,
      "language_loss": 0.69054985,
      "learning_rate": 2.0065894702661957e-08,
      "loss": 0.7119413,
      "num_input_tokens_seen": 171669980,
      "step": 7952,
      "time_per_iteration": 2.804614305496216
    },
    {
      "auxiliary_loss_clip": 0.01113722,
      "auxiliary_loss_mlp": 0.00761983,
      "balance_loss_clip": 1.03823471,
      "balance_loss_mlp": 1.00030136,
      "epoch": 0.9562917092526905,
      "flos": 26098510550400.0,
      "grad_norm": 1.865450349924282,
      "language_loss": 0.77898192,
      "learning_rate": 1.9955977443994577e-08,
      "loss": 0.79773897,
      "num_input_tokens_seen": 171689970,
      "step": 7953,
      "time_per_iteration": 2.6989500522613525
    },
    {
      "auxiliary_loss_clip": 0.01134742,
      "auxiliary_loss_mlp": 0.01029195,
      "balance_loss_clip": 1.04228652,
      "balance_loss_mlp": 1.02152991,
      "epoch": 0.9564119521433295,
      "flos": 24096742531200.0,
      "grad_norm": 2.265582349287895,
      "language_loss": 0.62432325,
      "learning_rate": 1.9846360556430965e-08,
      "loss": 0.64596266,
      "num_input_tokens_seen": 171708270,
      "step": 7954,
      "time_per_iteration": 2.6592764854431152
    },
    {
      "auxiliary_loss_clip": 0.01164635,
      "auxiliary_loss_mlp": 0.01026773,
      "balance_loss_clip": 1.04522729,
      "balance_loss_mlp": 1.01981139,
      "epoch": 0.9565321950339686,
      "flos": 32008903896960.0,
      "grad_norm": 2.2689334426236023,
      "language_loss": 0.61949772,
      "learning_rate": 1.973704405660004e-08,
      "loss": 0.64141178,
      "num_input_tokens_seen": 171729385,
      "step": 7955,
      "time_per_iteration": 2.643402338027954
    },
    {
      "auxiliary_loss_clip": 0.01093017,
      "auxiliary_loss_mlp": 0.01025781,
      "balance_loss_clip": 1.03783703,
      "balance_loss_mlp": 1.01883125,
      "epoch": 0.9566524379246077,
      "flos": 23588642695680.0,
      "grad_norm": 1.837561897287415,
      "language_loss": 0.77990878,
      "learning_rate": 1.9628027961085203e-08,
      "loss": 0.8010968,
      "num_input_tokens_seen": 171752615,
      "step": 7956,
      "time_per_iteration": 3.714174747467041
    },
    {
      "auxiliary_loss_clip": 0.01109381,
      "auxiliary_loss_mlp": 0.01029593,
      "balance_loss_clip": 1.03639674,
      "balance_loss_mlp": 1.02271497,
      "epoch": 0.9567726808152468,
      "flos": 38067716240640.0,
      "grad_norm": 1.8612571163574678,
      "language_loss": 0.83936429,
      "learning_rate": 1.9519312286423894e-08,
      "loss": 0.86075401,
      "num_input_tokens_seen": 171775810,
      "step": 7957,
      "time_per_iteration": 2.8357603549957275
    },
    {
      "auxiliary_loss_clip": 0.011489,
      "auxiliary_loss_mlp": 0.01018759,
      "balance_loss_clip": 1.04546845,
      "balance_loss_mlp": 1.0119133,
      "epoch": 0.9568929237058859,
      "flos": 22744059229440.0,
      "grad_norm": 3.077918035807891,
      "language_loss": 0.77598387,
      "learning_rate": 1.9410897049108255e-08,
      "loss": 0.79766047,
      "num_input_tokens_seen": 171795090,
      "step": 7958,
      "time_per_iteration": 2.684954881668091
    },
    {
      "auxiliary_loss_clip": 0.01173368,
      "auxiliary_loss_mlp": 0.01030786,
      "balance_loss_clip": 1.0500133,
      "balance_loss_mlp": 1.02299857,
      "epoch": 0.957013166596525,
      "flos": 23841633162240.0,
      "grad_norm": 9.349232641071948,
      "language_loss": 0.91332757,
      "learning_rate": 1.9302782265584905e-08,
      "loss": 0.93536913,
      "num_input_tokens_seen": 171815755,
      "step": 7959,
      "time_per_iteration": 3.577113389968872
    },
    {
      "auxiliary_loss_clip": 0.01098741,
      "auxiliary_loss_mlp": 0.01026395,
      "balance_loss_clip": 1.04040051,
      "balance_loss_mlp": 1.0194689,
      "epoch": 0.9571334094871641,
      "flos": 17639286071040.0,
      "grad_norm": 2.526377313424507,
      "language_loss": 0.87095004,
      "learning_rate": 1.9194967952254282e-08,
      "loss": 0.89220142,
      "num_input_tokens_seen": 171834330,
      "step": 7960,
      "time_per_iteration": 3.593712329864502
    },
    {
      "auxiliary_loss_clip": 0.01151683,
      "auxiliary_loss_mlp": 0.01020676,
      "balance_loss_clip": 1.04584432,
      "balance_loss_mlp": 1.01390767,
      "epoch": 0.9572536523778031,
      "flos": 15369623441280.0,
      "grad_norm": 3.4499541938474354,
      "language_loss": 0.804088,
      "learning_rate": 1.9087454125472635e-08,
      "loss": 0.82581162,
      "num_input_tokens_seen": 171848805,
      "step": 7961,
      "time_per_iteration": 2.601853370666504
    },
    {
      "auxiliary_loss_clip": 0.01166601,
      "auxiliary_loss_mlp": 0.01026958,
      "balance_loss_clip": 1.0459491,
      "balance_loss_mlp": 1.01954961,
      "epoch": 0.9573738952684423,
      "flos": 24969838417920.0,
      "grad_norm": 1.9441228474075551,
      "language_loss": 0.7898258,
      "learning_rate": 1.8980240801548696e-08,
      "loss": 0.81176144,
      "num_input_tokens_seen": 171867995,
      "step": 7962,
      "time_per_iteration": 2.5859153270721436
    },
    {
      "auxiliary_loss_clip": 0.01135575,
      "auxiliary_loss_mlp": 0.01023922,
      "balance_loss_clip": 1.04587114,
      "balance_loss_mlp": 1.01661491,
      "epoch": 0.9574941381590814,
      "flos": 25769461034880.0,
      "grad_norm": 1.6831816995819224,
      "language_loss": 0.74123371,
      "learning_rate": 1.8873327996747458e-08,
      "loss": 0.76282871,
      "num_input_tokens_seen": 171886495,
      "step": 7963,
      "time_per_iteration": 2.749119997024536
    },
    {
      "auxiliary_loss_clip": 0.01152934,
      "auxiliary_loss_mlp": 0.01030868,
      "balance_loss_clip": 1.04279149,
      "balance_loss_mlp": 1.02359676,
      "epoch": 0.9576143810497204,
      "flos": 32307178435200.0,
      "grad_norm": 1.7037481979176172,
      "language_loss": 0.65877098,
      "learning_rate": 1.8766715727287053e-08,
      "loss": 0.68060899,
      "num_input_tokens_seen": 171908200,
      "step": 7964,
      "time_per_iteration": 3.6807610988616943
    },
    {
      "auxiliary_loss_clip": 0.01155225,
      "auxiliary_loss_mlp": 0.00762037,
      "balance_loss_clip": 1.04367316,
      "balance_loss_mlp": 1.00032282,
      "epoch": 0.9577346239403596,
      "flos": 27745733376000.0,
      "grad_norm": 1.6066808825284544,
      "language_loss": 0.79603362,
      "learning_rate": 1.8660404009340546e-08,
      "loss": 0.81520617,
      "num_input_tokens_seen": 171928650,
      "step": 7965,
      "time_per_iteration": 2.642902135848999
    },
    {
      "auxiliary_loss_clip": 0.01054341,
      "auxiliary_loss_mlp": 0.01002992,
      "balance_loss_clip": 1.00972164,
      "balance_loss_mlp": 1.00194252,
      "epoch": 0.9578548668309986,
      "flos": 57468313710720.0,
      "grad_norm": 0.8891598496903537,
      "language_loss": 0.59491438,
      "learning_rate": 1.8554392859035485e-08,
      "loss": 0.61548769,
      "num_input_tokens_seen": 171986400,
      "step": 7966,
      "time_per_iteration": 3.209813117980957
    },
    {
      "auxiliary_loss_clip": 0.01083465,
      "auxiliary_loss_mlp": 0.01021541,
      "balance_loss_clip": 1.03600955,
      "balance_loss_mlp": 1.01448989,
      "epoch": 0.9579751097216377,
      "flos": 19756040503680.0,
      "grad_norm": 2.0209124305783197,
      "language_loss": 0.7890569,
      "learning_rate": 1.8448682292453444e-08,
      "loss": 0.81010687,
      "num_input_tokens_seen": 172005475,
      "step": 7967,
      "time_per_iteration": 2.7390992641448975
    },
    {
      "auxiliary_loss_clip": 0.01163366,
      "auxiliary_loss_mlp": 0.01023869,
      "balance_loss_clip": 1.04489541,
      "balance_loss_mlp": 1.0172534,
      "epoch": 0.9580953526122769,
      "flos": 18041270152320.0,
      "grad_norm": 1.6639292591412034,
      "language_loss": 0.66089725,
      "learning_rate": 1.8343272325631154e-08,
      "loss": 0.68276966,
      "num_input_tokens_seen": 172024420,
      "step": 7968,
      "time_per_iteration": 2.6867029666900635
    },
    {
      "auxiliary_loss_clip": 0.01085371,
      "auxiliary_loss_mlp": 0.00762467,
      "balance_loss_clip": 1.03813207,
      "balance_loss_mlp": 1.00033879,
      "epoch": 0.9582155955029159,
      "flos": 24270154416000.0,
      "grad_norm": 2.5464879130510307,
      "language_loss": 0.78319293,
      "learning_rate": 1.8238162974558492e-08,
      "loss": 0.80167127,
      "num_input_tokens_seen": 172038350,
      "step": 7969,
      "time_per_iteration": 2.801637887954712
    },
    {
      "auxiliary_loss_clip": 0.01133874,
      "auxiliary_loss_mlp": 0.01020667,
      "balance_loss_clip": 1.04277575,
      "balance_loss_mlp": 1.01396155,
      "epoch": 0.958335838393555,
      "flos": 22783309816320.0,
      "grad_norm": 1.9692470266974884,
      "language_loss": 0.74794221,
      "learning_rate": 1.8133354255181144e-08,
      "loss": 0.76948762,
      "num_input_tokens_seen": 172058665,
      "step": 7970,
      "time_per_iteration": 2.6603810787200928
    },
    {
      "auxiliary_loss_clip": 0.01145792,
      "auxiliary_loss_mlp": 0.01024424,
      "balance_loss_clip": 1.04238987,
      "balance_loss_mlp": 1.01788807,
      "epoch": 0.958456081284194,
      "flos": 16911484698240.0,
      "grad_norm": 1.839437656371464,
      "language_loss": 0.74608666,
      "learning_rate": 1.802884618339795e-08,
      "loss": 0.76778883,
      "num_input_tokens_seen": 172077470,
      "step": 7971,
      "time_per_iteration": 2.652435779571533
    },
    {
      "auxiliary_loss_clip": 0.01154154,
      "auxiliary_loss_mlp": 0.01021657,
      "balance_loss_clip": 1.0461545,
      "balance_loss_mlp": 1.01439762,
      "epoch": 0.9585763241748332,
      "flos": 19974951941760.0,
      "grad_norm": 3.5693207914651413,
      "language_loss": 0.8092317,
      "learning_rate": 1.7924638775062894e-08,
      "loss": 0.83098984,
      "num_input_tokens_seen": 172096590,
      "step": 7972,
      "time_per_iteration": 2.6020991802215576
    },
    {
      "auxiliary_loss_clip": 0.01115683,
      "auxiliary_loss_mlp": 0.01026111,
      "balance_loss_clip": 1.04120541,
      "balance_loss_mlp": 1.01955748,
      "epoch": 0.9586965670654722,
      "flos": 21395649646080.0,
      "grad_norm": 1.8516435253301873,
      "language_loss": 0.81781358,
      "learning_rate": 1.7820732045984444e-08,
      "loss": 0.83923149,
      "num_input_tokens_seen": 172116735,
      "step": 7973,
      "time_per_iteration": 2.6952946186065674
    },
    {
      "auxiliary_loss_clip": 0.01148949,
      "auxiliary_loss_mlp": 0.01029607,
      "balance_loss_clip": 1.04339683,
      "balance_loss_mlp": 1.02139318,
      "epoch": 0.9588168099561113,
      "flos": 21435115714560.0,
      "grad_norm": 2.2627666653327085,
      "language_loss": 0.74097651,
      "learning_rate": 1.7717126011924655e-08,
      "loss": 0.76276207,
      "num_input_tokens_seen": 172138320,
      "step": 7974,
      "time_per_iteration": 2.7159955501556396
    },
    {
      "auxiliary_loss_clip": 0.01100178,
      "auxiliary_loss_mlp": 0.01025243,
      "balance_loss_clip": 1.03521824,
      "balance_loss_mlp": 1.01772666,
      "epoch": 0.9589370528467505,
      "flos": 11763761852160.0,
      "grad_norm": 3.632036312309814,
      "language_loss": 0.76487684,
      "learning_rate": 1.7613820688600957e-08,
      "loss": 0.78613108,
      "num_input_tokens_seen": 172154225,
      "step": 7975,
      "time_per_iteration": 2.664675235748291
    },
    {
      "auxiliary_loss_clip": 0.01141906,
      "auxiliary_loss_mlp": 0.0102551,
      "balance_loss_clip": 1.04179764,
      "balance_loss_mlp": 1.01886725,
      "epoch": 0.9590572957373895,
      "flos": 23441516588160.0,
      "grad_norm": 1.9131523148593965,
      "language_loss": 0.78517592,
      "learning_rate": 1.7510816091684588e-08,
      "loss": 0.80685008,
      "num_input_tokens_seen": 172174150,
      "step": 7976,
      "time_per_iteration": 2.64736008644104
    },
    {
      "auxiliary_loss_clip": 0.01137704,
      "auxiliary_loss_mlp": 0.01023919,
      "balance_loss_clip": 1.04275465,
      "balance_loss_mlp": 1.01585412,
      "epoch": 0.9591775386280286,
      "flos": 22528272274560.0,
      "grad_norm": 2.627351993178401,
      "language_loss": 0.78693974,
      "learning_rate": 1.740811223680083e-08,
      "loss": 0.80855596,
      "num_input_tokens_seen": 172191005,
      "step": 7977,
      "time_per_iteration": 2.6496148109436035
    },
    {
      "auxiliary_loss_clip": 0.01164076,
      "auxiliary_loss_mlp": 0.01025097,
      "balance_loss_clip": 1.04376721,
      "balance_loss_mlp": 1.01816797,
      "epoch": 0.9592977815186677,
      "flos": 18186959715840.0,
      "grad_norm": 2.1560047048482254,
      "language_loss": 0.74194109,
      "learning_rate": 1.7305709139530334e-08,
      "loss": 0.76383281,
      "num_input_tokens_seen": 172209785,
      "step": 7978,
      "time_per_iteration": 2.5967350006103516
    },
    {
      "auxiliary_loss_clip": 0.01142963,
      "auxiliary_loss_mlp": 0.01027499,
      "balance_loss_clip": 1.03999114,
      "balance_loss_mlp": 1.02025151,
      "epoch": 0.9594180244093068,
      "flos": 16537797555840.0,
      "grad_norm": 2.2428925279422782,
      "language_loss": 0.74721503,
      "learning_rate": 1.7203606815407334e-08,
      "loss": 0.76891959,
      "num_input_tokens_seen": 172224380,
      "step": 7979,
      "time_per_iteration": 2.691249370574951
    },
    {
      "auxiliary_loss_clip": 0.01141145,
      "auxiliary_loss_mlp": 0.01028128,
      "balance_loss_clip": 1.04366446,
      "balance_loss_mlp": 1.0208447,
      "epoch": 0.9595382672999458,
      "flos": 20554334317440.0,
      "grad_norm": 1.781692717624645,
      "language_loss": 0.79113424,
      "learning_rate": 1.7101805279920557e-08,
      "loss": 0.81282699,
      "num_input_tokens_seen": 172242540,
      "step": 7980,
      "time_per_iteration": 2.6922121047973633
    },
    {
      "auxiliary_loss_clip": 0.01167407,
      "auxiliary_loss_mlp": 0.01023756,
      "balance_loss_clip": 1.04842281,
      "balance_loss_mlp": 1.01664853,
      "epoch": 0.959658510190585,
      "flos": 22638266697600.0,
      "grad_norm": 2.8286937508406833,
      "language_loss": 0.80885559,
      "learning_rate": 1.7000304548513643e-08,
      "loss": 0.83076727,
      "num_input_tokens_seen": 172262645,
      "step": 7981,
      "time_per_iteration": 2.6083686351776123
    },
    {
      "auxiliary_loss_clip": 0.01119429,
      "auxiliary_loss_mlp": 0.01026052,
      "balance_loss_clip": 1.03989744,
      "balance_loss_mlp": 1.01904213,
      "epoch": 0.9597787530812241,
      "flos": 19135252725120.0,
      "grad_norm": 2.1111562870176277,
      "language_loss": 0.82907569,
      "learning_rate": 1.6899104636583394e-08,
      "loss": 0.85053051,
      "num_input_tokens_seen": 172280695,
      "step": 7982,
      "time_per_iteration": 3.6461403369903564
    },
    {
      "auxiliary_loss_clip": 0.01054613,
      "auxiliary_loss_mlp": 0.01001313,
      "balance_loss_clip": 1.00931883,
      "balance_loss_mlp": 1.00031722,
      "epoch": 0.9598989959718631,
      "flos": 60098124055680.0,
      "grad_norm": 0.7469289332973486,
      "language_loss": 0.61889464,
      "learning_rate": 1.6798205559482638e-08,
      "loss": 0.63945395,
      "num_input_tokens_seen": 172343075,
      "step": 7983,
      "time_per_iteration": 3.3206589221954346
    },
    {
      "auxiliary_loss_clip": 0.01121098,
      "auxiliary_loss_mlp": 0.01024163,
      "balance_loss_clip": 1.04040742,
      "balance_loss_mlp": 1.01703167,
      "epoch": 0.9600192388625023,
      "flos": 20886795624960.0,
      "grad_norm": 2.218078167680888,
      "language_loss": 0.76757663,
      "learning_rate": 1.669760733251713e-08,
      "loss": 0.78902924,
      "num_input_tokens_seen": 172361950,
      "step": 7984,
      "time_per_iteration": 2.692739486694336
    },
    {
      "auxiliary_loss_clip": 0.01101996,
      "auxiliary_loss_mlp": 0.0102027,
      "balance_loss_clip": 1.03789282,
      "balance_loss_mlp": 1.01336741,
      "epoch": 0.9601394817531413,
      "flos": 20445740524800.0,
      "grad_norm": 1.710721910445185,
      "language_loss": 0.82278615,
      "learning_rate": 1.659730997094755e-08,
      "loss": 0.8440088,
      "num_input_tokens_seen": 172380440,
      "step": 7985,
      "time_per_iteration": 3.722456216812134
    },
    {
      "auxiliary_loss_clip": 0.0114326,
      "auxiliary_loss_mlp": 0.01025615,
      "balance_loss_clip": 1.04217744,
      "balance_loss_mlp": 1.01867414,
      "epoch": 0.9602597246437804,
      "flos": 21507152440320.0,
      "grad_norm": 1.7766881293499743,
      "language_loss": 0.62159342,
      "learning_rate": 1.6497313489989283e-08,
      "loss": 0.64328218,
      "num_input_tokens_seen": 172400265,
      "step": 7986,
      "time_per_iteration": 3.4650235176086426
    },
    {
      "auxiliary_loss_clip": 0.01105785,
      "auxiliary_loss_mlp": 0.01025749,
      "balance_loss_clip": 1.03422689,
      "balance_loss_mlp": 1.01836455,
      "epoch": 0.9603799675344196,
      "flos": 29935099152000.0,
      "grad_norm": 2.368701321346479,
      "language_loss": 0.70235276,
      "learning_rate": 1.639761790481131e-08,
      "loss": 0.7236681,
      "num_input_tokens_seen": 172421145,
      "step": 7987,
      "time_per_iteration": 2.789889097213745
    },
    {
      "auxiliary_loss_clip": 0.01154169,
      "auxiliary_loss_mlp": 0.01020458,
      "balance_loss_clip": 1.04555631,
      "balance_loss_mlp": 1.01361823,
      "epoch": 0.9605002104250586,
      "flos": 28001525103360.0,
      "grad_norm": 1.926154866188746,
      "language_loss": 0.79087281,
      "learning_rate": 1.6298223230537754e-08,
      "loss": 0.81261909,
      "num_input_tokens_seen": 172438945,
      "step": 7988,
      "time_per_iteration": 2.653923988342285
    },
    {
      "auxiliary_loss_clip": 0.01137804,
      "auxiliary_loss_mlp": 0.00762261,
      "balance_loss_clip": 1.04367614,
      "balance_loss_mlp": 1.00030327,
      "epoch": 0.9606204533156977,
      "flos": 35590490870400.0,
      "grad_norm": 2.0013344973134273,
      "language_loss": 0.69663686,
      "learning_rate": 1.619912948224611e-08,
      "loss": 0.71563751,
      "num_input_tokens_seen": 172460150,
      "step": 7989,
      "time_per_iteration": 2.765037775039673
    },
    {
      "auxiliary_loss_clip": 0.01118318,
      "auxiliary_loss_mlp": 0.01028363,
      "balance_loss_clip": 1.04088926,
      "balance_loss_mlp": 1.02146065,
      "epoch": 0.9607406962063368,
      "flos": 26574614346240.0,
      "grad_norm": 2.8015594465537212,
      "language_loss": 0.61202389,
      "learning_rate": 1.6100336674969682e-08,
      "loss": 0.63349068,
      "num_input_tokens_seen": 172478990,
      "step": 7990,
      "time_per_iteration": 3.6251304149627686
    },
    {
      "auxiliary_loss_clip": 0.01110073,
      "auxiliary_loss_mlp": 0.01025445,
      "balance_loss_clip": 1.03625524,
      "balance_loss_mlp": 1.01818514,
      "epoch": 0.9608609390969759,
      "flos": 25331781813120.0,
      "grad_norm": 2.010670085338739,
      "language_loss": 0.77040446,
      "learning_rate": 1.600184482369449e-08,
      "loss": 0.79175961,
      "num_input_tokens_seen": 172498905,
      "step": 7991,
      "time_per_iteration": 2.822733163833618
    },
    {
      "auxiliary_loss_clip": 0.01123903,
      "auxiliary_loss_mlp": 0.01022965,
      "balance_loss_clip": 1.04003131,
      "balance_loss_mlp": 1.01544857,
      "epoch": 0.960981181987615,
      "flos": 21069114082560.0,
      "grad_norm": 3.0632010564472405,
      "language_loss": 0.88919371,
      "learning_rate": 1.5903653943362126e-08,
      "loss": 0.91066241,
      "num_input_tokens_seen": 172517900,
      "step": 7992,
      "time_per_iteration": 2.7203872203826904
    },
    {
      "auxiliary_loss_clip": 0.01137873,
      "auxiliary_loss_mlp": 0.01022159,
      "balance_loss_clip": 1.04282689,
      "balance_loss_mlp": 1.01568937,
      "epoch": 0.9611014248782541,
      "flos": 17823256554240.0,
      "grad_norm": 1.8198943403964072,
      "language_loss": 0.77125287,
      "learning_rate": 1.580576404886802e-08,
      "loss": 0.79285312,
      "num_input_tokens_seen": 172536430,
      "step": 7993,
      "time_per_iteration": 2.624990701675415
    },
    {
      "auxiliary_loss_clip": 0.01151069,
      "auxiliary_loss_mlp": 0.01025915,
      "balance_loss_clip": 1.04448247,
      "balance_loss_mlp": 1.01963902,
      "epoch": 0.9612216677688932,
      "flos": 19354631040000.0,
      "grad_norm": 2.0825244820862183,
      "language_loss": 0.79749072,
      "learning_rate": 1.570817515506162e-08,
      "loss": 0.8192606,
      "num_input_tokens_seen": 172555120,
      "step": 7994,
      "time_per_iteration": 2.6581501960754395
    },
    {
      "auxiliary_loss_clip": 0.01162467,
      "auxiliary_loss_mlp": 0.01023089,
      "balance_loss_clip": 1.04585433,
      "balance_loss_mlp": 1.01629972,
      "epoch": 0.9613419106595322,
      "flos": 15808739207040.0,
      "grad_norm": 1.8995620997196834,
      "language_loss": 0.8106066,
      "learning_rate": 1.561088727674753e-08,
      "loss": 0.83246219,
      "num_input_tokens_seen": 172569330,
      "step": 7995,
      "time_per_iteration": 2.5471107959747314
    },
    {
      "auxiliary_loss_clip": 0.01122576,
      "auxiliary_loss_mlp": 0.01027639,
      "balance_loss_clip": 1.04143262,
      "balance_loss_mlp": 1.01967645,
      "epoch": 0.9614621535501714,
      "flos": 25702488126720.0,
      "grad_norm": 2.320742748792102,
      "language_loss": 0.71302617,
      "learning_rate": 1.551390042868417e-08,
      "loss": 0.7345283,
      "num_input_tokens_seen": 172591100,
      "step": 7996,
      "time_per_iteration": 2.8054592609405518
    },
    {
      "auxiliary_loss_clip": 0.01151149,
      "auxiliary_loss_mlp": 0.01024995,
      "balance_loss_clip": 1.04468036,
      "balance_loss_mlp": 1.01840234,
      "epoch": 0.9615823964408104,
      "flos": 17819054663040.0,
      "grad_norm": 1.8328745006846003,
      "language_loss": 0.70718777,
      "learning_rate": 1.5417214625584207e-08,
      "loss": 0.72894925,
      "num_input_tokens_seen": 172608755,
      "step": 7997,
      "time_per_iteration": 2.6152281761169434
    },
    {
      "auxiliary_loss_clip": 0.01144638,
      "auxiliary_loss_mlp": 0.01022239,
      "balance_loss_clip": 1.04149127,
      "balance_loss_mlp": 1.01549506,
      "epoch": 0.9617026393314495,
      "flos": 20190020624640.0,
      "grad_norm": 2.0073348714557024,
      "language_loss": 0.85418004,
      "learning_rate": 1.5320829882114806e-08,
      "loss": 0.87584883,
      "num_input_tokens_seen": 172626830,
      "step": 7998,
      "time_per_iteration": 2.6629140377044678
    },
    {
      "auxiliary_loss_clip": 0.01163439,
      "auxiliary_loss_mlp": 0.01025701,
      "balance_loss_clip": 1.04311359,
      "balance_loss_mlp": 1.01879561,
      "epoch": 0.9618228822220887,
      "flos": 20267013427200.0,
      "grad_norm": 1.8163262547604877,
      "language_loss": 0.7921797,
      "learning_rate": 1.5224746212897378e-08,
      "loss": 0.81407112,
      "num_input_tokens_seen": 172646125,
      "step": 7999,
      "time_per_iteration": 2.5940558910369873
    },
    {
      "auxiliary_loss_clip": 0.01162942,
      "auxiliary_loss_mlp": 0.01023302,
      "balance_loss_clip": 1.04655933,
      "balance_loss_mlp": 1.01624799,
      "epoch": 0.9619431251127277,
      "flos": 21031300039680.0,
      "grad_norm": 1.8112563599140874,
      "language_loss": 0.77298546,
      "learning_rate": 1.512896363250804e-08,
      "loss": 0.79484791,
      "num_input_tokens_seen": 172666235,
      "step": 8000,
      "time_per_iteration": 2.582535743713379
    },
    {
      "auxiliary_loss_clip": 0.01150253,
      "auxiliary_loss_mlp": 0.01023099,
      "balance_loss_clip": 1.04194808,
      "balance_loss_mlp": 1.01643503,
      "epoch": 0.9620633680033668,
      "flos": 22382654538240.0,
      "grad_norm": 1.7768613702076068,
      "language_loss": 0.75595975,
      "learning_rate": 1.503348215547673e-08,
      "loss": 0.77769333,
      "num_input_tokens_seen": 172687325,
      "step": 8001,
      "time_per_iteration": 2.6494333744049072
    },
    {
      "auxiliary_loss_clip": 0.01134913,
      "auxiliary_loss_mlp": 0.01019864,
      "balance_loss_clip": 1.04212213,
      "balance_loss_mlp": 1.01342082,
      "epoch": 0.962183610894006,
      "flos": 18471730740480.0,
      "grad_norm": 2.3705892872713057,
      "language_loss": 0.80750132,
      "learning_rate": 1.4938301796288078e-08,
      "loss": 0.82904911,
      "num_input_tokens_seen": 172703895,
      "step": 8002,
      "time_per_iteration": 2.6072072982788086
    },
    {
      "auxiliary_loss_clip": 0.0116593,
      "auxiliary_loss_mlp": 0.01023276,
      "balance_loss_clip": 1.0471977,
      "balance_loss_mlp": 1.01585484,
      "epoch": 0.962303853784645,
      "flos": 18435245500800.0,
      "grad_norm": 3.541091960720743,
      "language_loss": 0.81591272,
      "learning_rate": 1.4843422569380537e-08,
      "loss": 0.83780479,
      "num_input_tokens_seen": 172720650,
      "step": 8003,
      "time_per_iteration": 2.623051166534424
    },
    {
      "auxiliary_loss_clip": 0.01110566,
      "auxiliary_loss_mlp": 0.01023065,
      "balance_loss_clip": 1.03925347,
      "balance_loss_mlp": 1.01616585,
      "epoch": 0.9624240966752841,
      "flos": 26391074826240.0,
      "grad_norm": 1.8349393763503818,
      "language_loss": 0.82727313,
      "learning_rate": 1.4748844489147483e-08,
      "loss": 0.84860939,
      "num_input_tokens_seen": 172737640,
      "step": 8004,
      "time_per_iteration": 2.7408039569854736
    },
    {
      "auxiliary_loss_clip": 0.0113419,
      "auxiliary_loss_mlp": 0.01023357,
      "balance_loss_clip": 1.03928566,
      "balance_loss_mlp": 1.01658916,
      "epoch": 0.9625443395659231,
      "flos": 14647675985280.0,
      "grad_norm": 2.537741466259607,
      "language_loss": 0.71183968,
      "learning_rate": 1.4654567569936326e-08,
      "loss": 0.73341513,
      "num_input_tokens_seen": 172755215,
      "step": 8005,
      "time_per_iteration": 2.657639265060425
    },
    {
      "auxiliary_loss_clip": 0.01104182,
      "auxiliary_loss_mlp": 0.01028322,
      "balance_loss_clip": 1.03813744,
      "balance_loss_mlp": 1.02147651,
      "epoch": 0.9626645824565623,
      "flos": 18367626147840.0,
      "grad_norm": 2.2334991137656113,
      "language_loss": 0.83170491,
      "learning_rate": 1.456059182604874e-08,
      "loss": 0.85302997,
      "num_input_tokens_seen": 172774020,
      "step": 8006,
      "time_per_iteration": 2.7265191078186035
    },
    {
      "auxiliary_loss_clip": 0.01166513,
      "auxiliary_loss_mlp": 0.01025856,
      "balance_loss_clip": 1.04636157,
      "balance_loss_mlp": 1.01877475,
      "epoch": 0.9627848253472013,
      "flos": 16580424021120.0,
      "grad_norm": 1.8722201936962148,
      "language_loss": 0.76470441,
      "learning_rate": 1.4466917271740653e-08,
      "loss": 0.78662813,
      "num_input_tokens_seen": 172792220,
      "step": 8007,
      "time_per_iteration": 2.5845415592193604
    },
    {
      "auxiliary_loss_clip": 0.01134037,
      "auxiliary_loss_mlp": 0.01031014,
      "balance_loss_clip": 1.04188704,
      "balance_loss_mlp": 1.02339053,
      "epoch": 0.9629050682378404,
      "flos": 20886867452160.0,
      "grad_norm": 3.8124522565359285,
      "language_loss": 0.67631537,
      "learning_rate": 1.4373543921222697e-08,
      "loss": 0.69796586,
      "num_input_tokens_seen": 172811805,
      "step": 8008,
      "time_per_iteration": 3.5595290660858154
    },
    {
      "auxiliary_loss_clip": 0.01133305,
      "auxiliary_loss_mlp": 0.01025608,
      "balance_loss_clip": 1.0423975,
      "balance_loss_mlp": 1.01801729,
      "epoch": 0.9630253111284796,
      "flos": 17019252478080.0,
      "grad_norm": 2.029463318105329,
      "language_loss": 0.78006124,
      "learning_rate": 1.428047178865932e-08,
      "loss": 0.80165029,
      "num_input_tokens_seen": 172828595,
      "step": 8009,
      "time_per_iteration": 2.6777825355529785
    },
    {
      "auxiliary_loss_clip": 0.01132382,
      "auxiliary_loss_mlp": 0.01027452,
      "balance_loss_clip": 1.03803062,
      "balance_loss_mlp": 1.02010322,
      "epoch": 0.9631455540191186,
      "flos": 20338942412160.0,
      "grad_norm": 1.6202304805167445,
      "language_loss": 0.74572963,
      "learning_rate": 1.4187700888169451e-08,
      "loss": 0.76732796,
      "num_input_tokens_seen": 172847770,
      "step": 8010,
      "time_per_iteration": 3.7080330848693848
    },
    {
      "auxiliary_loss_clip": 0.01051711,
      "auxiliary_loss_mlp": 0.01002403,
      "balance_loss_clip": 1.00887561,
      "balance_loss_mlp": 1.00140727,
      "epoch": 0.9632657969097577,
      "flos": 65956700033280.0,
      "grad_norm": 0.7525406947577326,
      "language_loss": 0.56996626,
      "learning_rate": 1.40952312338265e-08,
      "loss": 0.59050739,
      "num_input_tokens_seen": 172912415,
      "step": 8011,
      "time_per_iteration": 3.285555601119995
    },
    {
      "auxiliary_loss_clip": 0.01124039,
      "auxiliary_loss_mlp": 0.01025275,
      "balance_loss_clip": 1.03929019,
      "balance_loss_mlp": 1.01828313,
      "epoch": 0.9633860398003968,
      "flos": 44419523823360.0,
      "grad_norm": 1.7748460554219798,
      "language_loss": 0.68407941,
      "learning_rate": 1.4003062839657909e-08,
      "loss": 0.70557249,
      "num_input_tokens_seen": 172934895,
      "step": 8012,
      "time_per_iteration": 3.77223801612854
    },
    {
      "auxiliary_loss_clip": 0.01127843,
      "auxiliary_loss_mlp": 0.01023175,
      "balance_loss_clip": 1.04232192,
      "balance_loss_mlp": 1.01612651,
      "epoch": 0.9635062826910359,
      "flos": 24827704300800.0,
      "grad_norm": 1.6230378002555563,
      "language_loss": 0.80023992,
      "learning_rate": 1.391119571964583e-08,
      "loss": 0.8217501,
      "num_input_tokens_seen": 172955835,
      "step": 8013,
      "time_per_iteration": 2.7802484035491943
    },
    {
      "auxiliary_loss_clip": 0.01149562,
      "auxiliary_loss_mlp": 0.01022892,
      "balance_loss_clip": 1.04519701,
      "balance_loss_mlp": 1.01613033,
      "epoch": 0.9636265255816749,
      "flos": 15961360095360.0,
      "grad_norm": 1.8674047931979112,
      "language_loss": 0.73369789,
      "learning_rate": 1.3819629887726225e-08,
      "loss": 0.75542241,
      "num_input_tokens_seen": 172973925,
      "step": 8014,
      "time_per_iteration": 2.6151063442230225
    },
    {
      "auxiliary_loss_clip": 0.011444,
      "auxiliary_loss_mlp": 0.01025796,
      "balance_loss_clip": 1.04609561,
      "balance_loss_mlp": 1.01857746,
      "epoch": 0.9637467684723141,
      "flos": 22601781457920.0,
      "grad_norm": 1.744242200874373,
      "language_loss": 0.76575577,
      "learning_rate": 1.3728365357789317e-08,
      "loss": 0.78745776,
      "num_input_tokens_seen": 172993290,
      "step": 8015,
      "time_per_iteration": 2.6990599632263184
    },
    {
      "auxiliary_loss_clip": 0.01085635,
      "auxiliary_loss_mlp": 0.01024676,
      "balance_loss_clip": 1.03660893,
      "balance_loss_mlp": 1.01762509,
      "epoch": 0.9638670113629532,
      "flos": 17565812801280.0,
      "grad_norm": 2.3513432089124815,
      "language_loss": 0.76292872,
      "learning_rate": 1.3637402143680254e-08,
      "loss": 0.78403187,
      "num_input_tokens_seen": 173008190,
      "step": 8016,
      "time_per_iteration": 3.602175235748291
    },
    {
      "auxiliary_loss_clip": 0.0103186,
      "auxiliary_loss_mlp": 0.0100243,
      "balance_loss_clip": 1.0119369,
      "balance_loss_mlp": 1.00133896,
      "epoch": 0.9639872542535922,
      "flos": 55072139379840.0,
      "grad_norm": 0.7179412822169149,
      "language_loss": 0.55074131,
      "learning_rate": 1.3546740259197998e-08,
      "loss": 0.5710842,
      "num_input_tokens_seen": 173061000,
      "step": 8017,
      "time_per_iteration": 3.2145140171051025
    },
    {
      "auxiliary_loss_clip": 0.01139926,
      "auxiliary_loss_mlp": 0.01028891,
      "balance_loss_clip": 1.04389083,
      "balance_loss_mlp": 1.02112174,
      "epoch": 0.9641074971442314,
      "flos": 24134484746880.0,
      "grad_norm": 2.144932955734405,
      "language_loss": 0.70055526,
      "learning_rate": 1.3456379718095989e-08,
      "loss": 0.72224343,
      "num_input_tokens_seen": 173081415,
      "step": 8018,
      "time_per_iteration": 2.719129800796509
    },
    {
      "auxiliary_loss_clip": 0.01039668,
      "auxiliary_loss_mlp": 0.01001488,
      "balance_loss_clip": 1.00819325,
      "balance_loss_mlp": 1.00045133,
      "epoch": 0.9642277400348704,
      "flos": 66747416077440.0,
      "grad_norm": 0.8778966797230715,
      "language_loss": 0.62056285,
      "learning_rate": 1.3366320534081487e-08,
      "loss": 0.6409744,
      "num_input_tokens_seen": 173144095,
      "step": 8019,
      "time_per_iteration": 3.230581283569336
    },
    {
      "auxiliary_loss_clip": 0.01152143,
      "auxiliary_loss_mlp": 0.01026319,
      "balance_loss_clip": 1.04512084,
      "balance_loss_mlp": 1.01860332,
      "epoch": 0.9643479829255095,
      "flos": 30920272450560.0,
      "grad_norm": 2.178203666010136,
      "language_loss": 0.76044309,
      "learning_rate": 1.3276562720816675e-08,
      "loss": 0.78222764,
      "num_input_tokens_seen": 173165605,
      "step": 8020,
      "time_per_iteration": 2.712013006210327
    },
    {
      "auxiliary_loss_clip": 0.01166438,
      "auxiliary_loss_mlp": 0.01022743,
      "balance_loss_clip": 1.04507923,
      "balance_loss_mlp": 1.01584387,
      "epoch": 0.9644682258161487,
      "flos": 20048245643520.0,
      "grad_norm": 2.4042986759231804,
      "language_loss": 0.8251158,
      "learning_rate": 1.3187106291917549e-08,
      "loss": 0.84700763,
      "num_input_tokens_seen": 173182595,
      "step": 8021,
      "time_per_iteration": 2.534327745437622
    },
    {
      "auxiliary_loss_clip": 0.01146227,
      "auxiliary_loss_mlp": 0.01024613,
      "balance_loss_clip": 1.04333234,
      "balance_loss_mlp": 1.01730514,
      "epoch": 0.9645884687067877,
      "flos": 21178713456000.0,
      "grad_norm": 1.7293010013559558,
      "language_loss": 0.70460147,
      "learning_rate": 1.309795126095503e-08,
      "loss": 0.7263099,
      "num_input_tokens_seen": 173200895,
      "step": 8022,
      "time_per_iteration": 2.629523992538452
    },
    {
      "auxiliary_loss_clip": 0.01074853,
      "auxiliary_loss_mlp": 0.01023066,
      "balance_loss_clip": 1.0337249,
      "balance_loss_mlp": 1.01625907,
      "epoch": 0.9647087115974268,
      "flos": 18945967029120.0,
      "grad_norm": 2.301512910725068,
      "language_loss": 0.80391186,
      "learning_rate": 1.3009097641453192e-08,
      "loss": 0.82489103,
      "num_input_tokens_seen": 173218745,
      "step": 8023,
      "time_per_iteration": 2.7175095081329346
    },
    {
      "auxiliary_loss_clip": 0.01134424,
      "auxiliary_loss_mlp": 0.01023735,
      "balance_loss_clip": 1.04222989,
      "balance_loss_mlp": 1.01633859,
      "epoch": 0.9648289544880659,
      "flos": 16545088016640.0,
      "grad_norm": 1.712894824164886,
      "language_loss": 0.76109838,
      "learning_rate": 1.2920545446891474e-08,
      "loss": 0.78267992,
      "num_input_tokens_seen": 173235465,
      "step": 8024,
      "time_per_iteration": 2.615529775619507
    },
    {
      "auxiliary_loss_clip": 0.01141498,
      "auxiliary_loss_mlp": 0.01024604,
      "balance_loss_clip": 1.04517996,
      "balance_loss_mlp": 1.01750791,
      "epoch": 0.964949197378705,
      "flos": 24057527857920.0,
      "grad_norm": 3.387999104897225,
      "language_loss": 0.70714265,
      "learning_rate": 1.2832294690703127e-08,
      "loss": 0.72880363,
      "num_input_tokens_seen": 173254440,
      "step": 8025,
      "time_per_iteration": 2.7022433280944824
    },
    {
      "auxiliary_loss_clip": 0.01150343,
      "auxiliary_loss_mlp": 0.01023983,
      "balance_loss_clip": 1.04480743,
      "balance_loss_mlp": 1.01710176,
      "epoch": 0.965069440269344,
      "flos": 23365565280000.0,
      "grad_norm": 1.8458012214942505,
      "language_loss": 0.77593189,
      "learning_rate": 1.2744345386275668e-08,
      "loss": 0.79767513,
      "num_input_tokens_seen": 173273980,
      "step": 8026,
      "time_per_iteration": 2.641173839569092
    },
    {
      "auxiliary_loss_clip": 0.01142862,
      "auxiliary_loss_mlp": 0.01028265,
      "balance_loss_clip": 1.0450635,
      "balance_loss_mlp": 1.02149343,
      "epoch": 0.9651896831599832,
      "flos": 25374875155200.0,
      "grad_norm": 1.6573422814019758,
      "language_loss": 0.78793257,
      "learning_rate": 1.265669754695109e-08,
      "loss": 0.80964386,
      "num_input_tokens_seen": 173293550,
      "step": 8027,
      "time_per_iteration": 2.6933484077453613
    },
    {
      "auxiliary_loss_clip": 0.01096924,
      "auxiliary_loss_mlp": 0.01025169,
      "balance_loss_clip": 1.03707695,
      "balance_loss_mlp": 1.01757264,
      "epoch": 0.9653099260506223,
      "flos": 22272875596800.0,
      "grad_norm": 1.8839111984153647,
      "language_loss": 0.82247138,
      "learning_rate": 1.2569351186025201e-08,
      "loss": 0.8436923,
      "num_input_tokens_seen": 173312005,
      "step": 8028,
      "time_per_iteration": 2.750873565673828
    },
    {
      "auxiliary_loss_clip": 0.01111866,
      "auxiliary_loss_mlp": 0.01021946,
      "balance_loss_clip": 1.03860319,
      "balance_loss_mlp": 1.01517487,
      "epoch": 0.9654301689412613,
      "flos": 26760847386240.0,
      "grad_norm": 1.5453276461507903,
      "language_loss": 0.75561267,
      "learning_rate": 1.2482306316748737e-08,
      "loss": 0.77695078,
      "num_input_tokens_seen": 173332450,
      "step": 8029,
      "time_per_iteration": 2.7223312854766846
    },
    {
      "auxiliary_loss_clip": 0.01155653,
      "auxiliary_loss_mlp": 0.01027291,
      "balance_loss_clip": 1.04321551,
      "balance_loss_mlp": 1.0198437,
      "epoch": 0.9655504118319005,
      "flos": 17412689122560.0,
      "grad_norm": 2.1070087389657797,
      "language_loss": 0.78429651,
      "learning_rate": 1.2395562952326021e-08,
      "loss": 0.80612594,
      "num_input_tokens_seen": 173349610,
      "step": 8030,
      "time_per_iteration": 2.6083874702453613
    },
    {
      "auxiliary_loss_clip": 0.01146624,
      "auxiliary_loss_mlp": 0.01033325,
      "balance_loss_clip": 1.04480958,
      "balance_loss_mlp": 1.02585924,
      "epoch": 0.9656706547225395,
      "flos": 22126970551680.0,
      "grad_norm": 2.2981038559048264,
      "language_loss": 0.80952996,
      "learning_rate": 1.2309121105916309e-08,
      "loss": 0.83132946,
      "num_input_tokens_seen": 173367900,
      "step": 8031,
      "time_per_iteration": 2.643375873565674
    },
    {
      "auxiliary_loss_clip": 0.01155251,
      "auxiliary_loss_mlp": 0.0102299,
      "balance_loss_clip": 1.04437721,
      "balance_loss_mlp": 1.01566756,
      "epoch": 0.9657908976131786,
      "flos": 37049289926400.0,
      "grad_norm": 2.024523766303704,
      "language_loss": 0.69332588,
      "learning_rate": 1.222298079063222e-08,
      "loss": 0.71510828,
      "num_input_tokens_seen": 173389040,
      "step": 8032,
      "time_per_iteration": 2.739582061767578
    },
    {
      "auxiliary_loss_clip": 0.01149568,
      "auxiliary_loss_mlp": 0.01022274,
      "balance_loss_clip": 1.04343045,
      "balance_loss_mlp": 1.01569998,
      "epoch": 0.9659111405038178,
      "flos": 24389809597440.0,
      "grad_norm": 1.9654888413784923,
      "language_loss": 0.72321904,
      "learning_rate": 1.2137142019541524e-08,
      "loss": 0.74493742,
      "num_input_tokens_seen": 173407595,
      "step": 8033,
      "time_per_iteration": 2.655346393585205
    },
    {
      "auxiliary_loss_clip": 0.01144101,
      "auxiliary_loss_mlp": 0.01028207,
      "balance_loss_clip": 1.04329085,
      "balance_loss_mlp": 1.02064884,
      "epoch": 0.9660313833944568,
      "flos": 25009412227200.0,
      "grad_norm": 3.7120683540943014,
      "language_loss": 0.736426,
      "learning_rate": 1.2051604805666027e-08,
      "loss": 0.75814909,
      "num_input_tokens_seen": 173424720,
      "step": 8034,
      "time_per_iteration": 3.6424593925476074
    },
    {
      "auxiliary_loss_clip": 0.01164754,
      "auxiliary_loss_mlp": 0.00762144,
      "balance_loss_clip": 1.04626024,
      "balance_loss_mlp": 1.00036824,
      "epoch": 0.9661516262850959,
      "flos": 11801575895040.0,
      "grad_norm": 2.7568415993569806,
      "language_loss": 0.78808224,
      "learning_rate": 1.196636916198135e-08,
      "loss": 0.80735117,
      "num_input_tokens_seen": 173442260,
      "step": 8035,
      "time_per_iteration": 2.558863878250122
    },
    {
      "auxiliary_loss_clip": 0.01165708,
      "auxiliary_loss_mlp": 0.01029968,
      "balance_loss_clip": 1.04463935,
      "balance_loss_mlp": 1.02240419,
      "epoch": 0.9662718691757349,
      "flos": 20047778766720.0,
      "grad_norm": 2.824265288219802,
      "language_loss": 0.7698856,
      "learning_rate": 1.1881435101418036e-08,
      "loss": 0.79184234,
      "num_input_tokens_seen": 173461675,
      "step": 8036,
      "time_per_iteration": 2.5809054374694824
    },
    {
      "auxiliary_loss_clip": 0.01042572,
      "auxiliary_loss_mlp": 0.0100124,
      "balance_loss_clip": 1.01007366,
      "balance_loss_mlp": 1.00023305,
      "epoch": 0.9663921120663741,
      "flos": 68027703517440.0,
      "grad_norm": 0.747614383187982,
      "language_loss": 0.65544033,
      "learning_rate": 1.1796802636860003e-08,
      "loss": 0.67587852,
      "num_input_tokens_seen": 173530205,
      "step": 8037,
      "time_per_iteration": 4.173290729522705
    },
    {
      "auxiliary_loss_clip": 0.01165896,
      "auxiliary_loss_mlp": 0.01018705,
      "balance_loss_clip": 1.04504919,
      "balance_loss_mlp": 1.01128745,
      "epoch": 0.9665123549570132,
      "flos": 26322916769280.0,
      "grad_norm": 1.9495079038882606,
      "language_loss": 0.74035221,
      "learning_rate": 1.1712471781146316e-08,
      "loss": 0.76219821,
      "num_input_tokens_seen": 173549540,
      "step": 8038,
      "time_per_iteration": 2.650090456008911
    },
    {
      "auxiliary_loss_clip": 0.01164449,
      "auxiliary_loss_mlp": 0.01021737,
      "balance_loss_clip": 1.04409933,
      "balance_loss_mlp": 1.01444447,
      "epoch": 0.9666325978476522,
      "flos": 43941121557120.0,
      "grad_norm": 1.9014546939053292,
      "language_loss": 0.66509074,
      "learning_rate": 1.1628442547069628e-08,
      "loss": 0.68695265,
      "num_input_tokens_seen": 173571740,
      "step": 8039,
      "time_per_iteration": 3.642157793045044
    },
    {
      "auxiliary_loss_clip": 0.01150941,
      "auxiliary_loss_mlp": 0.00762365,
      "balance_loss_clip": 1.04120684,
      "balance_loss_mlp": 1.00036812,
      "epoch": 0.9667528407382914,
      "flos": 21543422198400.0,
      "grad_norm": 1.9417482916881617,
      "language_loss": 0.77014983,
      "learning_rate": 1.1544714947377521e-08,
      "loss": 0.78928292,
      "num_input_tokens_seen": 173589425,
      "step": 8040,
      "time_per_iteration": 2.611628293991089
    },
    {
      "auxiliary_loss_clip": 0.01168671,
      "auxiliary_loss_mlp": 0.01023314,
      "balance_loss_clip": 1.04797626,
      "balance_loss_mlp": 1.01590514,
      "epoch": 0.9668730836289304,
      "flos": 23878585278720.0,
      "grad_norm": 2.342536777581491,
      "language_loss": 0.69975334,
      "learning_rate": 1.1461288994770945e-08,
      "loss": 0.72167319,
      "num_input_tokens_seen": 173608500,
      "step": 8041,
      "time_per_iteration": 2.6151280403137207
    },
    {
      "auxiliary_loss_clip": 0.01168572,
      "auxiliary_loss_mlp": 0.01029533,
      "balance_loss_clip": 1.04482222,
      "balance_loss_mlp": 1.02191567,
      "epoch": 0.9669933265195695,
      "flos": 28293011971200.0,
      "grad_norm": 1.980224275239243,
      "language_loss": 0.76884383,
      "learning_rate": 1.1378164701906002e-08,
      "loss": 0.79082489,
      "num_input_tokens_seen": 173630265,
      "step": 8042,
      "time_per_iteration": 3.563913583755493
    },
    {
      "auxiliary_loss_clip": 0.01168026,
      "auxiliary_loss_mlp": 0.01024696,
      "balance_loss_clip": 1.04676938,
      "balance_loss_mlp": 1.01722193,
      "epoch": 0.9671135694102087,
      "flos": 22454763091200.0,
      "grad_norm": 1.853249275628975,
      "language_loss": 0.66881925,
      "learning_rate": 1.1295342081392156e-08,
      "loss": 0.69074649,
      "num_input_tokens_seen": 173649625,
      "step": 8043,
      "time_per_iteration": 2.7985219955444336
    },
    {
      "auxiliary_loss_clip": 0.01137222,
      "auxiliary_loss_mlp": 0.01021183,
      "balance_loss_clip": 1.04185438,
      "balance_loss_mlp": 1.01472461,
      "epoch": 0.9672338123008477,
      "flos": 20155941596160.0,
      "grad_norm": 1.876236947266394,
      "language_loss": 0.69293773,
      "learning_rate": 1.1212821145793804e-08,
      "loss": 0.71452188,
      "num_input_tokens_seen": 173669240,
      "step": 8044,
      "time_per_iteration": 2.681973695755005
    },
    {
      "auxiliary_loss_clip": 0.0113714,
      "auxiliary_loss_mlp": 0.01022271,
      "balance_loss_clip": 1.04132724,
      "balance_loss_mlp": 1.01526737,
      "epoch": 0.9673540551914868,
      "flos": 16977487939200.0,
      "grad_norm": 1.917398420981964,
      "language_loss": 0.78900409,
      "learning_rate": 1.1130601907629156e-08,
      "loss": 0.81059813,
      "num_input_tokens_seen": 173686970,
      "step": 8045,
      "time_per_iteration": 2.5916202068328857
    },
    {
      "auxiliary_loss_clip": 0.01055026,
      "auxiliary_loss_mlp": 0.01000816,
      "balance_loss_clip": 1.00936127,
      "balance_loss_mlp": 0.99985677,
      "epoch": 0.9674742980821259,
      "flos": 61892903952000.0,
      "grad_norm": 0.8166778718475749,
      "language_loss": 0.64710641,
      "learning_rate": 1.1048684379370899e-08,
      "loss": 0.66766489,
      "num_input_tokens_seen": 173747655,
      "step": 8046,
      "time_per_iteration": 3.1994142532348633
    },
    {
      "auxiliary_loss_clip": 0.01126875,
      "auxiliary_loss_mlp": 0.01021223,
      "balance_loss_clip": 1.04233408,
      "balance_loss_mlp": 1.01466084,
      "epoch": 0.967594540972765,
      "flos": 18697824898560.0,
      "grad_norm": 2.0301954946402736,
      "language_loss": 0.74330336,
      "learning_rate": 1.0967068573445759e-08,
      "loss": 0.76478434,
      "num_input_tokens_seen": 173765140,
      "step": 8047,
      "time_per_iteration": 2.628218650817871
    },
    {
      "auxiliary_loss_clip": 0.01132321,
      "auxiliary_loss_mlp": 0.01027288,
      "balance_loss_clip": 1.04005826,
      "balance_loss_mlp": 1.01992655,
      "epoch": 0.967714783863404,
      "flos": 20777411733120.0,
      "grad_norm": 2.0974312225040124,
      "language_loss": 0.6514858,
      "learning_rate": 1.0885754502234945e-08,
      "loss": 0.67308187,
      "num_input_tokens_seen": 173784800,
      "step": 8048,
      "time_per_iteration": 2.660459041595459
    },
    {
      "auxiliary_loss_clip": 0.01122027,
      "auxiliary_loss_mlp": 0.01023819,
      "balance_loss_clip": 1.04201961,
      "balance_loss_mlp": 1.01660681,
      "epoch": 0.9678350267540432,
      "flos": 23185473465600.0,
      "grad_norm": 1.901181415538082,
      "language_loss": 0.78248084,
      "learning_rate": 1.08047421780737e-08,
      "loss": 0.80393922,
      "num_input_tokens_seen": 173803990,
      "step": 8049,
      "time_per_iteration": 2.6757004261016846
    },
    {
      "auxiliary_loss_clip": 0.0114076,
      "auxiliary_loss_mlp": 0.00761945,
      "balance_loss_clip": 1.04114389,
      "balance_loss_mlp": 1.00028479,
      "epoch": 0.9679552696446823,
      "flos": 21726063878400.0,
      "grad_norm": 2.2518972148924004,
      "language_loss": 0.74306607,
      "learning_rate": 1.0724031613251305e-08,
      "loss": 0.76209307,
      "num_input_tokens_seen": 173821890,
      "step": 8050,
      "time_per_iteration": 2.6551778316497803
    },
    {
      "auxiliary_loss_clip": 0.01156137,
      "auxiliary_loss_mlp": 0.01019181,
      "balance_loss_clip": 1.04417276,
      "balance_loss_mlp": 1.01191258,
      "epoch": 0.9680755125353213,
      "flos": 26869046129280.0,
      "grad_norm": 3.9160331405110878,
      "language_loss": 0.66761631,
      "learning_rate": 1.0643622820011744e-08,
      "loss": 0.68936944,
      "num_input_tokens_seen": 173842945,
      "step": 8051,
      "time_per_iteration": 2.643146276473999
    },
    {
      "auxiliary_loss_clip": 0.01167249,
      "auxiliary_loss_mlp": 0.01029486,
      "balance_loss_clip": 1.04555094,
      "balance_loss_mlp": 1.02150214,
      "epoch": 0.9681957554259605,
      "flos": 28325008010880.0,
      "grad_norm": 2.4415295607167544,
      "language_loss": 0.68050194,
      "learning_rate": 1.0563515810552814e-08,
      "loss": 0.70246923,
      "num_input_tokens_seen": 173859915,
      "step": 8052,
      "time_per_iteration": 2.6023592948913574
    },
    {
      "auxiliary_loss_clip": 0.01167081,
      "auxiliary_loss_mlp": 0.01031576,
      "balance_loss_clip": 1.04814434,
      "balance_loss_mlp": 1.02442932,
      "epoch": 0.9683159983165995,
      "flos": 20557674282240.0,
      "grad_norm": 1.5539629772815964,
      "language_loss": 0.73402488,
      "learning_rate": 1.0483710597026795e-08,
      "loss": 0.75601143,
      "num_input_tokens_seen": 173879775,
      "step": 8053,
      "time_per_iteration": 2.6072306632995605
    },
    {
      "auxiliary_loss_clip": 0.01125291,
      "auxiliary_loss_mlp": 0.0102333,
      "balance_loss_clip": 1.0417186,
      "balance_loss_mlp": 1.01717281,
      "epoch": 0.9684362412072386,
      "flos": 24207958016640.0,
      "grad_norm": 2.2027821143003576,
      "language_loss": 0.74048567,
      "learning_rate": 1.0404207191540227e-08,
      "loss": 0.76197189,
      "num_input_tokens_seen": 173900230,
      "step": 8054,
      "time_per_iteration": 2.7009434700012207
    },
    {
      "auxiliary_loss_clip": 0.01164125,
      "auxiliary_loss_mlp": 0.01024452,
      "balance_loss_clip": 1.04504132,
      "balance_loss_mlp": 1.01729107,
      "epoch": 0.9685564840978778,
      "flos": 22346241125760.0,
      "grad_norm": 2.1293004183647515,
      "language_loss": 0.74416661,
      "learning_rate": 1.0325005606153236e-08,
      "loss": 0.76605231,
      "num_input_tokens_seen": 173919690,
      "step": 8055,
      "time_per_iteration": 2.5539891719818115
    },
    {
      "auxiliary_loss_clip": 0.01112065,
      "auxiliary_loss_mlp": 0.01023856,
      "balance_loss_clip": 1.03888249,
      "balance_loss_mlp": 1.01623821,
      "epoch": 0.9686767269885168,
      "flos": 14386389477120.0,
      "grad_norm": 2.681923205374886,
      "language_loss": 0.79038817,
      "learning_rate": 1.0246105852881104e-08,
      "loss": 0.81174743,
      "num_input_tokens_seen": 173934790,
      "step": 8056,
      "time_per_iteration": 2.6863715648651123
    },
    {
      "auxiliary_loss_clip": 0.0116606,
      "auxiliary_loss_mlp": 0.01028171,
      "balance_loss_clip": 1.04550123,
      "balance_loss_mlp": 1.02061307,
      "epoch": 0.9687969698791559,
      "flos": 21287630471040.0,
      "grad_norm": 2.511242163612255,
      "language_loss": 0.78676265,
      "learning_rate": 1.0167507943692476e-08,
      "loss": 0.80870497,
      "num_input_tokens_seen": 173953875,
      "step": 8057,
      "time_per_iteration": 2.6977202892303467
    },
    {
      "auxiliary_loss_clip": 0.01151043,
      "auxiliary_loss_mlp": 0.01025636,
      "balance_loss_clip": 1.04656041,
      "balance_loss_mlp": 1.01837635,
      "epoch": 0.968917212769795,
      "flos": 19828328624640.0,
      "grad_norm": 2.068299777041304,
      "language_loss": 0.72007143,
      "learning_rate": 1.008921189051093e-08,
      "loss": 0.74183816,
      "num_input_tokens_seen": 173971220,
      "step": 8058,
      "time_per_iteration": 2.5898773670196533
    },
    {
      "auxiliary_loss_clip": 0.01167842,
      "auxiliary_loss_mlp": 0.01027703,
      "balance_loss_clip": 1.0472815,
      "balance_loss_mlp": 1.02033901,
      "epoch": 0.9690374556604341,
      "flos": 21681749473920.0,
      "grad_norm": 1.946884588445897,
      "language_loss": 0.77144003,
      "learning_rate": 1.0011217705213848e-08,
      "loss": 0.79339552,
      "num_input_tokens_seen": 173989095,
      "step": 8059,
      "time_per_iteration": 2.594555616378784
    },
    {
      "auxiliary_loss_clip": 0.01147468,
      "auxiliary_loss_mlp": 0.01019213,
      "balance_loss_clip": 1.04383397,
      "balance_loss_mlp": 1.01283264,
      "epoch": 0.9691576985510731,
      "flos": 32635437851520.0,
      "grad_norm": 1.8589786629043585,
      "language_loss": 0.74193037,
      "learning_rate": 9.933525399632658e-09,
      "loss": 0.76359719,
      "num_input_tokens_seen": 174007330,
      "step": 8060,
      "time_per_iteration": 0.9517033100128174
    },
    {
      "auxiliary_loss_clip": 0.01134807,
      "auxiliary_loss_mlp": 0.01026838,
      "balance_loss_clip": 1.04179978,
      "balance_loss_mlp": 1.01958132,
      "epoch": 0.9692779414417123,
      "flos": 35663174040960.0,
      "grad_norm": 1.8035651014514542,
      "language_loss": 0.64826035,
      "learning_rate": 9.856134985553488e-09,
      "loss": 0.66987681,
      "num_input_tokens_seen": 174027055,
      "step": 8061,
      "time_per_iteration": 2.7452144622802734
    },
    {
      "auxiliary_loss_clip": 0.01161965,
      "auxiliary_loss_mlp": 0.01027783,
      "balance_loss_clip": 1.04362106,
      "balance_loss_mlp": 1.02033782,
      "epoch": 0.9693981843323514,
      "flos": 28366952117760.0,
      "grad_norm": 1.7379530493341169,
      "language_loss": 0.73817348,
      "learning_rate": 9.77904647471628e-09,
      "loss": 0.76007092,
      "num_input_tokens_seen": 174050235,
      "step": 8062,
      "time_per_iteration": 2.6009695529937744
    },
    {
      "auxiliary_loss_clip": 0.01100117,
      "auxiliary_loss_mlp": 0.01025196,
      "balance_loss_clip": 1.03950405,
      "balance_loss_mlp": 1.0180757,
      "epoch": 0.9695184272229904,
      "flos": 23622865378560.0,
      "grad_norm": 1.518047285018217,
      "language_loss": 0.73866403,
      "learning_rate": 9.702259878815454e-09,
      "loss": 0.7599172,
      "num_input_tokens_seen": 174070560,
      "step": 8063,
      "time_per_iteration": 3.6935014724731445
    },
    {
      "auxiliary_loss_clip": 0.01156386,
      "auxiliary_loss_mlp": 0.01024544,
      "balance_loss_clip": 1.04684949,
      "balance_loss_mlp": 1.01728106,
      "epoch": 0.9696386701136296,
      "flos": 23294677789440.0,
      "grad_norm": 2.072325390451814,
      "language_loss": 0.74065,
      "learning_rate": 9.625775209499254e-09,
      "loss": 0.76245928,
      "num_input_tokens_seen": 174090565,
      "step": 8064,
      "time_per_iteration": 2.6438655853271484
    },
    {
      "auxiliary_loss_clip": 0.01117046,
      "auxiliary_loss_mlp": 0.01025518,
      "balance_loss_clip": 1.03953385,
      "balance_loss_mlp": 1.01893473,
      "epoch": 0.9697589130042686,
      "flos": 15121876360320.0,
      "grad_norm": 4.035470427254621,
      "language_loss": 0.7421124,
      "learning_rate": 9.549592478370172e-09,
      "loss": 0.763538,
      "num_input_tokens_seen": 174108745,
      "step": 8065,
      "time_per_iteration": 3.4915177822113037
    },
    {
      "auxiliary_loss_clip": 0.01154328,
      "auxiliary_loss_mlp": 0.01021917,
      "balance_loss_clip": 1.04498792,
      "balance_loss_mlp": 1.01486278,
      "epoch": 0.9698791558949077,
      "flos": 18879532824960.0,
      "grad_norm": 1.6908795780244605,
      "language_loss": 0.79451942,
      "learning_rate": 9.473711696985632e-09,
      "loss": 0.81628191,
      "num_input_tokens_seen": 174128075,
      "step": 8066,
      "time_per_iteration": 2.626004695892334
    },
    {
      "auxiliary_loss_clip": 0.01135748,
      "auxiliary_loss_mlp": 0.0102144,
      "balance_loss_clip": 1.0406878,
      "balance_loss_mlp": 1.01461554,
      "epoch": 0.9699993987855468,
      "flos": 17931455297280.0,
      "grad_norm": 2.3330391384843367,
      "language_loss": 0.75779998,
      "learning_rate": 9.398132876856201e-09,
      "loss": 0.77937192,
      "num_input_tokens_seen": 174147040,
      "step": 8067,
      "time_per_iteration": 2.624974012374878
    },
    {
      "auxiliary_loss_clip": 0.01023769,
      "auxiliary_loss_mlp": 0.0100252,
      "balance_loss_clip": 1.01081932,
      "balance_loss_mlp": 1.00144708,
      "epoch": 0.9701196416761859,
      "flos": 67182186297600.0,
      "grad_norm": 0.7753849550312911,
      "language_loss": 0.60755581,
      "learning_rate": 9.322856029447379e-09,
      "loss": 0.6278187,
      "num_input_tokens_seen": 174208225,
      "step": 8068,
      "time_per_iteration": 3.977970600128174
    },
    {
      "auxiliary_loss_clip": 0.01162434,
      "auxiliary_loss_mlp": 0.01024968,
      "balance_loss_clip": 1.04497433,
      "balance_loss_mlp": 1.01832843,
      "epoch": 0.970239884566825,
      "flos": 24277804012800.0,
      "grad_norm": 1.8743361939379348,
      "language_loss": 0.80321062,
      "learning_rate": 9.247881166178695e-09,
      "loss": 0.82508463,
      "num_input_tokens_seen": 174226935,
      "step": 8069,
      "time_per_iteration": 2.611100196838379
    },
    {
      "auxiliary_loss_clip": 0.01129415,
      "auxiliary_loss_mlp": 0.01022943,
      "balance_loss_clip": 1.04121614,
      "balance_loss_mlp": 1.01584375,
      "epoch": 0.970360127457464,
      "flos": 25301689194240.0,
      "grad_norm": 2.5902418377638488,
      "language_loss": 0.76421428,
      "learning_rate": 9.173208298423274e-09,
      "loss": 0.78573781,
      "num_input_tokens_seen": 174248140,
      "step": 8070,
      "time_per_iteration": 2.7611727714538574
    },
    {
      "auxiliary_loss_clip": 0.01106936,
      "auxiliary_loss_mlp": 0.00762092,
      "balance_loss_clip": 1.03950083,
      "balance_loss_mlp": 1.00031495,
      "epoch": 0.9704803703481032,
      "flos": 29572473398400.0,
      "grad_norm": 1.5806576840008906,
      "language_loss": 0.76121247,
      "learning_rate": 9.09883743750961e-09,
      "loss": 0.77990276,
      "num_input_tokens_seen": 174271030,
      "step": 8071,
      "time_per_iteration": 2.784320116043091
    },
    {
      "auxiliary_loss_clip": 0.01133536,
      "auxiliary_loss_mlp": 0.01020575,
      "balance_loss_clip": 1.04212511,
      "balance_loss_mlp": 1.01413465,
      "epoch": 0.9706006132387422,
      "flos": 17380046638080.0,
      "grad_norm": 1.7322376282775436,
      "language_loss": 0.83851153,
      "learning_rate": 9.024768594719124e-09,
      "loss": 0.86005265,
      "num_input_tokens_seen": 174289410,
      "step": 8072,
      "time_per_iteration": 2.628959894180298
    },
    {
      "auxiliary_loss_clip": 0.01123742,
      "auxiliary_loss_mlp": 0.01024496,
      "balance_loss_clip": 1.04281449,
      "balance_loss_mlp": 1.01736414,
      "epoch": 0.9707208561293813,
      "flos": 18186421011840.0,
      "grad_norm": 2.769440138187765,
      "language_loss": 0.72955739,
      "learning_rate": 8.95100178128816e-09,
      "loss": 0.75103974,
      "num_input_tokens_seen": 174308550,
      "step": 8073,
      "time_per_iteration": 2.676483392715454
    },
    {
      "auxiliary_loss_clip": 0.01140128,
      "auxiliary_loss_mlp": 0.0102507,
      "balance_loss_clip": 1.04468429,
      "balance_loss_mlp": 1.01740444,
      "epoch": 0.9708410990200205,
      "flos": 31248388212480.0,
      "grad_norm": 1.7436440375908924,
      "language_loss": 0.70424473,
      "learning_rate": 8.877537008407321e-09,
      "loss": 0.72589678,
      "num_input_tokens_seen": 174328600,
      "step": 8074,
      "time_per_iteration": 2.7069835662841797
    },
    {
      "auxiliary_loss_clip": 0.01139619,
      "auxiliary_loss_mlp": 0.01023038,
      "balance_loss_clip": 1.04338396,
      "balance_loss_mlp": 1.01607668,
      "epoch": 0.9709613419106595,
      "flos": 30554450386560.0,
      "grad_norm": 1.5595820856186635,
      "language_loss": 0.6845082,
      "learning_rate": 8.804374287221028e-09,
      "loss": 0.7061348,
      "num_input_tokens_seen": 174349835,
      "step": 8075,
      "time_per_iteration": 2.7215447425842285
    },
    {
      "auxiliary_loss_clip": 0.01113158,
      "auxiliary_loss_mlp": 0.01025464,
      "balance_loss_clip": 1.03433931,
      "balance_loss_mlp": 1.01825476,
      "epoch": 0.9710815848012986,
      "flos": 23730166281600.0,
      "grad_norm": 1.644648868879505,
      "language_loss": 0.84721917,
      "learning_rate": 8.731513628827958e-09,
      "loss": 0.86860543,
      "num_input_tokens_seen": 174369200,
      "step": 8076,
      "time_per_iteration": 2.6982333660125732
    },
    {
      "auxiliary_loss_clip": 0.01153336,
      "auxiliary_loss_mlp": 0.01022851,
      "balance_loss_clip": 1.04464388,
      "balance_loss_mlp": 1.01573992,
      "epoch": 0.9712018276919377,
      "flos": 23761875012480.0,
      "grad_norm": 2.578539503867079,
      "language_loss": 0.82285023,
      "learning_rate": 8.658955044280825e-09,
      "loss": 0.84461212,
      "num_input_tokens_seen": 174388125,
      "step": 8077,
      "time_per_iteration": 2.667250394821167
    },
    {
      "auxiliary_loss_clip": 0.01147591,
      "auxiliary_loss_mlp": 0.01022366,
      "balance_loss_clip": 1.04411829,
      "balance_loss_mlp": 1.01545751,
      "epoch": 0.9713220705825768,
      "flos": 23330983461120.0,
      "grad_norm": 1.5167362902029875,
      "language_loss": 0.77726847,
      "learning_rate": 8.586698544587268e-09,
      "loss": 0.79896802,
      "num_input_tokens_seen": 174409735,
      "step": 8078,
      "time_per_iteration": 2.6477038860321045
    },
    {
      "auxiliary_loss_clip": 0.01127711,
      "auxiliary_loss_mlp": 0.01029616,
      "balance_loss_clip": 1.04001045,
      "balance_loss_mlp": 1.02289581,
      "epoch": 0.9714423134732159,
      "flos": 22200946611840.0,
      "grad_norm": 2.201773842496642,
      "language_loss": 0.74260747,
      "learning_rate": 8.514744140707853e-09,
      "loss": 0.76418078,
      "num_input_tokens_seen": 174428875,
      "step": 8079,
      "time_per_iteration": 2.6375560760498047
    },
    {
      "auxiliary_loss_clip": 0.01161452,
      "auxiliary_loss_mlp": 0.01025989,
      "balance_loss_clip": 1.04412818,
      "balance_loss_mlp": 1.01938152,
      "epoch": 0.971562556363855,
      "flos": 20229917656320.0,
      "grad_norm": 1.5899492610616683,
      "language_loss": 0.76684022,
      "learning_rate": 8.443091843558515e-09,
      "loss": 0.78871465,
      "num_input_tokens_seen": 174447960,
      "step": 8080,
      "time_per_iteration": 2.5895440578460693
    },
    {
      "auxiliary_loss_clip": 0.0113192,
      "auxiliary_loss_mlp": 0.01028408,
      "balance_loss_clip": 1.04189384,
      "balance_loss_mlp": 1.02061749,
      "epoch": 0.9716827992544941,
      "flos": 24970197553920.0,
      "grad_norm": 2.0514231485775123,
      "language_loss": 0.64811784,
      "learning_rate": 8.37174166400878e-09,
      "loss": 0.66972113,
      "num_input_tokens_seen": 174463535,
      "step": 8081,
      "time_per_iteration": 2.727454423904419
    },
    {
      "auxiliary_loss_clip": 0.01167582,
      "auxiliary_loss_mlp": 0.01025824,
      "balance_loss_clip": 1.04913151,
      "balance_loss_mlp": 1.01872802,
      "epoch": 0.9718030421451331,
      "flos": 24681476033280.0,
      "grad_norm": 4.381018237474725,
      "language_loss": 0.85319853,
      "learning_rate": 8.300693612881992e-09,
      "loss": 0.87513262,
      "num_input_tokens_seen": 174483600,
      "step": 8082,
      "time_per_iteration": 2.633908748626709
    },
    {
      "auxiliary_loss_clip": 0.01147845,
      "auxiliary_loss_mlp": 0.0076189,
      "balance_loss_clip": 1.04408598,
      "balance_loss_mlp": 1.0003407,
      "epoch": 0.9719232850357723,
      "flos": 22090700793600.0,
      "grad_norm": 2.090306427210176,
      "language_loss": 0.81399083,
      "learning_rate": 8.22994770095664e-09,
      "loss": 0.83308816,
      "num_input_tokens_seen": 174502175,
      "step": 8083,
      "time_per_iteration": 2.6488468647003174
    },
    {
      "auxiliary_loss_clip": 0.01137183,
      "auxiliary_loss_mlp": 0.01025511,
      "balance_loss_clip": 1.04599082,
      "balance_loss_mlp": 1.01792049,
      "epoch": 0.9720435279264114,
      "flos": 23656908493440.0,
      "grad_norm": 2.1152249256931364,
      "language_loss": 0.75083023,
      "learning_rate": 8.159503938964585e-09,
      "loss": 0.77245718,
      "num_input_tokens_seen": 174519495,
      "step": 8084,
      "time_per_iteration": 2.721585273742676
    },
    {
      "auxiliary_loss_clip": 0.01114561,
      "auxiliary_loss_mlp": 0.01028038,
      "balance_loss_clip": 1.0394069,
      "balance_loss_mlp": 1.02158833,
      "epoch": 0.9721637708170504,
      "flos": 28365910623360.0,
      "grad_norm": 1.8368540400125888,
      "language_loss": 0.70764446,
      "learning_rate": 8.089362337592164e-09,
      "loss": 0.72907043,
      "num_input_tokens_seen": 174543120,
      "step": 8085,
      "time_per_iteration": 2.718135118484497
    },
    {
      "auxiliary_loss_clip": 0.01132508,
      "auxiliary_loss_mlp": 0.01023029,
      "balance_loss_clip": 1.04196,
      "balance_loss_mlp": 1.01658845,
      "epoch": 0.9722840137076896,
      "flos": 29130807767040.0,
      "grad_norm": 3.46885969041535,
      "language_loss": 0.72249019,
      "learning_rate": 8.019522907479536e-09,
      "loss": 0.74404556,
      "num_input_tokens_seen": 174563480,
      "step": 8086,
      "time_per_iteration": 3.63275408744812
    },
    {
      "auxiliary_loss_clip": 0.01153868,
      "auxiliary_loss_mlp": 0.01029175,
      "balance_loss_clip": 1.04507995,
      "balance_loss_mlp": 1.02229667,
      "epoch": 0.9724042565983286,
      "flos": 19243954258560.0,
      "grad_norm": 2.2276532267569378,
      "language_loss": 0.77472007,
      "learning_rate": 7.949985659221558e-09,
      "loss": 0.79655051,
      "num_input_tokens_seen": 174580745,
      "step": 8087,
      "time_per_iteration": 2.604414224624634
    },
    {
      "auxiliary_loss_clip": 0.01141328,
      "auxiliary_loss_mlp": 0.01018154,
      "balance_loss_clip": 1.04419255,
      "balance_loss_mlp": 1.01144791,
      "epoch": 0.9725244994889677,
      "flos": 23039676161280.0,
      "grad_norm": 2.138563278236607,
      "language_loss": 0.79001051,
      "learning_rate": 7.880750603366904e-09,
      "loss": 0.81160533,
      "num_input_tokens_seen": 174599615,
      "step": 8088,
      "time_per_iteration": 2.6154944896698
    },
    {
      "auxiliary_loss_clip": 0.01132434,
      "auxiliary_loss_mlp": 0.01024739,
      "balance_loss_clip": 1.0410893,
      "balance_loss_mlp": 1.01737511,
      "epoch": 0.9726447423796069,
      "flos": 23367468700800.0,
      "grad_norm": 1.8956207337363327,
      "language_loss": 0.79512954,
      "learning_rate": 7.811817750418282e-09,
      "loss": 0.81670129,
      "num_input_tokens_seen": 174618375,
      "step": 8089,
      "time_per_iteration": 3.671121835708618
    },
    {
      "auxiliary_loss_clip": 0.01122122,
      "auxiliary_loss_mlp": 0.01026904,
      "balance_loss_clip": 1.04384184,
      "balance_loss_mlp": 1.01919174,
      "epoch": 0.9727649852702459,
      "flos": 26541648639360.0,
      "grad_norm": 1.647168837692988,
      "language_loss": 0.80164206,
      "learning_rate": 7.743187110833105e-09,
      "loss": 0.82313234,
      "num_input_tokens_seen": 174641135,
      "step": 8090,
      "time_per_iteration": 2.7457826137542725
    },
    {
      "auxiliary_loss_clip": 0.01139181,
      "auxiliary_loss_mlp": 0.01022722,
      "balance_loss_clip": 1.04079485,
      "balance_loss_mlp": 1.0157218,
      "epoch": 0.972885228160885,
      "flos": 20522338277760.0,
      "grad_norm": 1.5981395618738707,
      "language_loss": 0.80344498,
      "learning_rate": 7.674858695022602e-09,
      "loss": 0.82506406,
      "num_input_tokens_seen": 174659490,
      "step": 8091,
      "time_per_iteration": 3.752051830291748
    },
    {
      "auxiliary_loss_clip": 0.01170337,
      "auxiliary_loss_mlp": 0.01025332,
      "balance_loss_clip": 1.0484587,
      "balance_loss_mlp": 1.01738024,
      "epoch": 0.9730054710515241,
      "flos": 17566064196480.0,
      "grad_norm": 2.840334889928141,
      "language_loss": 0.76064342,
      "learning_rate": 7.606832513351591e-09,
      "loss": 0.7826001,
      "num_input_tokens_seen": 174677440,
      "step": 8092,
      "time_per_iteration": 2.59678316116333
    },
    {
      "auxiliary_loss_clip": 0.01062624,
      "auxiliary_loss_mlp": 0.00753441,
      "balance_loss_clip": 1.00899005,
      "balance_loss_mlp": 1.00016236,
      "epoch": 0.9731257139421632,
      "flos": 68972010117120.0,
      "grad_norm": 0.8261800955079551,
      "language_loss": 0.63907009,
      "learning_rate": 7.539108576140264e-09,
      "loss": 0.65723073,
      "num_input_tokens_seen": 174741550,
      "step": 8093,
      "time_per_iteration": 3.216064929962158
    },
    {
      "auxiliary_loss_clip": 0.01108867,
      "auxiliary_loss_mlp": 0.01024843,
      "balance_loss_clip": 1.04023433,
      "balance_loss_mlp": 1.01836085,
      "epoch": 0.9732459568328022,
      "flos": 18478841633280.0,
      "grad_norm": 2.0504127563384755,
      "language_loss": 0.7038753,
      "learning_rate": 7.471686893661732e-09,
      "loss": 0.7252124,
      "num_input_tokens_seen": 174759845,
      "step": 8094,
      "time_per_iteration": 3.6348602771759033
    },
    {
      "auxiliary_loss_clip": 0.01134704,
      "auxiliary_loss_mlp": 0.01026226,
      "balance_loss_clip": 1.04401946,
      "balance_loss_mlp": 1.01898372,
      "epoch": 0.9733661997234414,
      "flos": 20883886623360.0,
      "grad_norm": 1.7846509784246194,
      "language_loss": 0.64235985,
      "learning_rate": 7.4045674761442636e-09,
      "loss": 0.66396916,
      "num_input_tokens_seen": 174777175,
      "step": 8095,
      "time_per_iteration": 2.6099915504455566
    },
    {
      "auxiliary_loss_clip": 0.0116434,
      "auxiliary_loss_mlp": 0.00762111,
      "balance_loss_clip": 1.04541135,
      "balance_loss_mlp": 1.00033295,
      "epoch": 0.9734864426140805,
      "flos": 23766795175680.0,
      "grad_norm": 1.6945926792280717,
      "language_loss": 0.74286604,
      "learning_rate": 7.337750333769488e-09,
      "loss": 0.76213056,
      "num_input_tokens_seen": 174796980,
      "step": 8096,
      "time_per_iteration": 2.6134121417999268
    },
    {
      "auxiliary_loss_clip": 0.0114104,
      "auxiliary_loss_mlp": 0.01023215,
      "balance_loss_clip": 1.03988278,
      "balance_loss_mlp": 1.0160625,
      "epoch": 0.9736066855047195,
      "flos": 35042422176000.0,
      "grad_norm": 1.7554185162905058,
      "language_loss": 0.72705424,
      "learning_rate": 7.2712354766737425e-09,
      "loss": 0.7486968,
      "num_input_tokens_seen": 174817310,
      "step": 8097,
      "time_per_iteration": 2.76782488822937
    },
    {
      "auxiliary_loss_clip": 0.01113991,
      "auxiliary_loss_mlp": 0.01023467,
      "balance_loss_clip": 1.0431515,
      "balance_loss_mlp": 1.01661837,
      "epoch": 0.9737269283953586,
      "flos": 20410620001920.0,
      "grad_norm": 1.575664873686886,
      "language_loss": 0.80920982,
      "learning_rate": 7.2050229149469565e-09,
      "loss": 0.83058441,
      "num_input_tokens_seen": 174837320,
      "step": 8098,
      "time_per_iteration": 2.6716487407684326
    },
    {
      "auxiliary_loss_clip": 0.0112504,
      "auxiliary_loss_mlp": 0.01030298,
      "balance_loss_clip": 1.03831649,
      "balance_loss_mlp": 1.02287161,
      "epoch": 0.9738471712859977,
      "flos": 28911680847360.0,
      "grad_norm": 1.7679828878158725,
      "language_loss": 0.63426232,
      "learning_rate": 7.139112658633984e-09,
      "loss": 0.65581572,
      "num_input_tokens_seen": 174857470,
      "step": 8099,
      "time_per_iteration": 2.749349594116211
    },
    {
      "auxiliary_loss_clip": 0.01118199,
      "auxiliary_loss_mlp": 0.0102359,
      "balance_loss_clip": 1.04044533,
      "balance_loss_mlp": 1.01632762,
      "epoch": 0.9739674141766368,
      "flos": 27782326356480.0,
      "grad_norm": 2.0629026505610195,
      "language_loss": 0.70097077,
      "learning_rate": 7.073504717733048e-09,
      "loss": 0.72238868,
      "num_input_tokens_seen": 174877035,
      "step": 8100,
      "time_per_iteration": 2.7755634784698486
    },
    {
      "auxiliary_loss_clip": 0.01024955,
      "auxiliary_loss_mlp": 0.0100166,
      "balance_loss_clip": 1.01240873,
      "balance_loss_mlp": 1.00064123,
      "epoch": 0.9740876570672758,
      "flos": 68863057188480.0,
      "grad_norm": 0.7339323358517948,
      "language_loss": 0.57175195,
      "learning_rate": 7.008199102196855e-09,
      "loss": 0.59201813,
      "num_input_tokens_seen": 174938460,
      "step": 8101,
      "time_per_iteration": 3.2815632820129395
    },
    {
      "auxiliary_loss_clip": 0.01032645,
      "auxiliary_loss_mlp": 0.01000333,
      "balance_loss_clip": 1.00675654,
      "balance_loss_mlp": 0.99934959,
      "epoch": 0.974207899957915,
      "flos": 58236622646400.0,
      "grad_norm": 0.8249126286341723,
      "language_loss": 0.58906424,
      "learning_rate": 6.9431958219321464e-09,
      "loss": 0.60939395,
      "num_input_tokens_seen": 174994625,
      "step": 8102,
      "time_per_iteration": 3.2441694736480713
    },
    {
      "auxiliary_loss_clip": 0.01135501,
      "auxiliary_loss_mlp": 0.0102135,
      "balance_loss_clip": 1.04233003,
      "balance_loss_mlp": 1.01429582,
      "epoch": 0.9743281428485541,
      "flos": 22600057605120.0,
      "grad_norm": 1.4945001756368512,
      "language_loss": 0.77813798,
      "learning_rate": 6.878494886800146e-09,
      "loss": 0.79970652,
      "num_input_tokens_seen": 175015400,
      "step": 8103,
      "time_per_iteration": 2.7675180435180664
    },
    {
      "auxiliary_loss_clip": 0.01138428,
      "auxiliary_loss_mlp": 0.01029489,
      "balance_loss_clip": 1.0442431,
      "balance_loss_mlp": 1.02197289,
      "epoch": 0.9744483857391931,
      "flos": 20008815488640.0,
      "grad_norm": 2.002344926655349,
      "language_loss": 0.76143825,
      "learning_rate": 6.814096306615669e-09,
      "loss": 0.78311741,
      "num_input_tokens_seen": 175033540,
      "step": 8104,
      "time_per_iteration": 2.7644031047821045
    },
    {
      "auxiliary_loss_clip": 0.01140892,
      "auxiliary_loss_mlp": 0.01024959,
      "balance_loss_clip": 1.0399853,
      "balance_loss_mlp": 1.01796186,
      "epoch": 0.9745686286298323,
      "flos": 17675268520320.0,
      "grad_norm": 2.2891246721604803,
      "language_loss": 0.65606594,
      "learning_rate": 6.750000091148011e-09,
      "loss": 0.67772448,
      "num_input_tokens_seen": 175050835,
      "step": 8105,
      "time_per_iteration": 2.714402437210083
    },
    {
      "auxiliary_loss_clip": 0.01166987,
      "auxiliary_loss_mlp": 0.01025511,
      "balance_loss_clip": 1.04796791,
      "balance_loss_mlp": 1.01796508,
      "epoch": 0.9746888715204713,
      "flos": 29460252332160.0,
      "grad_norm": 2.0108515934955977,
      "language_loss": 0.72857213,
      "learning_rate": 6.686206250120729e-09,
      "loss": 0.7504971,
      "num_input_tokens_seen": 175072330,
      "step": 8106,
      "time_per_iteration": 2.7278361320495605
    },
    {
      "auxiliary_loss_clip": 0.01125647,
      "auxiliary_loss_mlp": 0.01022082,
      "balance_loss_clip": 1.03930378,
      "balance_loss_mlp": 1.01516449,
      "epoch": 0.9748091144111104,
      "flos": 18479308510080.0,
      "grad_norm": 1.6854119134987964,
      "language_loss": 0.74587727,
      "learning_rate": 6.622714793210749e-09,
      "loss": 0.76735461,
      "num_input_tokens_seen": 175091250,
      "step": 8107,
      "time_per_iteration": 2.689077138900757
    },
    {
      "auxiliary_loss_clip": 0.01165123,
      "auxiliary_loss_mlp": 0.01026138,
      "balance_loss_clip": 1.04551291,
      "balance_loss_mlp": 1.01892316,
      "epoch": 0.9749293573017496,
      "flos": 20665154753280.0,
      "grad_norm": 1.7145719279573164,
      "language_loss": 0.7860502,
      "learning_rate": 6.559525730050364e-09,
      "loss": 0.80796278,
      "num_input_tokens_seen": 175111350,
      "step": 8108,
      "time_per_iteration": 2.6869115829467773
    },
    {
      "auxiliary_loss_clip": 0.01125431,
      "auxiliary_loss_mlp": 0.01024877,
      "balance_loss_clip": 1.04283524,
      "balance_loss_mlp": 1.01794791,
      "epoch": 0.9750496001923886,
      "flos": 18478590238080.0,
      "grad_norm": 1.9009229129543723,
      "language_loss": 0.75472569,
      "learning_rate": 6.496639070224574e-09,
      "loss": 0.77622873,
      "num_input_tokens_seen": 175129835,
      "step": 8109,
      "time_per_iteration": 2.731154441833496
    },
    {
      "auxiliary_loss_clip": 0.01154906,
      "auxiliary_loss_mlp": 0.01025906,
      "balance_loss_clip": 1.04640293,
      "balance_loss_mlp": 1.01886415,
      "epoch": 0.9751698430830277,
      "flos": 19572967860480.0,
      "grad_norm": 2.09903623272339,
      "language_loss": 0.8371563,
      "learning_rate": 6.4340548232739714e-09,
      "loss": 0.85896438,
      "num_input_tokens_seen": 175146035,
      "step": 8110,
      "time_per_iteration": 2.696333169937134
    },
    {
      "auxiliary_loss_clip": 0.01128742,
      "auxiliary_loss_mlp": 0.01021259,
      "balance_loss_clip": 1.04198575,
      "balance_loss_mlp": 1.01379299,
      "epoch": 0.9752900859736668,
      "flos": 23550325862400.0,
      "grad_norm": 2.8387876805632923,
      "language_loss": 0.79138166,
      "learning_rate": 6.371772998692071e-09,
      "loss": 0.81288165,
      "num_input_tokens_seen": 175165290,
      "step": 8111,
      "time_per_iteration": 3.7230780124664307
    },
    {
      "auxiliary_loss_clip": 0.01125927,
      "auxiliary_loss_mlp": 0.01024113,
      "balance_loss_clip": 1.04051101,
      "balance_loss_mlp": 1.01685643,
      "epoch": 0.9754103288643059,
      "flos": 20303211358080.0,
      "grad_norm": 3.019132009698342,
      "language_loss": 0.64907354,
      "learning_rate": 6.309793605927094e-09,
      "loss": 0.67057395,
      "num_input_tokens_seen": 175183610,
      "step": 8112,
      "time_per_iteration": 2.7475764751434326
    },
    {
      "auxiliary_loss_clip": 0.01143398,
      "auxiliary_loss_mlp": 0.01024055,
      "balance_loss_clip": 1.04349804,
      "balance_loss_mlp": 1.01744795,
      "epoch": 0.975530571754945,
      "flos": 19350680544000.0,
      "grad_norm": 1.8785991568822407,
      "language_loss": 0.80222952,
      "learning_rate": 6.248116654381297e-09,
      "loss": 0.82390404,
      "num_input_tokens_seen": 175202080,
      "step": 8113,
      "time_per_iteration": 2.715064525604248
    },
    {
      "auxiliary_loss_clip": 0.01140204,
      "auxiliary_loss_mlp": 0.01023254,
      "balance_loss_clip": 1.04084516,
      "balance_loss_mlp": 1.01638806,
      "epoch": 0.9756508146455841,
      "flos": 23583399310080.0,
      "grad_norm": 1.8300342499621864,
      "language_loss": 0.72910714,
      "learning_rate": 6.186742153410751e-09,
      "loss": 0.75074172,
      "num_input_tokens_seen": 175221575,
      "step": 8114,
      "time_per_iteration": 2.7126028537750244
    },
    {
      "auxiliary_loss_clip": 0.0113594,
      "auxiliary_loss_mlp": 0.01026205,
      "balance_loss_clip": 1.04264486,
      "balance_loss_mlp": 1.01879561,
      "epoch": 0.9757710575362232,
      "flos": 22966921163520.0,
      "grad_norm": 2.41917967084347,
      "language_loss": 0.87562561,
      "learning_rate": 6.125670112326453e-09,
      "loss": 0.89724708,
      "num_input_tokens_seen": 175240835,
      "step": 8115,
      "time_per_iteration": 3.6568708419799805
    },
    {
      "auxiliary_loss_clip": 0.01151161,
      "auxiliary_loss_mlp": 0.01026614,
      "balance_loss_clip": 1.04149795,
      "balance_loss_mlp": 1.01925898,
      "epoch": 0.9758913004268622,
      "flos": 27966009530880.0,
      "grad_norm": 1.7284911357517998,
      "language_loss": 0.70486683,
      "learning_rate": 6.064900540392548e-09,
      "loss": 0.72664458,
      "num_input_tokens_seen": 175262930,
      "step": 8116,
      "time_per_iteration": 2.716320753097534
    },
    {
      "auxiliary_loss_clip": 0.01130613,
      "auxiliary_loss_mlp": 0.01023022,
      "balance_loss_clip": 1.04308248,
      "balance_loss_mlp": 1.016469,
      "epoch": 0.9760115433175014,
      "flos": 22200156512640.0,
      "grad_norm": 4.728989930940953,
      "language_loss": 0.79073912,
      "learning_rate": 6.0044334468278835e-09,
      "loss": 0.81227547,
      "num_input_tokens_seen": 175282275,
      "step": 8117,
      "time_per_iteration": 2.669156312942505
    },
    {
      "auxiliary_loss_clip": 0.01109827,
      "auxiliary_loss_mlp": 0.01021667,
      "balance_loss_clip": 1.03847146,
      "balance_loss_mlp": 1.01469326,
      "epoch": 0.9761317862081405,
      "flos": 26250736389120.0,
      "grad_norm": 2.040465469355562,
      "language_loss": 0.7147063,
      "learning_rate": 5.944268840805345e-09,
      "loss": 0.73602128,
      "num_input_tokens_seen": 175303020,
      "step": 8118,
      "time_per_iteration": 3.776001453399658
    },
    {
      "auxiliary_loss_clip": 0.01116743,
      "auxiliary_loss_mlp": 0.01022024,
      "balance_loss_clip": 1.03886795,
      "balance_loss_mlp": 1.01520491,
      "epoch": 0.9762520290987795,
      "flos": 26575440359040.0,
      "grad_norm": 1.989429660408004,
      "language_loss": 0.63801301,
      "learning_rate": 5.88440673145163e-09,
      "loss": 0.6594007,
      "num_input_tokens_seen": 175324070,
      "step": 8119,
      "time_per_iteration": 2.734724998474121
    },
    {
      "auxiliary_loss_clip": 0.01149557,
      "auxiliary_loss_mlp": 0.01026851,
      "balance_loss_clip": 1.04575324,
      "balance_loss_mlp": 1.02028859,
      "epoch": 0.9763722719894187,
      "flos": 18005036307840.0,
      "grad_norm": 2.0426077305484363,
      "language_loss": 0.8259353,
      "learning_rate": 5.824847127848142e-09,
      "loss": 0.8476994,
      "num_input_tokens_seen": 175342595,
      "step": 8120,
      "time_per_iteration": 3.487734079360962
    },
    {
      "auxiliary_loss_clip": 0.01112863,
      "auxiliary_loss_mlp": 0.01025172,
      "balance_loss_clip": 1.04198945,
      "balance_loss_mlp": 1.01790285,
      "epoch": 0.9764925148800577,
      "flos": 22455660931200.0,
      "grad_norm": 1.8685419702788058,
      "language_loss": 0.78859097,
      "learning_rate": 5.765590039029433e-09,
      "loss": 0.80997127,
      "num_input_tokens_seen": 175361915,
      "step": 8121,
      "time_per_iteration": 2.7872447967529297
    },
    {
      "auxiliary_loss_clip": 0.01166608,
      "auxiliary_loss_mlp": 0.01026873,
      "balance_loss_clip": 1.04851007,
      "balance_loss_mlp": 1.01954818,
      "epoch": 0.9766127577706968,
      "flos": 36757084786560.0,
      "grad_norm": 1.7874118830098242,
      "language_loss": 0.71315247,
      "learning_rate": 5.706635473985422e-09,
      "loss": 0.73508728,
      "num_input_tokens_seen": 175385785,
      "step": 8122,
      "time_per_iteration": 2.7344343662261963
    },
    {
      "auxiliary_loss_clip": 0.01151924,
      "auxiliary_loss_mlp": 0.01022649,
      "balance_loss_clip": 1.04526532,
      "balance_loss_mlp": 1.01608014,
      "epoch": 0.976733000661336,
      "flos": 22309971367680.0,
      "grad_norm": 2.3064775039918843,
      "language_loss": 0.84983075,
      "learning_rate": 5.6479834416591764e-09,
      "loss": 0.87157643,
      "num_input_tokens_seen": 175405145,
      "step": 8123,
      "time_per_iteration": 2.646451234817505
    },
    {
      "auxiliary_loss_clip": 0.01150085,
      "auxiliary_loss_mlp": 0.00762808,
      "balance_loss_clip": 1.04493523,
      "balance_loss_mlp": 1.00034297,
      "epoch": 0.976853243551975,
      "flos": 25810938264960.0,
      "grad_norm": 1.829347339933807,
      "language_loss": 0.68508476,
      "learning_rate": 5.589633950947803e-09,
      "loss": 0.70421368,
      "num_input_tokens_seen": 175422645,
      "step": 8124,
      "time_per_iteration": 2.705111503601074
    },
    {
      "auxiliary_loss_clip": 0.01133998,
      "auxiliary_loss_mlp": 0.0102894,
      "balance_loss_clip": 1.04187322,
      "balance_loss_mlp": 1.02106035,
      "epoch": 0.9769734864426141,
      "flos": 21397445326080.0,
      "grad_norm": 2.189332987874214,
      "language_loss": 0.69934523,
      "learning_rate": 5.5315870107035535e-09,
      "loss": 0.72097468,
      "num_input_tokens_seen": 175440695,
      "step": 8125,
      "time_per_iteration": 2.6548855304718018
    },
    {
      "auxiliary_loss_clip": 0.01134196,
      "auxiliary_loss_mlp": 0.01025833,
      "balance_loss_clip": 1.0448879,
      "balance_loss_mlp": 1.01911521,
      "epoch": 0.9770937293332532,
      "flos": 13990977584640.0,
      "grad_norm": 2.115374199716958,
      "language_loss": 0.78898406,
      "learning_rate": 5.473842629731607e-09,
      "loss": 0.81058431,
      "num_input_tokens_seen": 175459195,
      "step": 8126,
      "time_per_iteration": 2.8073999881744385
    },
    {
      "auxiliary_loss_clip": 0.01143917,
      "auxiliary_loss_mlp": 0.0076242,
      "balance_loss_clip": 1.04227924,
      "balance_loss_mlp": 1.00035322,
      "epoch": 0.9772139722238923,
      "flos": 17931994001280.0,
      "grad_norm": 2.1692997309203936,
      "language_loss": 0.77924281,
      "learning_rate": 5.416400816792066e-09,
      "loss": 0.79830623,
      "num_input_tokens_seen": 175476710,
      "step": 8127,
      "time_per_iteration": 2.6686689853668213
    },
    {
      "auxiliary_loss_clip": 0.01162865,
      "auxiliary_loss_mlp": 0.01027232,
      "balance_loss_clip": 1.04358149,
      "balance_loss_mlp": 1.02013326,
      "epoch": 0.9773342151145313,
      "flos": 20446171488000.0,
      "grad_norm": 2.8843343434377693,
      "language_loss": 0.7865994,
      "learning_rate": 5.359261580598407e-09,
      "loss": 0.80850035,
      "num_input_tokens_seen": 175492550,
      "step": 8128,
      "time_per_iteration": 2.570988178253174
    },
    {
      "auxiliary_loss_clip": 0.01153326,
      "auxiliary_loss_mlp": 0.01025125,
      "balance_loss_clip": 1.04546785,
      "balance_loss_mlp": 1.01720929,
      "epoch": 0.9774544580051704,
      "flos": 11837306949120.0,
      "grad_norm": 3.3427491954543767,
      "language_loss": 0.78530645,
      "learning_rate": 5.302424929819027e-09,
      "loss": 0.807091,
      "num_input_tokens_seen": 175506560,
      "step": 8129,
      "time_per_iteration": 2.6053876876831055
    },
    {
      "auxiliary_loss_clip": 0.01153821,
      "auxiliary_loss_mlp": 0.01025449,
      "balance_loss_clip": 1.04183173,
      "balance_loss_mlp": 1.01781917,
      "epoch": 0.9775747008958096,
      "flos": 13479932833920.0,
      "grad_norm": 2.20199729506144,
      "language_loss": 0.73013997,
      "learning_rate": 5.24589087307592e-09,
      "loss": 0.75193268,
      "num_input_tokens_seen": 175524180,
      "step": 8130,
      "time_per_iteration": 2.5992488861083984
    },
    {
      "auxiliary_loss_clip": 0.01166126,
      "auxiliary_loss_mlp": 0.01027368,
      "balance_loss_clip": 1.04419351,
      "balance_loss_mlp": 1.01988792,
      "epoch": 0.9776949437864486,
      "flos": 59532314042880.0,
      "grad_norm": 1.578165977394243,
      "language_loss": 0.65154976,
      "learning_rate": 5.189659418944891e-09,
      "loss": 0.67348468,
      "num_input_tokens_seen": 175554355,
      "step": 8131,
      "time_per_iteration": 2.973022937774658
    },
    {
      "auxiliary_loss_clip": 0.01163874,
      "auxiliary_loss_mlp": 0.01021509,
      "balance_loss_clip": 1.04551375,
      "balance_loss_mlp": 1.01459217,
      "epoch": 0.9778151866770877,
      "flos": 21178605715200.0,
      "grad_norm": 2.3269979611978053,
      "language_loss": 0.78295535,
      "learning_rate": 5.133730575956674e-09,
      "loss": 0.80480915,
      "num_input_tokens_seen": 175574025,
      "step": 8132,
      "time_per_iteration": 2.673265218734741
    },
    {
      "auxiliary_loss_clip": 0.01139395,
      "auxiliary_loss_mlp": 0.01026787,
      "balance_loss_clip": 1.04282904,
      "balance_loss_mlp": 1.01973605,
      "epoch": 0.9779354295677268,
      "flos": 20886795624960.0,
      "grad_norm": 2.0552605638702452,
      "language_loss": 0.72262281,
      "learning_rate": 5.0781043525953696e-09,
      "loss": 0.74428463,
      "num_input_tokens_seen": 175592090,
      "step": 8133,
      "time_per_iteration": 2.657054901123047
    },
    {
      "auxiliary_loss_clip": 0.01133866,
      "auxiliary_loss_mlp": 0.01025816,
      "balance_loss_clip": 1.04449224,
      "balance_loss_mlp": 1.01935816,
      "epoch": 0.9780556724583659,
      "flos": 23440618748160.0,
      "grad_norm": 1.6606710159370175,
      "language_loss": 0.74264109,
      "learning_rate": 5.0227807572995605e-09,
      "loss": 0.76423794,
      "num_input_tokens_seen": 175614065,
      "step": 8134,
      "time_per_iteration": 2.6526482105255127
    },
    {
      "auxiliary_loss_clip": 0.01139703,
      "auxiliary_loss_mlp": 0.01022487,
      "balance_loss_clip": 1.04274726,
      "balance_loss_mlp": 1.01537919,
      "epoch": 0.9781759153490049,
      "flos": 20923244951040.0,
      "grad_norm": 2.3448828436760505,
      "language_loss": 0.67721325,
      "learning_rate": 4.967759798461646e-09,
      "loss": 0.69883519,
      "num_input_tokens_seen": 175632410,
      "step": 8135,
      "time_per_iteration": 2.6077494621276855
    },
    {
      "auxiliary_loss_clip": 0.01162097,
      "auxiliary_loss_mlp": 0.01023642,
      "balance_loss_clip": 1.04491401,
      "balance_loss_mlp": 1.01690614,
      "epoch": 0.9782961582396441,
      "flos": 28293191539200.0,
      "grad_norm": 3.6575794373133674,
      "language_loss": 0.75376016,
      "learning_rate": 4.913041484428282e-09,
      "loss": 0.7756176,
      "num_input_tokens_seen": 175652885,
      "step": 8136,
      "time_per_iteration": 2.672907829284668
    },
    {
      "auxiliary_loss_clip": 0.01152819,
      "auxiliary_loss_mlp": 0.01025443,
      "balance_loss_clip": 1.04418004,
      "balance_loss_mlp": 1.01909506,
      "epoch": 0.9784164011302832,
      "flos": 25552955808000.0,
      "grad_norm": 2.1075836981924225,
      "language_loss": 0.74337792,
      "learning_rate": 4.858625823500384e-09,
      "loss": 0.76516056,
      "num_input_tokens_seen": 175670585,
      "step": 8137,
      "time_per_iteration": 2.5978081226348877
    },
    {
      "auxiliary_loss_clip": 0.01155742,
      "auxiliary_loss_mlp": 0.01023795,
      "balance_loss_clip": 1.04347515,
      "balance_loss_mlp": 1.01701546,
      "epoch": 0.9785366440209222,
      "flos": 29965945956480.0,
      "grad_norm": 3.1853053279706005,
      "language_loss": 0.73423994,
      "learning_rate": 4.80451282393246e-09,
      "loss": 0.75603533,
      "num_input_tokens_seen": 175690570,
      "step": 8138,
      "time_per_iteration": 3.8139994144439697
    },
    {
      "auxiliary_loss_clip": 0.0114037,
      "auxiliary_loss_mlp": 0.01027798,
      "balance_loss_clip": 1.04282141,
      "balance_loss_mlp": 1.02091014,
      "epoch": 0.9786568869115614,
      "flos": 32343591847680.0,
      "grad_norm": 1.9869304619092698,
      "language_loss": 0.67708361,
      "learning_rate": 4.750702493933722e-09,
      "loss": 0.69876528,
      "num_input_tokens_seen": 175710455,
      "step": 8139,
      "time_per_iteration": 2.7291347980499268
    },
    {
      "auxiliary_loss_clip": 0.0113469,
      "auxiliary_loss_mlp": 0.00761959,
      "balance_loss_clip": 1.04277849,
      "balance_loss_mlp": 1.00031424,
      "epoch": 0.9787771298022004,
      "flos": 23331414424320.0,
      "grad_norm": 1.8392969592954262,
      "language_loss": 0.85545707,
      "learning_rate": 4.697194841666974e-09,
      "loss": 0.87442362,
      "num_input_tokens_seen": 175729380,
      "step": 8140,
      "time_per_iteration": 2.7855002880096436
    },
    {
      "auxiliary_loss_clip": 0.01153598,
      "auxiliary_loss_mlp": 0.01030994,
      "balance_loss_clip": 1.04464889,
      "balance_loss_mlp": 1.02316809,
      "epoch": 0.9788973726928395,
      "flos": 21468548298240.0,
      "grad_norm": 1.7732525107583232,
      "language_loss": 0.81643063,
      "learning_rate": 4.6439898752492764e-09,
      "loss": 0.83827651,
      "num_input_tokens_seen": 175749520,
      "step": 8141,
      "time_per_iteration": 3.8103091716766357
    },
    {
      "auxiliary_loss_clip": 0.01054469,
      "auxiliary_loss_mlp": 0.0075339,
      "balance_loss_clip": 1.01008213,
      "balance_loss_mlp": 1.00018883,
      "epoch": 0.9790176155834787,
      "flos": 68897459439360.0,
      "grad_norm": 0.7504176875704255,
      "language_loss": 0.63695931,
      "learning_rate": 4.591087602751731e-09,
      "loss": 0.65503788,
      "num_input_tokens_seen": 175811380,
      "step": 8142,
      "time_per_iteration": 3.3274405002593994
    },
    {
      "auxiliary_loss_clip": 0.01149014,
      "auxiliary_loss_mlp": 0.01027644,
      "balance_loss_clip": 1.04427743,
      "balance_loss_mlp": 1.02076519,
      "epoch": 0.9791378584741177,
      "flos": 21430877909760.0,
      "grad_norm": 1.7510427093260563,
      "language_loss": 0.72092056,
      "learning_rate": 4.538488032199916e-09,
      "loss": 0.74268711,
      "num_input_tokens_seen": 175829480,
      "step": 8143,
      "time_per_iteration": 2.612973928451538
    },
    {
      "auxiliary_loss_clip": 0.01154885,
      "auxiliary_loss_mlp": 0.01022447,
      "balance_loss_clip": 1.04258204,
      "balance_loss_mlp": 1.0153358,
      "epoch": 0.9792581013647568,
      "flos": 20153032594560.0,
      "grad_norm": 1.9712248597606088,
      "language_loss": 0.68668032,
      "learning_rate": 4.486191171572784e-09,
      "loss": 0.70845366,
      "num_input_tokens_seen": 175846750,
      "step": 8144,
      "time_per_iteration": 3.537792921066284
    },
    {
      "auxiliary_loss_clip": 0.01153754,
      "auxiliary_loss_mlp": 0.01028266,
      "balance_loss_clip": 1.04490113,
      "balance_loss_mlp": 1.02094316,
      "epoch": 0.9793783442553959,
      "flos": 23728191033600.0,
      "grad_norm": 1.5234476759349203,
      "language_loss": 0.77439106,
      "learning_rate": 4.434197028803766e-09,
      "loss": 0.79621124,
      "num_input_tokens_seen": 175865975,
      "step": 8145,
      "time_per_iteration": 2.611265182495117
    },
    {
      "auxiliary_loss_clip": 0.01129741,
      "auxiliary_loss_mlp": 0.01026582,
      "balance_loss_clip": 1.04106593,
      "balance_loss_mlp": 1.0198704,
      "epoch": 0.979498587146035,
      "flos": 23038742407680.0,
      "grad_norm": 2.041650679665563,
      "language_loss": 0.82066077,
      "learning_rate": 4.3825056117805514e-09,
      "loss": 0.842224,
      "num_input_tokens_seen": 175881860,
      "step": 8146,
      "time_per_iteration": 3.6784520149230957
    },
    {
      "auxiliary_loss_clip": 0.01165425,
      "auxiliary_loss_mlp": 0.01030603,
      "balance_loss_clip": 1.04501474,
      "balance_loss_mlp": 1.02219307,
      "epoch": 0.979618830036674,
      "flos": 14318841951360.0,
      "grad_norm": 2.716447414150671,
      "language_loss": 0.79114163,
      "learning_rate": 4.331116928344425e-09,
      "loss": 0.81310189,
      "num_input_tokens_seen": 175898175,
      "step": 8147,
      "time_per_iteration": 2.5783259868621826
    },
    {
      "auxiliary_loss_clip": 0.01141224,
      "auxiliary_loss_mlp": 0.0076312,
      "balance_loss_clip": 1.04218304,
      "balance_loss_mlp": 1.0003314,
      "epoch": 0.9797390729273132,
      "flos": 16727514215040.0,
      "grad_norm": 2.2437734598130556,
      "language_loss": 0.62738454,
      "learning_rate": 4.28003098629115e-09,
      "loss": 0.64642793,
      "num_input_tokens_seen": 175914310,
      "step": 8148,
      "time_per_iteration": 2.615752935409546
    },
    {
      "auxiliary_loss_clip": 0.01119415,
      "auxiliary_loss_mlp": 0.01023214,
      "balance_loss_clip": 1.03613353,
      "balance_loss_mlp": 1.01664305,
      "epoch": 0.9798593158179523,
      "flos": 24532661986560.0,
      "grad_norm": 2.0433918952550782,
      "language_loss": 0.78633165,
      "learning_rate": 4.229247793370305e-09,
      "loss": 0.80775797,
      "num_input_tokens_seen": 175933435,
      "step": 8149,
      "time_per_iteration": 2.7146196365356445
    },
    {
      "auxiliary_loss_clip": 0.01168512,
      "auxiliary_loss_mlp": 0.01027161,
      "balance_loss_clip": 1.04755712,
      "balance_loss_mlp": 1.02014565,
      "epoch": 0.9799795587085913,
      "flos": 27308808339840.0,
      "grad_norm": 1.724979543815382,
      "language_loss": 0.70522732,
      "learning_rate": 4.178767357285951e-09,
      "loss": 0.72718406,
      "num_input_tokens_seen": 175955065,
      "step": 8150,
      "time_per_iteration": 2.681532382965088
    },
    {
      "auxiliary_loss_clip": 0.01150771,
      "auxiliary_loss_mlp": 0.00761865,
      "balance_loss_clip": 1.04304504,
      "balance_loss_mlp": 1.00035512,
      "epoch": 0.9800998015992305,
      "flos": 26286575184000.0,
      "grad_norm": 2.170184530775519,
      "language_loss": 0.71411103,
      "learning_rate": 4.128589685695516e-09,
      "loss": 0.73323739,
      "num_input_tokens_seen": 175975490,
      "step": 8151,
      "time_per_iteration": 2.6793222427368164
    },
    {
      "auxiliary_loss_clip": 0.01167683,
      "auxiliary_loss_mlp": 0.01024958,
      "balance_loss_clip": 1.04742479,
      "balance_loss_mlp": 1.01735806,
      "epoch": 0.9802200444898695,
      "flos": 16723635546240.0,
      "grad_norm": 1.8054180062239564,
      "language_loss": 0.84315479,
      "learning_rate": 4.078714786211135e-09,
      "loss": 0.86508119,
      "num_input_tokens_seen": 175991340,
      "step": 8152,
      "time_per_iteration": 2.573296070098877
    },
    {
      "auxiliary_loss_clip": 0.0114703,
      "auxiliary_loss_mlp": 0.01020789,
      "balance_loss_clip": 1.04323006,
      "balance_loss_mlp": 1.01462889,
      "epoch": 0.9803402873805086,
      "flos": 24900459298560.0,
      "grad_norm": 1.7545253710231965,
      "language_loss": 0.76518261,
      "learning_rate": 4.029142666398977e-09,
      "loss": 0.78686088,
      "num_input_tokens_seen": 176011505,
      "step": 8153,
      "time_per_iteration": 2.631821393966675
    },
    {
      "auxiliary_loss_clip": 0.01162652,
      "auxiliary_loss_mlp": 0.01026527,
      "balance_loss_clip": 1.0458796,
      "balance_loss_mlp": 1.01963651,
      "epoch": 0.9804605302711478,
      "flos": 22564937082240.0,
      "grad_norm": 2.0869413702250035,
      "language_loss": 0.8021487,
      "learning_rate": 3.979873333778805e-09,
      "loss": 0.82404053,
      "num_input_tokens_seen": 176029680,
      "step": 8154,
      "time_per_iteration": 2.588289260864258
    },
    {
      "auxiliary_loss_clip": 0.01143204,
      "auxiliary_loss_mlp": 0.01024481,
      "balance_loss_clip": 1.04434109,
      "balance_loss_mlp": 1.01730776,
      "epoch": 0.9805807731617868,
      "flos": 38905368382080.0,
      "grad_norm": 2.278337699844078,
      "language_loss": 0.73996007,
      "learning_rate": 3.930906795824862e-09,
      "loss": 0.76163691,
      "num_input_tokens_seen": 176050355,
      "step": 8155,
      "time_per_iteration": 2.784166097640991
    },
    {
      "auxiliary_loss_clip": 0.0114653,
      "auxiliary_loss_mlp": 0.01022604,
      "balance_loss_clip": 1.04263759,
      "balance_loss_mlp": 1.01569867,
      "epoch": 0.9807010160524259,
      "flos": 17821999578240.0,
      "grad_norm": 3.131827168891957,
      "language_loss": 0.77028275,
      "learning_rate": 3.882243059965207e-09,
      "loss": 0.79197407,
      "num_input_tokens_seen": 176068070,
      "step": 8156,
      "time_per_iteration": 2.5459935665130615
    },
    {
      "auxiliary_loss_clip": 0.01144888,
      "auxiliary_loss_mlp": 0.01025085,
      "balance_loss_clip": 1.04129124,
      "balance_loss_mlp": 1.01779234,
      "epoch": 0.980821258943065,
      "flos": 13552975140480.0,
      "grad_norm": 2.240921791403321,
      "language_loss": 0.65468901,
      "learning_rate": 3.833882133582156e-09,
      "loss": 0.67638874,
      "num_input_tokens_seen": 176083730,
      "step": 8157,
      "time_per_iteration": 2.6086947917938232
    },
    {
      "auxiliary_loss_clip": 0.0115451,
      "auxiliary_loss_mlp": 0.01024857,
      "balance_loss_clip": 1.04410458,
      "balance_loss_mlp": 1.01769519,
      "epoch": 0.9809415018337041,
      "flos": 21689794120320.0,
      "grad_norm": 2.2103862025596097,
      "language_loss": 0.78523493,
      "learning_rate": 3.785824024012285e-09,
      "loss": 0.80702865,
      "num_input_tokens_seen": 176102730,
      "step": 8158,
      "time_per_iteration": 2.606032133102417
    },
    {
      "auxiliary_loss_clip": 0.01131412,
      "auxiliary_loss_mlp": 0.01024078,
      "balance_loss_clip": 1.04399967,
      "balance_loss_mlp": 1.01765585,
      "epoch": 0.9810617447243432,
      "flos": 23294857357440.0,
      "grad_norm": 1.716505167231159,
      "language_loss": 0.78595376,
      "learning_rate": 3.738068738545541e-09,
      "loss": 0.80750871,
      "num_input_tokens_seen": 176121815,
      "step": 8159,
      "time_per_iteration": 2.676825523376465
    },
    {
      "auxiliary_loss_clip": 0.01154198,
      "auxiliary_loss_mlp": 0.01027006,
      "balance_loss_clip": 1.04370856,
      "balance_loss_mlp": 1.01957309,
      "epoch": 0.9811819876149822,
      "flos": 18332038748160.0,
      "grad_norm": 2.1914962391641635,
      "language_loss": 0.78598744,
      "learning_rate": 3.6906162844265733e-09,
      "loss": 0.80779952,
      "num_input_tokens_seen": 176138900,
      "step": 8160,
      "time_per_iteration": 2.5867042541503906
    },
    {
      "auxiliary_loss_clip": 0.0113178,
      "auxiliary_loss_mlp": 0.01023684,
      "balance_loss_clip": 1.04132652,
      "balance_loss_mlp": 1.0158366,
      "epoch": 0.9813022305056214,
      "flos": 22601961025920.0,
      "grad_norm": 1.8135674108100337,
      "language_loss": 0.70616913,
      "learning_rate": 3.643466668853845e-09,
      "loss": 0.72772378,
      "num_input_tokens_seen": 176156925,
      "step": 8161,
      "time_per_iteration": 2.6601109504699707
    },
    {
      "auxiliary_loss_clip": 0.0113848,
      "auxiliary_loss_mlp": 0.0102534,
      "balance_loss_clip": 1.04208255,
      "balance_loss_mlp": 1.01816964,
      "epoch": 0.9814224733962604,
      "flos": 25413335642880.0,
      "grad_norm": 1.9445784665678776,
      "language_loss": 0.75234807,
      "learning_rate": 3.59661989898008e-09,
      "loss": 0.77398622,
      "num_input_tokens_seen": 176177980,
      "step": 8162,
      "time_per_iteration": 2.7560431957244873
    },
    {
      "auxiliary_loss_clip": 0.01118281,
      "auxiliary_loss_mlp": 0.01024843,
      "balance_loss_clip": 1.04307997,
      "balance_loss_mlp": 1.0184145,
      "epoch": 0.9815427162868995,
      "flos": 25007185584000.0,
      "grad_norm": 1.6659894771764059,
      "language_loss": 0.7681272,
      "learning_rate": 3.5500759819115934e-09,
      "loss": 0.78955847,
      "num_input_tokens_seen": 176198345,
      "step": 8163,
      "time_per_iteration": 3.583566904067993
    },
    {
      "auxiliary_loss_clip": 0.0116746,
      "auxiliary_loss_mlp": 0.0102327,
      "balance_loss_clip": 1.04692888,
      "balance_loss_mlp": 1.01607835,
      "epoch": 0.9816629591775387,
      "flos": 20662604887680.0,
      "grad_norm": 2.151600704198693,
      "language_loss": 0.81294388,
      "learning_rate": 3.5038349247094034e-09,
      "loss": 0.83485121,
      "num_input_tokens_seen": 176215605,
      "step": 8164,
      "time_per_iteration": 2.5667433738708496
    },
    {
      "auxiliary_loss_clip": 0.01137423,
      "auxiliary_loss_mlp": 0.01027245,
      "balance_loss_clip": 1.04263759,
      "balance_loss_mlp": 1.02066433,
      "epoch": 0.9817832020681777,
      "flos": 17712220636800.0,
      "grad_norm": 2.023422904360436,
      "language_loss": 0.7771191,
      "learning_rate": 3.4578967343878994e-09,
      "loss": 0.79876578,
      "num_input_tokens_seen": 176231810,
      "step": 8165,
      "time_per_iteration": 2.6130032539367676
    },
    {
      "auxiliary_loss_clip": 0.01135123,
      "auxiliary_loss_mlp": 0.01028815,
      "balance_loss_clip": 1.04269052,
      "balance_loss_mlp": 1.02179396,
      "epoch": 0.9819034449588168,
      "flos": 22530032040960.0,
      "grad_norm": 1.7365857324865317,
      "language_loss": 0.80904609,
      "learning_rate": 3.4122614179161733e-09,
      "loss": 0.8306855,
      "num_input_tokens_seen": 176251770,
      "step": 8166,
      "time_per_iteration": 2.6961042881011963
    },
    {
      "auxiliary_loss_clip": 0.01108399,
      "auxiliary_loss_mlp": 0.01022677,
      "balance_loss_clip": 1.03665566,
      "balance_loss_mlp": 1.01570892,
      "epoch": 0.9820236878494559,
      "flos": 20011221699840.0,
      "grad_norm": 1.6523674337983187,
      "language_loss": 0.78132474,
      "learning_rate": 3.36692898221691e-09,
      "loss": 0.80263549,
      "num_input_tokens_seen": 176270135,
      "step": 8167,
      "time_per_iteration": 3.6654086112976074
    },
    {
      "auxiliary_loss_clip": 0.01150697,
      "auxiliary_loss_mlp": 0.01026574,
      "balance_loss_clip": 1.04295719,
      "balance_loss_mlp": 1.01986802,
      "epoch": 0.982143930740095,
      "flos": 18807316531200.0,
      "grad_norm": 3.6790937630067595,
      "language_loss": 0.73610651,
      "learning_rate": 3.3218994341668305e-09,
      "loss": 0.75787914,
      "num_input_tokens_seen": 176289065,
      "step": 8168,
      "time_per_iteration": 2.687021017074585
    },
    {
      "auxiliary_loss_clip": 0.01163809,
      "auxiliary_loss_mlp": 0.01028423,
      "balance_loss_clip": 1.04731798,
      "balance_loss_mlp": 1.02141356,
      "epoch": 0.982264173630734,
      "flos": 26578026138240.0,
      "grad_norm": 1.769443498432387,
      "language_loss": 0.75702226,
      "learning_rate": 3.2771727805971373e-09,
      "loss": 0.77894461,
      "num_input_tokens_seen": 176310450,
      "step": 8169,
      "time_per_iteration": 2.646388530731201
    },
    {
      "auxiliary_loss_clip": 0.01101166,
      "auxiliary_loss_mlp": 0.01024212,
      "balance_loss_clip": 1.03610098,
      "balance_loss_mlp": 1.01682115,
      "epoch": 0.9823844165213732,
      "flos": 22014462176640.0,
      "grad_norm": 1.6671323148999326,
      "language_loss": 0.76979601,
      "learning_rate": 3.232749028292847e-09,
      "loss": 0.79104978,
      "num_input_tokens_seen": 176327415,
      "step": 8170,
      "time_per_iteration": 3.6121017932891846
    },
    {
      "auxiliary_loss_clip": 0.01166394,
      "auxiliary_loss_mlp": 0.01023486,
      "balance_loss_clip": 1.04530311,
      "balance_loss_mlp": 1.0158025,
      "epoch": 0.9825046594120123,
      "flos": 21908166854400.0,
      "grad_norm": 2.146470327826388,
      "language_loss": 0.88312715,
      "learning_rate": 3.188628183992792e-09,
      "loss": 0.90502596,
      "num_input_tokens_seen": 176347680,
      "step": 8171,
      "time_per_iteration": 2.6168394088745117
    },
    {
      "auxiliary_loss_clip": 0.01054955,
      "auxiliary_loss_mlp": 0.01001964,
      "balance_loss_clip": 1.00927854,
      "balance_loss_mlp": 1.00093925,
      "epoch": 0.9826249023026513,
      "flos": 59494610718720.0,
      "grad_norm": 0.7336843186397584,
      "language_loss": 0.62459928,
      "learning_rate": 3.1448102543902844e-09,
      "loss": 0.64516848,
      "num_input_tokens_seen": 176411595,
      "step": 8172,
      "time_per_iteration": 4.111817121505737
    },
    {
      "auxiliary_loss_clip": 0.01132792,
      "auxiliary_loss_mlp": 0.01022126,
      "balance_loss_clip": 1.04299104,
      "balance_loss_mlp": 1.01517284,
      "epoch": 0.9827451451932905,
      "flos": 16071031296000.0,
      "grad_norm": 2.101722486819739,
      "language_loss": 0.67670482,
      "learning_rate": 3.1012952461324515e-09,
      "loss": 0.69825399,
      "num_input_tokens_seen": 176430570,
      "step": 8173,
      "time_per_iteration": 2.6133477687835693
    },
    {
      "auxiliary_loss_clip": 0.01147431,
      "auxiliary_loss_mlp": 0.01027757,
      "balance_loss_clip": 1.04527426,
      "balance_loss_mlp": 1.02082229,
      "epoch": 0.9828653880839295,
      "flos": 20262775622400.0,
      "grad_norm": 2.245755828588393,
      "language_loss": 0.73695624,
      "learning_rate": 3.0580831658204575e-09,
      "loss": 0.75870812,
      "num_input_tokens_seen": 176448150,
      "step": 8174,
      "time_per_iteration": 2.630075454711914
    },
    {
      "auxiliary_loss_clip": 0.01147295,
      "auxiliary_loss_mlp": 0.01019051,
      "balance_loss_clip": 1.04444718,
      "balance_loss_mlp": 1.01251793,
      "epoch": 0.9829856309745686,
      "flos": 21616141282560.0,
      "grad_norm": 1.6508148042219315,
      "language_loss": 0.77834862,
      "learning_rate": 3.015174020009281e-09,
      "loss": 0.80001211,
      "num_input_tokens_seen": 176467475,
      "step": 8175,
      "time_per_iteration": 2.6214518547058105
    },
    {
      "auxiliary_loss_clip": 0.01124273,
      "auxiliary_loss_mlp": 0.01031842,
      "balance_loss_clip": 1.03906798,
      "balance_loss_mlp": 1.02500272,
      "epoch": 0.9831058738652078,
      "flos": 23764209396480.0,
      "grad_norm": 2.0312714935722607,
      "language_loss": 0.74871171,
      "learning_rate": 2.9725678152086043e-09,
      "loss": 0.77027291,
      "num_input_tokens_seen": 176486045,
      "step": 8176,
      "time_per_iteration": 2.7512712478637695
    },
    {
      "auxiliary_loss_clip": 0.01126134,
      "auxiliary_loss_mlp": 0.0102973,
      "balance_loss_clip": 1.04046559,
      "balance_loss_mlp": 1.02238345,
      "epoch": 0.9832261167558468,
      "flos": 11320911072000.0,
      "grad_norm": 2.6214574966197413,
      "language_loss": 0.82106793,
      "learning_rate": 2.930264557881257e-09,
      "loss": 0.84262657,
      "num_input_tokens_seen": 176501230,
      "step": 8177,
      "time_per_iteration": 2.5971169471740723
    },
    {
      "auxiliary_loss_clip": 0.01063101,
      "auxiliary_loss_mlp": 0.01002485,
      "balance_loss_clip": 1.00931525,
      "balance_loss_mlp": 1.0014534,
      "epoch": 0.9833463596464859,
      "flos": 60000304343040.0,
      "grad_norm": 0.8355664303080811,
      "language_loss": 0.5815894,
      "learning_rate": 2.8882642544452163e-09,
      "loss": 0.60224527,
      "num_input_tokens_seen": 176565955,
      "step": 8178,
      "time_per_iteration": 3.1807310581207275
    },
    {
      "auxiliary_loss_clip": 0.01128343,
      "auxiliary_loss_mlp": 0.01021883,
      "balance_loss_clip": 1.04048455,
      "balance_loss_mlp": 1.01508164,
      "epoch": 0.983466602537125,
      "flos": 13626699805440.0,
      "grad_norm": 2.4619763312879477,
      "language_loss": 0.74798977,
      "learning_rate": 2.8465669112716083e-09,
      "loss": 0.76949197,
      "num_input_tokens_seen": 176583480,
      "step": 8179,
      "time_per_iteration": 2.6626791954040527
    },
    {
      "auxiliary_loss_clip": 0.01153368,
      "auxiliary_loss_mlp": 0.00762445,
      "balance_loss_clip": 1.04414117,
      "balance_loss_mlp": 1.00032139,
      "epoch": 0.9835868454277641,
      "flos": 22926844563840.0,
      "grad_norm": 2.1377485226672217,
      "language_loss": 0.76632738,
      "learning_rate": 2.8051725346858177e-09,
      "loss": 0.78548551,
      "num_input_tokens_seen": 176603740,
      "step": 8180,
      "time_per_iteration": 2.6797678470611572
    },
    {
      "auxiliary_loss_clip": 0.01164841,
      "auxiliary_loss_mlp": 0.0102717,
      "balance_loss_clip": 1.04305196,
      "balance_loss_mlp": 1.02005041,
      "epoch": 0.9837070883184031,
      "flos": 27673409341440.0,
      "grad_norm": 2.4013370722064864,
      "language_loss": 0.70984572,
      "learning_rate": 2.7640811309674883e-09,
      "loss": 0.73176587,
      "num_input_tokens_seen": 176623240,
      "step": 8181,
      "time_per_iteration": 2.6368725299835205
    },
    {
      "auxiliary_loss_clip": 0.01113293,
      "auxiliary_loss_mlp": 0.01021423,
      "balance_loss_clip": 1.04053319,
      "balance_loss_mlp": 1.0144875,
      "epoch": 0.9838273312090423,
      "flos": 29241951425280.0,
      "grad_norm": 1.5891921844419812,
      "language_loss": 0.81033969,
      "learning_rate": 2.7232927063498557e-09,
      "loss": 0.83168685,
      "num_input_tokens_seen": 176643615,
      "step": 8182,
      "time_per_iteration": 2.7740886211395264
    },
    {
      "auxiliary_loss_clip": 0.01150438,
      "auxiliary_loss_mlp": 0.01024916,
      "balance_loss_clip": 1.04337502,
      "balance_loss_mlp": 1.0178231,
      "epoch": 0.9839475740996814,
      "flos": 40110207304320.0,
      "grad_norm": 1.80264757106684,
      "language_loss": 0.68442631,
      "learning_rate": 2.682807267020859e-09,
      "loss": 0.70617986,
      "num_input_tokens_seen": 176666375,
      "step": 8183,
      "time_per_iteration": 2.7796945571899414
    },
    {
      "auxiliary_loss_clip": 0.01151021,
      "auxiliary_loss_mlp": 0.01020791,
      "balance_loss_clip": 1.04413521,
      "balance_loss_mlp": 1.01413631,
      "epoch": 0.9840678169903204,
      "flos": 24169389788160.0,
      "grad_norm": 1.6380351457562843,
      "language_loss": 0.62336528,
      "learning_rate": 2.642624819121808e-09,
      "loss": 0.64508337,
      "num_input_tokens_seen": 176686525,
      "step": 8184,
      "time_per_iteration": 2.6401853561401367
    },
    {
      "auxiliary_loss_clip": 0.01133665,
      "auxiliary_loss_mlp": 0.01024827,
      "balance_loss_clip": 1.04231811,
      "balance_loss_mlp": 1.01866651,
      "epoch": 0.9841880598809596,
      "flos": 14684484447360.0,
      "grad_norm": 2.1810914647554642,
      "language_loss": 0.61945438,
      "learning_rate": 2.6027453687487154e-09,
      "loss": 0.64103925,
      "num_input_tokens_seen": 176703615,
      "step": 8185,
      "time_per_iteration": 2.6136081218719482
    },
    {
      "auxiliary_loss_clip": 0.01137268,
      "auxiliary_loss_mlp": 0.01028476,
      "balance_loss_clip": 1.04200494,
      "balance_loss_mlp": 1.02113318,
      "epoch": 0.9843083027715986,
      "flos": 22344768668160.0,
      "grad_norm": 2.8251043799353033,
      "language_loss": 0.53834027,
      "learning_rate": 2.5631689219509643e-09,
      "loss": 0.55999768,
      "num_input_tokens_seen": 176722295,
      "step": 8186,
      "time_per_iteration": 2.6506192684173584
    },
    {
      "auxiliary_loss_clip": 0.01137055,
      "auxiliary_loss_mlp": 0.01022846,
      "balance_loss_clip": 1.04408693,
      "balance_loss_mlp": 1.01623893,
      "epoch": 0.9844285456622377,
      "flos": 21800111765760.0,
      "grad_norm": 1.7416033447823667,
      "language_loss": 0.83046269,
      "learning_rate": 2.523895484732197e-09,
      "loss": 0.85206163,
      "num_input_tokens_seen": 176741750,
      "step": 8187,
      "time_per_iteration": 2.6487839221954346
    },
    {
      "auxiliary_loss_clip": 0.01156342,
      "auxiliary_loss_mlp": 0.01024429,
      "balance_loss_clip": 1.0439806,
      "balance_loss_mlp": 1.01732397,
      "epoch": 0.9845487885528769,
      "flos": 18035380321920.0,
      "grad_norm": 1.8671000699746907,
      "language_loss": 0.7460767,
      "learning_rate": 2.4849250630505357e-09,
      "loss": 0.76788449,
      "num_input_tokens_seen": 176759995,
      "step": 8188,
      "time_per_iteration": 2.6001384258270264
    },
    {
      "auxiliary_loss_clip": 0.01070606,
      "auxiliary_loss_mlp": 0.01027332,
      "balance_loss_clip": 1.03725362,
      "balance_loss_mlp": 1.01961064,
      "epoch": 0.9846690314435159,
      "flos": 25228610974080.0,
      "grad_norm": 1.753641547483148,
      "language_loss": 0.73336017,
      "learning_rate": 2.4462576628172528e-09,
      "loss": 0.75433958,
      "num_input_tokens_seen": 176778625,
      "step": 8189,
      "time_per_iteration": 3.8289053440093994
    },
    {
      "auxiliary_loss_clip": 0.01147847,
      "auxiliary_loss_mlp": 0.01029227,
      "balance_loss_clip": 1.0447042,
      "balance_loss_mlp": 1.02221739,
      "epoch": 0.984789274334155,
      "flos": 18552171248640.0,
      "grad_norm": 1.825743406046752,
      "language_loss": 0.7415266,
      "learning_rate": 2.407893289898766e-09,
      "loss": 0.76329732,
      "num_input_tokens_seen": 176797655,
      "step": 8190,
      "time_per_iteration": 2.8211781978607178
    },
    {
      "auxiliary_loss_clip": 0.01115539,
      "auxiliary_loss_mlp": 0.01023922,
      "balance_loss_clip": 1.03878415,
      "balance_loss_mlp": 1.01629257,
      "epoch": 0.984909517224794,
      "flos": 27345437233920.0,
      "grad_norm": 1.8505153955887452,
      "language_loss": 0.84099972,
      "learning_rate": 2.3698319501144202e-09,
      "loss": 0.86239433,
      "num_input_tokens_seen": 176818640,
      "step": 8191,
      "time_per_iteration": 2.7914187908172607
    },
    {
      "auxiliary_loss_clip": 0.01156943,
      "auxiliary_loss_mlp": 0.01027377,
      "balance_loss_clip": 1.04443455,
      "balance_loss_mlp": 1.01961613,
      "epoch": 0.9850297601154332,
      "flos": 18734058743040.0,
      "grad_norm": 3.1157922905679327,
      "language_loss": 0.73227274,
      "learning_rate": 2.3320736492382644e-09,
      "loss": 0.75411594,
      "num_input_tokens_seen": 176837475,
      "step": 8192,
      "time_per_iteration": 2.6233291625976562
    },
    {
      "auxiliary_loss_clip": 0.01162346,
      "auxiliary_loss_mlp": 0.01022919,
      "balance_loss_clip": 1.04567409,
      "balance_loss_mlp": 1.01617765,
      "epoch": 0.9851500030060723,
      "flos": 22308247514880.0,
      "grad_norm": 1.6939747690078113,
      "language_loss": 0.67836297,
      "learning_rate": 2.29461839299816e-09,
      "loss": 0.7002157,
      "num_input_tokens_seen": 176857190,
      "step": 8193,
      "time_per_iteration": 3.517348527908325
    },
    {
      "auxiliary_loss_clip": 0.01125642,
      "auxiliary_loss_mlp": 0.01025409,
      "balance_loss_clip": 1.04257941,
      "balance_loss_mlp": 1.01834249,
      "epoch": 0.9852702458967113,
      "flos": 26353691746560.0,
      "grad_norm": 1.6351836323521844,
      "language_loss": 0.79852664,
      "learning_rate": 2.257466187076229e-09,
      "loss": 0.82003713,
      "num_input_tokens_seen": 176876395,
      "step": 8194,
      "time_per_iteration": 2.7107551097869873
    },
    {
      "auxiliary_loss_clip": 0.01154156,
      "auxiliary_loss_mlp": 0.00761897,
      "balance_loss_clip": 1.0427506,
      "balance_loss_mlp": 1.00034714,
      "epoch": 0.9853904887873505,
      "flos": 20883599314560.0,
      "grad_norm": 1.837531527125937,
      "language_loss": 0.7121501,
      "learning_rate": 2.2206170371081854e-09,
      "loss": 0.73131061,
      "num_input_tokens_seen": 176894980,
      "step": 8195,
      "time_per_iteration": 2.630932331085205
    },
    {
      "auxiliary_loss_clip": 0.0113714,
      "auxiliary_loss_mlp": 0.01030804,
      "balance_loss_clip": 1.04150319,
      "balance_loss_mlp": 1.0231384,
      "epoch": 0.9855107316779895,
      "flos": 25263444188160.0,
      "grad_norm": 1.6575777512702246,
      "language_loss": 0.84696949,
      "learning_rate": 2.1840709486842247e-09,
      "loss": 0.86864889,
      "num_input_tokens_seen": 176914600,
      "step": 8196,
      "time_per_iteration": 3.5751090049743652
    },
    {
      "auxiliary_loss_clip": 0.01131266,
      "auxiliary_loss_mlp": 0.01024056,
      "balance_loss_clip": 1.04220366,
      "balance_loss_mlp": 1.01661158,
      "epoch": 0.9856309745686286,
      "flos": 19062102677760.0,
      "grad_norm": 2.1287475084017884,
      "language_loss": 0.79313177,
      "learning_rate": 2.1478279273481335e-09,
      "loss": 0.81468499,
      "num_input_tokens_seen": 176933085,
      "step": 8197,
      "time_per_iteration": 2.663273572921753
    },
    {
      "auxiliary_loss_clip": 0.01151783,
      "auxiliary_loss_mlp": 0.01024342,
      "balance_loss_clip": 1.04651535,
      "balance_loss_mlp": 1.01712406,
      "epoch": 0.9857512174592677,
      "flos": 34130758060800.0,
      "grad_norm": 2.076350395259823,
      "language_loss": 0.80227679,
      "learning_rate": 2.1118879785981815e-09,
      "loss": 0.82403803,
      "num_input_tokens_seen": 176953225,
      "step": 8198,
      "time_per_iteration": 3.607318878173828
    },
    {
      "auxiliary_loss_clip": 0.01136031,
      "auxiliary_loss_mlp": 0.01023112,
      "balance_loss_clip": 1.04191542,
      "balance_loss_mlp": 1.01663589,
      "epoch": 0.9858714603499068,
      "flos": 25994693266560.0,
      "grad_norm": 1.7752136164443189,
      "language_loss": 0.7936942,
      "learning_rate": 2.0762511078862288e-09,
      "loss": 0.81528562,
      "num_input_tokens_seen": 176973570,
      "step": 8199,
      "time_per_iteration": 2.6537671089172363
    },
    {
      "auxiliary_loss_clip": 0.01144722,
      "auxiliary_loss_mlp": 0.01023872,
      "balance_loss_clip": 1.04306602,
      "balance_loss_mlp": 1.01662409,
      "epoch": 0.9859917032405459,
      "flos": 23696230907520.0,
      "grad_norm": 2.0730517718179033,
      "language_loss": 0.65106535,
      "learning_rate": 2.0409173206186183e-09,
      "loss": 0.67275125,
      "num_input_tokens_seen": 176992810,
      "step": 8200,
      "time_per_iteration": 2.6518001556396484
    },
    {
      "auxiliary_loss_clip": 0.01122173,
      "auxiliary_loss_mlp": 0.01026394,
      "balance_loss_clip": 1.0446595,
      "balance_loss_mlp": 1.0199213,
      "epoch": 0.986111946131185,
      "flos": 19938287134080.0,
      "grad_norm": 2.166288425391624,
      "language_loss": 0.87267649,
      "learning_rate": 2.0058866221550617e-09,
      "loss": 0.89416218,
      "num_input_tokens_seen": 177011050,
      "step": 8201,
      "time_per_iteration": 2.6962101459503174
    },
    {
      "auxiliary_loss_clip": 0.01163334,
      "auxiliary_loss_mlp": 0.01024525,
      "balance_loss_clip": 1.04418325,
      "balance_loss_mlp": 1.01772666,
      "epoch": 0.9862321890218241,
      "flos": 19828831415040.0,
      "grad_norm": 2.199917338205368,
      "language_loss": 0.75504541,
      "learning_rate": 1.971159017809976e-09,
      "loss": 0.77692401,
      "num_input_tokens_seen": 177029340,
      "step": 8202,
      "time_per_iteration": 2.63451886177063
    },
    {
      "auxiliary_loss_clip": 0.0115209,
      "auxiliary_loss_mlp": 0.01027011,
      "balance_loss_clip": 1.04489708,
      "balance_loss_mlp": 1.01950085,
      "epoch": 0.9863524319124631,
      "flos": 21652051904640.0,
      "grad_norm": 2.3903495524143863,
      "language_loss": 0.7777651,
      "learning_rate": 1.93673451285159e-09,
      "loss": 0.79955608,
      "num_input_tokens_seen": 177048390,
      "step": 8203,
      "time_per_iteration": 2.851266860961914
    },
    {
      "auxiliary_loss_clip": 0.01045756,
      "auxiliary_loss_mlp": 0.01001301,
      "balance_loss_clip": 1.00909233,
      "balance_loss_mlp": 1.00027561,
      "epoch": 0.9864726748031023,
      "flos": 52769977920000.0,
      "grad_norm": 0.7512770990135484,
      "language_loss": 0.56543112,
      "learning_rate": 1.9026131125019495e-09,
      "loss": 0.58590168,
      "num_input_tokens_seen": 177105760,
      "step": 8204,
      "time_per_iteration": 3.1936943531036377
    },
    {
      "auxiliary_loss_clip": 0.01148512,
      "auxiliary_loss_mlp": 0.01023263,
      "balance_loss_clip": 1.04599726,
      "balance_loss_mlp": 1.01626503,
      "epoch": 0.9865929176937414,
      "flos": 23364631526400.0,
      "grad_norm": 1.7230017627197731,
      "language_loss": 0.86866522,
      "learning_rate": 1.8687948219371363e-09,
      "loss": 0.89038295,
      "num_input_tokens_seen": 177124985,
      "step": 8205,
      "time_per_iteration": 2.6303586959838867
    },
    {
      "auxiliary_loss_clip": 0.01167442,
      "auxiliary_loss_mlp": 0.01026188,
      "balance_loss_clip": 1.0442344,
      "balance_loss_mlp": 1.01837969,
      "epoch": 0.9867131605843804,
      "flos": 21616679986560.0,
      "grad_norm": 2.2330630714545534,
      "language_loss": 0.88534141,
      "learning_rate": 1.835279646287491e-09,
      "loss": 0.9072777,
      "num_input_tokens_seen": 177142995,
      "step": 8206,
      "time_per_iteration": 2.62031626701355
    },
    {
      "auxiliary_loss_clip": 0.01157344,
      "auxiliary_loss_mlp": 0.01031288,
      "balance_loss_clip": 1.04581451,
      "balance_loss_mlp": 1.02272892,
      "epoch": 0.9868334034750196,
      "flos": 22271403139200.0,
      "grad_norm": 1.8556187601100032,
      "language_loss": 0.76604408,
      "learning_rate": 1.8020675906371685e-09,
      "loss": 0.78793037,
      "num_input_tokens_seen": 177162390,
      "step": 8207,
      "time_per_iteration": 2.6151347160339355
    },
    {
      "auxiliary_loss_clip": 0.01105063,
      "auxiliary_loss_mlp": 0.01024254,
      "balance_loss_clip": 1.03823864,
      "balance_loss_mlp": 1.01751316,
      "epoch": 0.9869536463656586,
      "flos": 25809573548160.0,
      "grad_norm": 2.0278918065458673,
      "language_loss": 0.75402498,
      "learning_rate": 1.7691586600243612e-09,
      "loss": 0.77531815,
      "num_input_tokens_seen": 177181290,
      "step": 8208,
      "time_per_iteration": 2.7702324390411377
    },
    {
      "auxiliary_loss_clip": 0.01134817,
      "auxiliary_loss_mlp": 0.01025637,
      "balance_loss_clip": 1.0446856,
      "balance_loss_mlp": 1.01866603,
      "epoch": 0.9870738892562977,
      "flos": 16398500613120.0,
      "grad_norm": 4.627438356642214,
      "language_loss": 0.87153566,
      "learning_rate": 1.7365528594415202e-09,
      "loss": 0.8931402,
      "num_input_tokens_seen": 177195360,
      "step": 8209,
      "time_per_iteration": 2.669318914413452
    },
    {
      "auxiliary_loss_clip": 0.01152573,
      "auxiliary_loss_mlp": 0.00762421,
      "balance_loss_clip": 1.04256511,
      "balance_loss_mlp": 1.00030255,
      "epoch": 0.9871941321469369,
      "flos": 35481358373760.0,
      "grad_norm": 1.7926468698917222,
      "language_loss": 0.67693591,
      "learning_rate": 1.7042501938346888e-09,
      "loss": 0.69608587,
      "num_input_tokens_seen": 177218090,
      "step": 8210,
      "time_per_iteration": 2.7942373752593994
    },
    {
      "auxiliary_loss_clip": 0.01125472,
      "auxiliary_loss_mlp": 0.01024814,
      "balance_loss_clip": 1.03876638,
      "balance_loss_mlp": 1.01835871,
      "epoch": 0.9873143750375759,
      "flos": 21434217874560.0,
      "grad_norm": 1.8447189468019352,
      "language_loss": 0.76804209,
      "learning_rate": 1.6722506681043913e-09,
      "loss": 0.78954494,
      "num_input_tokens_seen": 177237050,
      "step": 8211,
      "time_per_iteration": 2.6475439071655273
    },
    {
      "auxiliary_loss_clip": 0.01140667,
      "auxiliary_loss_mlp": 0.01023984,
      "balance_loss_clip": 1.04276061,
      "balance_loss_mlp": 1.01658678,
      "epoch": 0.987434617928215,
      "flos": 16326499800960.0,
      "grad_norm": 2.3209216894156564,
      "language_loss": 0.69450235,
      "learning_rate": 1.640554287104745e-09,
      "loss": 0.71614885,
      "num_input_tokens_seen": 177255325,
      "step": 8212,
      "time_per_iteration": 2.6391592025756836
    },
    {
      "auxiliary_loss_clip": 0.01125057,
      "auxiliary_loss_mlp": 0.01027118,
      "balance_loss_clip": 1.03778267,
      "balance_loss_mlp": 1.01933384,
      "epoch": 0.9875548608188541,
      "flos": 17851984456320.0,
      "grad_norm": 2.1623568561927775,
      "language_loss": 0.80417943,
      "learning_rate": 1.609161055644348e-09,
      "loss": 0.82570124,
      "num_input_tokens_seen": 177271250,
      "step": 8213,
      "time_per_iteration": 2.668924570083618
    },
    {
      "auxiliary_loss_clip": 0.01157055,
      "auxiliary_loss_mlp": 0.01023837,
      "balance_loss_clip": 1.0431571,
      "balance_loss_mlp": 1.01636565,
      "epoch": 0.9876751037094932,
      "flos": 26132876887680.0,
      "grad_norm": 2.0562683961233845,
      "language_loss": 0.68574506,
      "learning_rate": 1.5780709784849467e-09,
      "loss": 0.70755398,
      "num_input_tokens_seen": 177288270,
      "step": 8214,
      "time_per_iteration": 2.7031219005584717
    },
    {
      "auxiliary_loss_clip": 0.01101786,
      "auxiliary_loss_mlp": 0.01027768,
      "balance_loss_clip": 1.04431117,
      "balance_loss_mlp": 1.02084744,
      "epoch": 0.9877953466001322,
      "flos": 15991344973440.0,
      "grad_norm": 2.171285822249106,
      "language_loss": 0.82560349,
      "learning_rate": 1.5472840603436565e-09,
      "loss": 0.84689903,
      "num_input_tokens_seen": 177305500,
      "step": 8215,
      "time_per_iteration": 2.7734599113464355
    },
    {
      "auxiliary_loss_clip": 0.01139279,
      "auxiliary_loss_mlp": 0.01023407,
      "balance_loss_clip": 1.04151797,
      "balance_loss_mlp": 1.01646316,
      "epoch": 0.9879155894907714,
      "flos": 18806777827200.0,
      "grad_norm": 6.816878680382445,
      "language_loss": 0.78024602,
      "learning_rate": 1.5168003058900757e-09,
      "loss": 0.80187291,
      "num_input_tokens_seen": 177323500,
      "step": 8216,
      "time_per_iteration": 3.5101113319396973
    },
    {
      "auxiliary_loss_clip": 0.01123671,
      "auxiliary_loss_mlp": 0.01020638,
      "balance_loss_clip": 1.0419178,
      "balance_loss_mlp": 1.01369667,
      "epoch": 0.9880358323814105,
      "flos": 22382044007040.0,
      "grad_norm": 2.276018993755789,
      "language_loss": 0.91952837,
      "learning_rate": 1.4866197197491715e-09,
      "loss": 0.94097137,
      "num_input_tokens_seen": 177342860,
      "step": 8217,
      "time_per_iteration": 2.7531497478485107
    },
    {
      "auxiliary_loss_clip": 0.01156114,
      "auxiliary_loss_mlp": 0.00763098,
      "balance_loss_clip": 1.04544497,
      "balance_loss_mlp": 1.00037646,
      "epoch": 0.9881560752720495,
      "flos": 15668831733120.0,
      "grad_norm": 4.896192217562307,
      "language_loss": 0.78440309,
      "learning_rate": 1.4567423064988371e-09,
      "loss": 0.80359519,
      "num_input_tokens_seen": 177360210,
      "step": 8218,
      "time_per_iteration": 2.7097654342651367
    },
    {
      "auxiliary_loss_clip": 0.01167128,
      "auxiliary_loss_mlp": 0.01024742,
      "balance_loss_clip": 1.04564285,
      "balance_loss_mlp": 1.01701772,
      "epoch": 0.9882763181626887,
      "flos": 21500113374720.0,
      "grad_norm": 1.9461186630868361,
      "language_loss": 0.77925992,
      "learning_rate": 1.4271680706718913e-09,
      "loss": 0.80117863,
      "num_input_tokens_seen": 177377885,
      "step": 8219,
      "time_per_iteration": 3.6073474884033203
    },
    {
      "auxiliary_loss_clip": 0.01153535,
      "auxiliary_loss_mlp": 0.01027212,
      "balance_loss_clip": 1.04688251,
      "balance_loss_mlp": 1.01958859,
      "epoch": 0.9883965610533277,
      "flos": 28034598551040.0,
      "grad_norm": 1.6734866495682483,
      "language_loss": 0.82445961,
      "learning_rate": 1.3978970167543013e-09,
      "loss": 0.8462671,
      "num_input_tokens_seen": 177398065,
      "step": 8220,
      "time_per_iteration": 2.720426321029663
    },
    {
      "auxiliary_loss_clip": 0.01128921,
      "auxiliary_loss_mlp": 0.01025796,
      "balance_loss_clip": 1.04115772,
      "balance_loss_mlp": 1.01853919,
      "epoch": 0.9885168039439668,
      "flos": 14098601710080.0,
      "grad_norm": 2.1171537946094157,
      "language_loss": 0.77595997,
      "learning_rate": 1.3689291491867372e-09,
      "loss": 0.79750711,
      "num_input_tokens_seen": 177416380,
      "step": 8221,
      "time_per_iteration": 2.622058629989624
    },
    {
      "auxiliary_loss_clip": 0.01166721,
      "auxiliary_loss_mlp": 0.01022509,
      "balance_loss_clip": 1.04528093,
      "balance_loss_mlp": 1.01520097,
      "epoch": 0.988637046834606,
      "flos": 26432013352320.0,
      "grad_norm": 3.6757263638257474,
      "language_loss": 0.73436809,
      "learning_rate": 1.3402644723636836e-09,
      "loss": 0.7562604,
      "num_input_tokens_seen": 177438410,
      "step": 8222,
      "time_per_iteration": 3.504518985748291
    },
    {
      "auxiliary_loss_clip": 0.0113526,
      "auxiliary_loss_mlp": 0.0102954,
      "balance_loss_clip": 1.04471874,
      "balance_loss_mlp": 1.02202964,
      "epoch": 0.988757289725245,
      "flos": 25229113764480.0,
      "grad_norm": 1.9090439636361296,
      "language_loss": 0.83721244,
      "learning_rate": 1.311902990633218e-09,
      "loss": 0.85886043,
      "num_input_tokens_seen": 177457375,
      "step": 8223,
      "time_per_iteration": 2.761946678161621
    },
    {
      "auxiliary_loss_clip": 0.01128152,
      "auxiliary_loss_mlp": 0.01025538,
      "balance_loss_clip": 1.03753233,
      "balance_loss_mlp": 1.01861811,
      "epoch": 0.9888775326158841,
      "flos": 26359042872960.0,
      "grad_norm": 1.5250581634388247,
      "language_loss": 0.71540105,
      "learning_rate": 1.2838447082978987e-09,
      "loss": 0.73693794,
      "num_input_tokens_seen": 177478530,
      "step": 8224,
      "time_per_iteration": 3.571662187576294
    },
    {
      "auxiliary_loss_clip": 0.01146909,
      "auxiliary_loss_mlp": 0.01023435,
      "balance_loss_clip": 1.04236698,
      "balance_loss_mlp": 1.01668739,
      "epoch": 0.9889977755065231,
      "flos": 24316120846080.0,
      "grad_norm": 2.325418192375166,
      "language_loss": 0.83036387,
      "learning_rate": 1.2560896296143208e-09,
      "loss": 0.85206735,
      "num_input_tokens_seen": 177496995,
      "step": 8225,
      "time_per_iteration": 2.6467814445495605
    },
    {
      "auxiliary_loss_clip": 0.01164854,
      "auxiliary_loss_mlp": 0.01028327,
      "balance_loss_clip": 1.0443002,
      "balance_loss_mlp": 1.02158582,
      "epoch": 0.9891180183971623,
      "flos": 18951066760320.0,
      "grad_norm": 3.046580887460009,
      "language_loss": 0.82645762,
      "learning_rate": 1.2286377587926722e-09,
      "loss": 0.84838939,
      "num_input_tokens_seen": 177513785,
      "step": 8226,
      "time_per_iteration": 2.5863096714019775
    },
    {
      "auxiliary_loss_clip": 0.01163988,
      "auxiliary_loss_mlp": 0.01021356,
      "balance_loss_clip": 1.04430628,
      "balance_loss_mlp": 1.01431668,
      "epoch": 0.9892382612878013,
      "flos": 26176580760960.0,
      "grad_norm": 2.346190601250753,
      "language_loss": 0.75136042,
      "learning_rate": 1.2014890999973992e-09,
      "loss": 0.7732138,
      "num_input_tokens_seen": 177530705,
      "step": 8227,
      "time_per_iteration": 2.634845733642578
    },
    {
      "auxiliary_loss_clip": 0.01162054,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 1.0429554,
      "balance_loss_mlp": 1.02023685,
      "epoch": 0.9893585041784404,
      "flos": 25449605400960.0,
      "grad_norm": 1.6799168444075059,
      "language_loss": 0.78251827,
      "learning_rate": 1.1746436573472073e-09,
      "loss": 0.80440593,
      "num_input_tokens_seen": 177552440,
      "step": 8228,
      "time_per_iteration": 2.6243631839752197
    },
    {
      "auxiliary_loss_clip": 0.01146772,
      "auxiliary_loss_mlp": 0.01027599,
      "balance_loss_clip": 1.0433526,
      "balance_loss_mlp": 1.0202949,
      "epoch": 0.9894787470690796,
      "flos": 20189302352640.0,
      "grad_norm": 1.9837860376900223,
      "language_loss": 0.68930101,
      "learning_rate": 1.1481014349141726e-09,
      "loss": 0.71104473,
      "num_input_tokens_seen": 177569660,
      "step": 8229,
      "time_per_iteration": 2.641768217086792
    },
    {
      "auxiliary_loss_clip": 0.01141387,
      "auxiliary_loss_mlp": 0.01033633,
      "balance_loss_clip": 1.04423285,
      "balance_loss_mlp": 1.02593184,
      "epoch": 0.9895989899597186,
      "flos": 24644308435200.0,
      "grad_norm": 1.9965793650963444,
      "language_loss": 0.84310603,
      "learning_rate": 1.121862436724852e-09,
      "loss": 0.86485624,
      "num_input_tokens_seen": 177588500,
      "step": 8230,
      "time_per_iteration": 2.8094396591186523
    },
    {
      "auxiliary_loss_clip": 0.01153283,
      "auxiliary_loss_mlp": 0.01025287,
      "balance_loss_clip": 1.04657638,
      "balance_loss_mlp": 1.01826608,
      "epoch": 0.9897192328503577,
      "flos": 21799034357760.0,
      "grad_norm": 1.9215596031962985,
      "language_loss": 0.70307648,
      "learning_rate": 1.0959266667598388e-09,
      "loss": 0.72486222,
      "num_input_tokens_seen": 177607315,
      "step": 8231,
      "time_per_iteration": 2.6665995121002197
    },
    {
      "auxiliary_loss_clip": 0.01126102,
      "auxiliary_loss_mlp": 0.01027505,
      "balance_loss_clip": 1.04264259,
      "balance_loss_mlp": 1.01988769,
      "epoch": 0.9898394757409968,
      "flos": 21325229032320.0,
      "grad_norm": 1.9293956194215527,
      "language_loss": 0.74666214,
      "learning_rate": 1.0702941289533196e-09,
      "loss": 0.76819819,
      "num_input_tokens_seen": 177625990,
      "step": 8232,
      "time_per_iteration": 2.727536916732788
    },
    {
      "auxiliary_loss_clip": 0.01121775,
      "auxiliary_loss_mlp": 0.01021358,
      "balance_loss_clip": 1.04258728,
      "balance_loss_mlp": 1.01436639,
      "epoch": 0.9899597186316359,
      "flos": 18545024442240.0,
      "grad_norm": 1.931322662818884,
      "language_loss": 0.88878369,
      "learning_rate": 1.0449648271939615e-09,
      "loss": 0.91021502,
      "num_input_tokens_seen": 177642335,
      "step": 8233,
      "time_per_iteration": 2.660128116607666
    },
    {
      "auxiliary_loss_clip": 0.01109889,
      "auxiliary_loss_mlp": 0.00761709,
      "balance_loss_clip": 1.04158711,
      "balance_loss_mlp": 1.00037718,
      "epoch": 0.990079961522275,
      "flos": 23766723348480.0,
      "grad_norm": 1.4934013580916503,
      "language_loss": 0.7304132,
      "learning_rate": 1.0199387653240243e-09,
      "loss": 0.74912918,
      "num_input_tokens_seen": 177662025,
      "step": 8234,
      "time_per_iteration": 2.7905023097991943
    },
    {
      "auxiliary_loss_clip": 0.01132325,
      "auxiliary_loss_mlp": 0.01023837,
      "balance_loss_clip": 1.04399872,
      "balance_loss_mlp": 1.01764727,
      "epoch": 0.9902002044129141,
      "flos": 16399182971520.0,
      "grad_norm": 1.6670187027437622,
      "language_loss": 0.70620829,
      "learning_rate": 9.952159471400267e-10,
      "loss": 0.72776991,
      "num_input_tokens_seen": 177679065,
      "step": 8235,
      "time_per_iteration": 2.815988540649414
    },
    {
      "auxiliary_loss_clip": 0.0115364,
      "auxiliary_loss_mlp": 0.00761379,
      "balance_loss_clip": 1.04367268,
      "balance_loss_mlp": 1.0003705,
      "epoch": 0.9903204473035532,
      "flos": 22559657783040.0,
      "grad_norm": 1.9161639801957944,
      "language_loss": 0.84601307,
      "learning_rate": 9.707963763923022e-10,
      "loss": 0.86516333,
      "num_input_tokens_seen": 177698115,
      "step": 8236,
      "time_per_iteration": 2.7349956035614014
    },
    {
      "auxiliary_loss_clip": 0.01134214,
      "auxiliary_loss_mlp": 0.01024552,
      "balance_loss_clip": 1.03999233,
      "balance_loss_mlp": 1.01772976,
      "epoch": 0.9904406901941922,
      "flos": 16144001775360.0,
      "grad_norm": 1.9246566794650046,
      "language_loss": 0.79188627,
      "learning_rate": 9.466800567854427e-10,
      "loss": 0.81347394,
      "num_input_tokens_seen": 177716715,
      "step": 8237,
      "time_per_iteration": 2.6396920680999756
    },
    {
      "auxiliary_loss_clip": 0.0112262,
      "auxiliary_loss_mlp": 0.01027196,
      "balance_loss_clip": 1.03826654,
      "balance_loss_mlp": 1.0193553,
      "epoch": 0.9905609330848314,
      "flos": 26651499408000.0,
      "grad_norm": 1.8180485382088207,
      "language_loss": 0.68299592,
      "learning_rate": 9.228669919778553e-10,
      "loss": 0.70449406,
      "num_input_tokens_seen": 177735640,
      "step": 8238,
      "time_per_iteration": 2.6910207271575928
    },
    {
      "auxiliary_loss_clip": 0.01134263,
      "auxiliary_loss_mlp": 0.01032566,
      "balance_loss_clip": 1.04250503,
      "balance_loss_mlp": 1.02506471,
      "epoch": 0.9906811759754705,
      "flos": 23111820627840.0,
      "grad_norm": 2.053715609679284,
      "language_loss": 0.79871535,
      "learning_rate": 8.993571855817617e-10,
      "loss": 0.82038367,
      "num_input_tokens_seen": 177754470,
      "step": 8239,
      "time_per_iteration": 2.6386706829071045
    },
    {
      "auxiliary_loss_clip": 0.01148978,
      "auxiliary_loss_mlp": 0.010257,
      "balance_loss_clip": 1.04169559,
      "balance_loss_mlp": 1.01863694,
      "epoch": 0.9908014188661095,
      "flos": 22090593052800.0,
      "grad_norm": 1.7821081644255254,
      "language_loss": 0.74804413,
      "learning_rate": 8.761506411638642e-10,
      "loss": 0.76979095,
      "num_input_tokens_seen": 177773935,
      "step": 8240,
      "time_per_iteration": 2.5869617462158203
    },
    {
      "auxiliary_loss_clip": 0.01134169,
      "auxiliary_loss_mlp": 0.01024073,
      "balance_loss_clip": 1.04379582,
      "balance_loss_mlp": 1.0168519,
      "epoch": 0.9909216617567487,
      "flos": 19242948677760.0,
      "grad_norm": 1.7230859344297593,
      "language_loss": 0.73569053,
      "learning_rate": 8.53247362244236e-10,
      "loss": 0.7572729,
      "num_input_tokens_seen": 177792745,
      "step": 8241,
      "time_per_iteration": 2.723297119140625
    },
    {
      "auxiliary_loss_clip": 0.0113698,
      "auxiliary_loss_mlp": 0.01029711,
      "balance_loss_clip": 1.04325807,
      "balance_loss_mlp": 1.02288365,
      "epoch": 0.9910419046473877,
      "flos": 23621213352960.0,
      "grad_norm": 1.7275401500072267,
      "language_loss": 0.68030429,
      "learning_rate": 8.306473522976532e-10,
      "loss": 0.70197117,
      "num_input_tokens_seen": 177812150,
      "step": 8242,
      "time_per_iteration": 3.5957937240600586
    },
    {
      "auxiliary_loss_clip": 0.01164243,
      "auxiliary_loss_mlp": 0.01024723,
      "balance_loss_clip": 1.04545462,
      "balance_loss_mlp": 1.01810122,
      "epoch": 0.9911621475380268,
      "flos": 22711380831360.0,
      "grad_norm": 1.785503892722512,
      "language_loss": 0.71853459,
      "learning_rate": 8.083506147522623e-10,
      "loss": 0.74042422,
      "num_input_tokens_seen": 177831545,
      "step": 8243,
      "time_per_iteration": 2.665158271789551
    },
    {
      "auxiliary_loss_clip": 0.01145129,
      "auxiliary_loss_mlp": 0.01025562,
      "balance_loss_clip": 1.04214406,
      "balance_loss_mlp": 1.01822448,
      "epoch": 0.991282390428666,
      "flos": 13516956777600.0,
      "grad_norm": 2.2884752447212957,
      "language_loss": 0.84966099,
      "learning_rate": 7.863571529906909e-10,
      "loss": 0.87136793,
      "num_input_tokens_seen": 177847130,
      "step": 8244,
      "time_per_iteration": 2.730184316635132
    },
    {
      "auxiliary_loss_clip": 0.01054069,
      "auxiliary_loss_mlp": 0.0100104,
      "balance_loss_clip": 1.00949264,
      "balance_loss_mlp": 0.99997312,
      "epoch": 0.991402633319305,
      "flos": 61830492071040.0,
      "grad_norm": 0.7263403823788498,
      "language_loss": 0.59721565,
      "learning_rate": 7.646669703489372e-10,
      "loss": 0.61776674,
      "num_input_tokens_seen": 177911440,
      "step": 8245,
      "time_per_iteration": 4.293559789657593
    },
    {
      "auxiliary_loss_clip": 0.01059886,
      "auxiliary_loss_mlp": 0.01025469,
      "balance_loss_clip": 1.03371191,
      "balance_loss_mlp": 1.01849222,
      "epoch": 0.9915228762099441,
      "flos": 18770148933120.0,
      "grad_norm": 2.0980225612717045,
      "language_loss": 0.57417083,
      "learning_rate": 7.432800701177023e-10,
      "loss": 0.59502435,
      "num_input_tokens_seen": 177929440,
      "step": 8246,
      "time_per_iteration": 3.319733142852783
    },
    {
      "auxiliary_loss_clip": 0.01045886,
      "auxiliary_loss_mlp": 0.01002611,
      "balance_loss_clip": 1.01066363,
      "balance_loss_mlp": 1.00157976,
      "epoch": 0.9916431191005832,
      "flos": 65936660244480.0,
      "grad_norm": 0.7926023061662805,
      "language_loss": 0.57773662,
      "learning_rate": 7.221964555415017e-10,
      "loss": 0.59822154,
      "num_input_tokens_seen": 177989100,
      "step": 8247,
      "time_per_iteration": 3.3543989658355713
    },
    {
      "auxiliary_loss_clip": 0.0113477,
      "auxiliary_loss_mlp": 0.01020832,
      "balance_loss_clip": 1.04307318,
      "balance_loss_mlp": 1.0146066,
      "epoch": 0.9917633619912223,
      "flos": 16581573256320.0,
      "grad_norm": 2.041771259741335,
      "language_loss": 0.74865007,
      "learning_rate": 7.01416129818222e-10,
      "loss": 0.77020609,
      "num_input_tokens_seen": 178006720,
      "step": 8248,
      "time_per_iteration": 3.619908571243286
    },
    {
      "auxiliary_loss_clip": 0.01131546,
      "auxiliary_loss_mlp": 0.01026558,
      "balance_loss_clip": 1.04182124,
      "balance_loss_mlp": 1.01999235,
      "epoch": 0.9918836048818613,
      "flos": 25411108999680.0,
      "grad_norm": 2.120926805841478,
      "language_loss": 0.57963955,
      "learning_rate": 6.809390961006745e-10,
      "loss": 0.60122061,
      "num_input_tokens_seen": 178026850,
      "step": 8249,
      "time_per_iteration": 2.8643739223480225
    },
    {
      "auxiliary_loss_clip": 0.01137593,
      "auxiliary_loss_mlp": 0.01025376,
      "balance_loss_clip": 1.04354298,
      "balance_loss_mlp": 1.01818728,
      "epoch": 0.9920038477725005,
      "flos": 25046867134080.0,
      "grad_norm": 2.153785593934072,
      "language_loss": 0.68529844,
      "learning_rate": 6.607653574948191e-10,
      "loss": 0.70692813,
      "num_input_tokens_seen": 178047630,
      "step": 8250,
      "time_per_iteration": 3.6667323112487793
    },
    {
      "auxiliary_loss_clip": 0.011419,
      "auxiliary_loss_mlp": 0.01023746,
      "balance_loss_clip": 1.04051471,
      "balance_loss_mlp": 1.01735651,
      "epoch": 0.9921240906631396,
      "flos": 21829773421440.0,
      "grad_norm": 2.449558430084373,
      "language_loss": 0.81838977,
      "learning_rate": 6.408949170613187e-10,
      "loss": 0.84004617,
      "num_input_tokens_seen": 178066895,
      "step": 8251,
      "time_per_iteration": 2.6726372241973877
    },
    {
      "auxiliary_loss_clip": 0.01135575,
      "auxiliary_loss_mlp": 0.0102886,
      "balance_loss_clip": 1.04236174,
      "balance_loss_mlp": 1.0212965,
      "epoch": 0.9922443335537786,
      "flos": 24864225454080.0,
      "grad_norm": 29.299835320607137,
      "language_loss": 0.81621772,
      "learning_rate": 6.213277778144288e-10,
      "loss": 0.83786201,
      "num_input_tokens_seen": 178088540,
      "step": 8252,
      "time_per_iteration": 2.7124154567718506
    },
    {
      "auxiliary_loss_clip": 0.01096148,
      "auxiliary_loss_mlp": 0.01026666,
      "balance_loss_clip": 1.0359776,
      "balance_loss_mlp": 1.01901305,
      "epoch": 0.9923645764444178,
      "flos": 21613088626560.0,
      "grad_norm": 1.93972722094967,
      "language_loss": 0.66734904,
      "learning_rate": 6.020639427224416e-10,
      "loss": 0.68857718,
      "num_input_tokens_seen": 178106185,
      "step": 8253,
      "time_per_iteration": 2.762369155883789
    },
    {
      "auxiliary_loss_clip": 0.01137798,
      "auxiliary_loss_mlp": 0.01026346,
      "balance_loss_clip": 1.04321635,
      "balance_loss_mlp": 1.01902342,
      "epoch": 0.9924848193350568,
      "flos": 25001798544000.0,
      "grad_norm": 2.6326052591299764,
      "language_loss": 0.72553515,
      "learning_rate": 5.831034147076864e-10,
      "loss": 0.74717653,
      "num_input_tokens_seen": 178123435,
      "step": 8254,
      "time_per_iteration": 2.696176528930664
    },
    {
      "auxiliary_loss_clip": 0.01049001,
      "auxiliary_loss_mlp": 0.01000233,
      "balance_loss_clip": 1.00892162,
      "balance_loss_mlp": 0.99922603,
      "epoch": 0.9926050622256959,
      "flos": 68912543151360.0,
      "grad_norm": 0.6859192240318006,
      "language_loss": 0.55744421,
      "learning_rate": 5.644461966463065e-10,
      "loss": 0.57793653,
      "num_input_tokens_seen": 178191045,
      "step": 8255,
      "time_per_iteration": 3.2910313606262207
    },
    {
      "auxiliary_loss_clip": 0.01134806,
      "auxiliary_loss_mlp": 0.01021221,
      "balance_loss_clip": 1.04268408,
      "balance_loss_mlp": 1.01451612,
      "epoch": 0.9927253051163349,
      "flos": 20923675914240.0,
      "grad_norm": 1.8596646511894117,
      "language_loss": 0.75850368,
      "learning_rate": 5.460922913687049e-10,
      "loss": 0.78006393,
      "num_input_tokens_seen": 178210135,
      "step": 8256,
      "time_per_iteration": 2.673096179962158
    },
    {
      "auxiliary_loss_clip": 0.01105439,
      "auxiliary_loss_mlp": 0.0076254,
      "balance_loss_clip": 1.03698802,
      "balance_loss_mlp": 1.00034523,
      "epoch": 0.9928455480069741,
      "flos": 22308211601280.0,
      "grad_norm": 2.527157677597883,
      "language_loss": 0.75494218,
      "learning_rate": 5.280417016593208e-10,
      "loss": 0.77362198,
      "num_input_tokens_seen": 178229925,
      "step": 8257,
      "time_per_iteration": 2.910029172897339
    },
    {
      "auxiliary_loss_clip": 0.01149994,
      "auxiliary_loss_mlp": 0.00762291,
      "balance_loss_clip": 1.04601192,
      "balance_loss_mlp": 1.00041211,
      "epoch": 0.9929657908976132,
      "flos": 17383889393280.0,
      "grad_norm": 1.8309486992685582,
      "language_loss": 0.74487162,
      "learning_rate": 5.102944302559642e-10,
      "loss": 0.76399446,
      "num_input_tokens_seen": 178247420,
      "step": 8258,
      "time_per_iteration": 2.6656014919281006
    },
    {
      "auxiliary_loss_clip": 0.0110143,
      "auxiliary_loss_mlp": 0.01026072,
      "balance_loss_clip": 1.03971851,
      "balance_loss_mlp": 1.01803088,
      "epoch": 0.9930860337882522,
      "flos": 22674680110080.0,
      "grad_norm": 1.9593407122295374,
      "language_loss": 0.8016333,
      "learning_rate": 4.9285047985137e-10,
      "loss": 0.82290834,
      "num_input_tokens_seen": 178266840,
      "step": 8259,
      "time_per_iteration": 2.807893991470337
    },
    {
      "auxiliary_loss_clip": 0.01156592,
      "auxiliary_loss_mlp": 0.01025392,
      "balance_loss_clip": 1.04626274,
      "balance_loss_mlp": 1.01813185,
      "epoch": 0.9932062766788914,
      "flos": 28147789284480.0,
      "grad_norm": 1.7229196818811334,
      "language_loss": 0.74588919,
      "learning_rate": 4.757098530916436e-10,
      "loss": 0.76770902,
      "num_input_tokens_seen": 178287285,
      "step": 8260,
      "time_per_iteration": 2.6834499835968018
    },
    {
      "auxiliary_loss_clip": 0.01153565,
      "auxiliary_loss_mlp": 0.01028891,
      "balance_loss_clip": 1.04600394,
      "balance_loss_mlp": 1.02126169,
      "epoch": 0.9933265195695304,
      "flos": 20156659868160.0,
      "grad_norm": 3.156397996066343,
      "language_loss": 0.77526313,
      "learning_rate": 4.5887255257670563e-10,
      "loss": 0.79708767,
      "num_input_tokens_seen": 178304325,
      "step": 8261,
      "time_per_iteration": 2.616689920425415
    },
    {
      "auxiliary_loss_clip": 0.01165061,
      "auxiliary_loss_mlp": 0.01022567,
      "balance_loss_clip": 1.04495239,
      "balance_loss_mlp": 1.0156436,
      "epoch": 0.9934467624601695,
      "flos": 21362037494400.0,
      "grad_norm": 1.8677009051034443,
      "language_loss": 0.77351236,
      "learning_rate": 4.4233858086117906e-10,
      "loss": 0.79538864,
      "num_input_tokens_seen": 178322850,
      "step": 8262,
      "time_per_iteration": 2.6168174743652344
    },
    {
      "auxiliary_loss_clip": 0.01107739,
      "auxiliary_loss_mlp": 0.0102385,
      "balance_loss_clip": 1.04491675,
      "balance_loss_mlp": 1.0168643,
      "epoch": 0.9935670053508087,
      "flos": 19756040503680.0,
      "grad_norm": 2.5213051479223516,
      "language_loss": 0.67453682,
      "learning_rate": 4.261079404528356e-10,
      "loss": 0.69585264,
      "num_input_tokens_seen": 178342330,
      "step": 8263,
      "time_per_iteration": 2.7505593299865723
    },
    {
      "auxiliary_loss_clip": 0.01147809,
      "auxiliary_loss_mlp": 0.01031693,
      "balance_loss_clip": 1.0431385,
      "balance_loss_mlp": 1.02382505,
      "epoch": 0.9936872482414477,
      "flos": 21978838863360.0,
      "grad_norm": 1.9496254205845174,
      "language_loss": 0.68788922,
      "learning_rate": 4.1018063381437205e-10,
      "loss": 0.70968425,
      "num_input_tokens_seen": 178362715,
      "step": 8264,
      "time_per_iteration": 2.6474568843841553
    },
    {
      "auxiliary_loss_clip": 0.01051548,
      "auxiliary_loss_mlp": 0.01001446,
      "balance_loss_clip": 1.01301003,
      "balance_loss_mlp": 1.00039661,
      "epoch": 0.9938074911320868,
      "flos": 69810667839360.0,
      "grad_norm": 0.8608330842132922,
      "language_loss": 0.61067891,
      "learning_rate": 3.9455666336141167e-10,
      "loss": 0.6312089,
      "num_input_tokens_seen": 178426495,
      "step": 8265,
      "time_per_iteration": 3.2728450298309326
    },
    {
      "auxiliary_loss_clip": 0.01164398,
      "auxiliary_loss_mlp": 0.01027079,
      "balance_loss_clip": 1.04628491,
      "balance_loss_mlp": 1.01956856,
      "epoch": 0.9939277340227259,
      "flos": 15084170058240.0,
      "grad_norm": 3.1600897682490494,
      "language_loss": 0.83566642,
      "learning_rate": 3.7923603146450267e-10,
      "loss": 0.85758114,
      "num_input_tokens_seen": 178442555,
      "step": 8266,
      "time_per_iteration": 2.5377769470214844
    },
    {
      "auxiliary_loss_clip": 0.01124173,
      "auxiliary_loss_mlp": 0.01022877,
      "balance_loss_clip": 1.0385989,
      "balance_loss_mlp": 1.01601672,
      "epoch": 0.994047976913365,
      "flos": 17712364291200.0,
      "grad_norm": 1.8197416043191126,
      "language_loss": 0.80693406,
      "learning_rate": 3.642187404473418e-10,
      "loss": 0.82840455,
      "num_input_tokens_seen": 178460715,
      "step": 8267,
      "time_per_iteration": 2.687286853790283
    },
    {
      "auxiliary_loss_clip": 0.01152293,
      "auxiliary_loss_mlp": 0.01024052,
      "balance_loss_clip": 1.04391623,
      "balance_loss_mlp": 1.01697695,
      "epoch": 0.994168219804004,
      "flos": 19171558396800.0,
      "grad_norm": 2.13717424866532,
      "language_loss": 0.86103696,
      "learning_rate": 3.495047925885508e-10,
      "loss": 0.8828004,
      "num_input_tokens_seen": 178479050,
      "step": 8268,
      "time_per_iteration": 3.6111183166503906
    },
    {
      "auxiliary_loss_clip": 0.01134464,
      "auxiliary_loss_mlp": 0.0102537,
      "balance_loss_clip": 1.04131174,
      "balance_loss_mlp": 1.01779127,
      "epoch": 0.9942884626946432,
      "flos": 17851589406720.0,
      "grad_norm": 2.1821276831131646,
      "language_loss": 0.82758796,
      "learning_rate": 3.350941901199e-10,
      "loss": 0.8491863,
      "num_input_tokens_seen": 178495970,
      "step": 8269,
      "time_per_iteration": 2.6740124225616455
    },
    {
      "auxiliary_loss_clip": 0.0114051,
      "auxiliary_loss_mlp": 0.01027111,
      "balance_loss_clip": 1.04167426,
      "balance_loss_mlp": 1.01942229,
      "epoch": 0.9944087055852823,
      "flos": 18796578364800.0,
      "grad_norm": 2.3058357555644986,
      "language_loss": 0.83707488,
      "learning_rate": 3.2098693522764066e-10,
      "loss": 0.85875106,
      "num_input_tokens_seen": 178509170,
      "step": 8270,
      "time_per_iteration": 2.679692268371582
    },
    {
      "auxiliary_loss_clip": 0.01143565,
      "auxiliary_loss_mlp": 0.00762174,
      "balance_loss_clip": 1.04250193,
      "balance_loss_mlp": 1.0003643,
      "epoch": 0.9945289484759213,
      "flos": 20996969616000.0,
      "grad_norm": 1.8105039846711797,
      "language_loss": 0.81174302,
      "learning_rate": 3.071830300516165e-10,
      "loss": 0.83080041,
      "num_input_tokens_seen": 178527000,
      "step": 8271,
      "time_per_iteration": 3.6483724117279053
    },
    {
      "auxiliary_loss_clip": 0.01156319,
      "auxiliary_loss_mlp": 0.01028815,
      "balance_loss_clip": 1.04430127,
      "balance_loss_mlp": 1.0209533,
      "epoch": 0.9946491913665605,
      "flos": 14756952136320.0,
      "grad_norm": 2.039092436970155,
      "language_loss": 0.70821285,
      "learning_rate": 2.9368247668615234e-10,
      "loss": 0.73006421,
      "num_input_tokens_seen": 178545590,
      "step": 8272,
      "time_per_iteration": 2.64585280418396
    },
    {
      "auxiliary_loss_clip": 0.01171628,
      "auxiliary_loss_mlp": 0.01026373,
      "balance_loss_clip": 1.04854131,
      "balance_loss_mlp": 1.01874709,
      "epoch": 0.9947694342571995,
      "flos": 12669931186560.0,
      "grad_norm": 2.0213017151741814,
      "language_loss": 0.61117983,
      "learning_rate": 2.804852771789434e-10,
      "loss": 0.63315988,
      "num_input_tokens_seen": 178558890,
      "step": 8273,
      "time_per_iteration": 2.5425117015838623
    },
    {
      "auxiliary_loss_clip": 0.01161782,
      "auxiliary_loss_mlp": 0.01020662,
      "balance_loss_clip": 1.04429448,
      "balance_loss_mlp": 1.01441288,
      "epoch": 0.9948896771478386,
      "flos": 18843442634880.0,
      "grad_norm": 1.67259591540126,
      "language_loss": 0.55985832,
      "learning_rate": 2.675914335321661e-10,
      "loss": 0.58168274,
      "num_input_tokens_seen": 178577645,
      "step": 8274,
      "time_per_iteration": 3.6576621532440186
    },
    {
      "auxiliary_loss_clip": 0.01154575,
      "auxiliary_loss_mlp": 0.01030338,
      "balance_loss_clip": 1.04412675,
      "balance_loss_mlp": 1.02201152,
      "epoch": 0.9950099200384778,
      "flos": 24900207903360.0,
      "grad_norm": 2.272377143364523,
      "language_loss": 0.7940824,
      "learning_rate": 2.550009477018111e-10,
      "loss": 0.81593156,
      "num_input_tokens_seen": 178596415,
      "step": 8275,
      "time_per_iteration": 2.654515266418457
    },
    {
      "auxiliary_loss_clip": 0.01135923,
      "auxiliary_loss_mlp": 0.00762318,
      "balance_loss_clip": 1.04284811,
      "balance_loss_mlp": 1.00036311,
      "epoch": 0.9951301629291168,
      "flos": 23733613987200.0,
      "grad_norm": 1.9078164531823198,
      "language_loss": 0.63270736,
      "learning_rate": 2.4271382159790634e-10,
      "loss": 0.65168977,
      "num_input_tokens_seen": 178613845,
      "step": 8276,
      "time_per_iteration": 3.6025702953338623
    },
    {
      "auxiliary_loss_clip": 0.01106215,
      "auxiliary_loss_mlp": 0.0102173,
      "balance_loss_clip": 1.03911185,
      "balance_loss_mlp": 1.01425266,
      "epoch": 0.9952504058197559,
      "flos": 22236893147520.0,
      "grad_norm": 1.7931669125561356,
      "language_loss": 0.85946321,
      "learning_rate": 2.3073005708429406e-10,
      "loss": 0.88074267,
      "num_input_tokens_seen": 178633490,
      "step": 8277,
      "time_per_iteration": 2.788496494293213
    },
    {
      "auxiliary_loss_clip": 0.01119962,
      "auxiliary_loss_mlp": 0.01025521,
      "balance_loss_clip": 1.04342961,
      "balance_loss_mlp": 1.01919663,
      "epoch": 0.995370648710395,
      "flos": 21211032718080.0,
      "grad_norm": 1.931039701810572,
      "language_loss": 0.72466695,
      "learning_rate": 2.190496559788535e-10,
      "loss": 0.74612182,
      "num_input_tokens_seen": 178651775,
      "step": 8278,
      "time_per_iteration": 2.711463451385498
    },
    {
      "auxiliary_loss_clip": 0.01134942,
      "auxiliary_loss_mlp": 0.01024952,
      "balance_loss_clip": 1.04224014,
      "balance_loss_mlp": 1.01779366,
      "epoch": 0.9954908916010341,
      "flos": 14866731077760.0,
      "grad_norm": 5.34408705512711,
      "language_loss": 0.76626676,
      "learning_rate": 2.0767262005372265e-10,
      "loss": 0.7878657,
      "num_input_tokens_seen": 178669290,
      "step": 8279,
      "time_per_iteration": 2.7115743160247803
    },
    {
      "auxiliary_loss_clip": 0.01128405,
      "auxiliary_loss_mlp": 0.01025758,
      "balance_loss_clip": 1.03960705,
      "balance_loss_mlp": 1.01854587,
      "epoch": 0.9956111344916732,
      "flos": 19208259118080.0,
      "grad_norm": 2.4513972587133654,
      "language_loss": 0.7531454,
      "learning_rate": 1.965989510346322e-10,
      "loss": 0.77468705,
      "num_input_tokens_seen": 178688410,
      "step": 8280,
      "time_per_iteration": 2.879134178161621
    },
    {
      "auxiliary_loss_clip": 0.01103891,
      "auxiliary_loss_mlp": 0.01025001,
      "balance_loss_clip": 1.03892517,
      "balance_loss_mlp": 1.01763093,
      "epoch": 0.9957313773823123,
      "flos": 20047060494720.0,
      "grad_norm": 2.0250146487167013,
      "language_loss": 0.71266687,
      "learning_rate": 1.8582865060134955e-10,
      "loss": 0.7339558,
      "num_input_tokens_seen": 178706600,
      "step": 8281,
      "time_per_iteration": 2.704165458679199
    },
    {
      "auxiliary_loss_clip": 0.01062851,
      "auxiliary_loss_mlp": 0.01000859,
      "balance_loss_clip": 1.00910902,
      "balance_loss_mlp": 0.99982828,
      "epoch": 0.9958516202729514,
      "flos": 57483253768320.0,
      "grad_norm": 0.776159037872063,
      "language_loss": 0.55714875,
      "learning_rate": 1.7536172038790098e-10,
      "loss": 0.57778585,
      "num_input_tokens_seen": 178766910,
      "step": 8282,
      "time_per_iteration": 3.252211570739746
    },
    {
      "auxiliary_loss_clip": 0.01136756,
      "auxiliary_loss_mlp": 0.01021768,
      "balance_loss_clip": 1.04288018,
      "balance_loss_mlp": 1.01506817,
      "epoch": 0.9959718631635904,
      "flos": 27782900974080.0,
      "grad_norm": 2.515770258635666,
      "language_loss": 0.69169611,
      "learning_rate": 1.651981619819054e-10,
      "loss": 0.71328133,
      "num_input_tokens_seen": 178784060,
      "step": 8283,
      "time_per_iteration": 2.7074215412139893
    },
    {
      "auxiliary_loss_clip": 0.01111898,
      "auxiliary_loss_mlp": 0.01023927,
      "balance_loss_clip": 1.04071546,
      "balance_loss_mlp": 1.01692355,
      "epoch": 0.9960921060542296,
      "flos": 24024095274240.0,
      "grad_norm": 2.306204724408792,
      "language_loss": 0.70557344,
      "learning_rate": 1.5533797692546257e-10,
      "loss": 0.72693169,
      "num_input_tokens_seen": 178802795,
      "step": 8284,
      "time_per_iteration": 2.7203524112701416
    },
    {
      "auxiliary_loss_clip": 0.01146414,
      "auxiliary_loss_mlp": 0.0102722,
      "balance_loss_clip": 1.04125977,
      "balance_loss_mlp": 1.01942933,
      "epoch": 0.9962123489448687,
      "flos": 18697393935360.0,
      "grad_norm": 2.368049033661081,
      "language_loss": 0.84500968,
      "learning_rate": 1.4578116671404296e-10,
      "loss": 0.86674601,
      "num_input_tokens_seen": 178821075,
      "step": 8285,
      "time_per_iteration": 2.6732044219970703
    },
    {
      "auxiliary_loss_clip": 0.01149426,
      "auxiliary_loss_mlp": 0.01022052,
      "balance_loss_clip": 1.04688859,
      "balance_loss_mlp": 1.01529324,
      "epoch": 0.9963325918355077,
      "flos": 20010754823040.0,
      "grad_norm": 2.3684620748172804,
      "language_loss": 0.71088624,
      "learning_rate": 1.3652773279759777e-10,
      "loss": 0.73260111,
      "num_input_tokens_seen": 178837725,
      "step": 8286,
      "time_per_iteration": 2.6329290866851807
    },
    {
      "auxiliary_loss_clip": 0.01151097,
      "auxiliary_loss_mlp": 0.01022907,
      "balance_loss_clip": 1.04354715,
      "balance_loss_mlp": 1.01559019,
      "epoch": 0.9964528347261468,
      "flos": 33108488991360.0,
      "grad_norm": 1.9394108053929289,
      "language_loss": 0.6324122,
      "learning_rate": 1.2757767657989305e-10,
      "loss": 0.65415227,
      "num_input_tokens_seen": 178861515,
      "step": 8287,
      "time_per_iteration": 2.6845993995666504
    },
    {
      "auxiliary_loss_clip": 0.01149889,
      "auxiliary_loss_mlp": 0.01022652,
      "balance_loss_clip": 1.04450202,
      "balance_loss_mlp": 1.01559472,
      "epoch": 0.9965730776167859,
      "flos": 23109342589440.0,
      "grad_norm": 2.0979292953952147,
      "language_loss": 0.87196302,
      "learning_rate": 1.1893099941850948e-10,
      "loss": 0.89368844,
      "num_input_tokens_seen": 178880410,
      "step": 8288,
      "time_per_iteration": 2.599811315536499
    },
    {
      "auxiliary_loss_clip": 0.01137267,
      "auxiliary_loss_mlp": 0.01028236,
      "balance_loss_clip": 1.03967619,
      "balance_loss_mlp": 1.02072561,
      "epoch": 0.996693320507425,
      "flos": 22965843755520.0,
      "grad_norm": 2.151922603428805,
      "language_loss": 0.77658784,
      "learning_rate": 1.105877026252866e-10,
      "loss": 0.79824287,
      "num_input_tokens_seen": 178898740,
      "step": 8289,
      "time_per_iteration": 2.658766508102417
    },
    {
      "auxiliary_loss_clip": 0.01164469,
      "auxiliary_loss_mlp": 0.01025277,
      "balance_loss_clip": 1.04340112,
      "balance_loss_mlp": 1.01776123,
      "epoch": 0.996813563398064,
      "flos": 13222740476160.0,
      "grad_norm": 1.9815175837354986,
      "language_loss": 0.72699326,
      "learning_rate": 1.0254778746565663e-10,
      "loss": 0.74889076,
      "num_input_tokens_seen": 178914015,
      "step": 8290,
      "time_per_iteration": 2.5652120113372803
    },
    {
      "auxiliary_loss_clip": 0.0111918,
      "auxiliary_loss_mlp": 0.01029394,
      "balance_loss_clip": 1.0405736,
      "balance_loss_mlp": 1.02196121,
      "epoch": 0.9969338062887032,
      "flos": 14647855553280.0,
      "grad_norm": 2.105536251175274,
      "language_loss": 0.73626292,
      "learning_rate": 9.481125515953259e-11,
      "loss": 0.7577486,
      "num_input_tokens_seen": 178932075,
      "step": 8291,
      "time_per_iteration": 2.8078479766845703
    },
    {
      "auxiliary_loss_clip": 0.01108911,
      "auxiliary_loss_mlp": 0.01025409,
      "balance_loss_clip": 1.03675628,
      "balance_loss_mlp": 1.01853108,
      "epoch": 0.9970540491793423,
      "flos": 25735741142400.0,
      "grad_norm": 1.9299218039496595,
      "language_loss": 0.79806781,
      "learning_rate": 8.737810688064228e-11,
      "loss": 0.81941104,
      "num_input_tokens_seen": 178951910,
      "step": 8292,
      "time_per_iteration": 2.782075881958008
    },
    {
      "auxiliary_loss_clip": 0.01116829,
      "auxiliary_loss_mlp": 0.01028204,
      "balance_loss_clip": 1.04110861,
      "balance_loss_mlp": 1.02002621,
      "epoch": 0.9971742920699813,
      "flos": 21470236237440.0,
      "grad_norm": 1.7882314026611343,
      "language_loss": 0.78994077,
      "learning_rate": 8.024834375608414e-11,
      "loss": 0.81139112,
      "num_input_tokens_seen": 178970500,
      "step": 8293,
      "time_per_iteration": 2.7414486408233643
    },
    {
      "auxiliary_loss_clip": 0.01062894,
      "auxiliary_loss_mlp": 0.01000819,
      "balance_loss_clip": 1.00918603,
      "balance_loss_mlp": 0.99976367,
      "epoch": 0.9972945349606205,
      "flos": 72211223629440.0,
      "grad_norm": 0.8182221056166156,
      "language_loss": 0.62845063,
      "learning_rate": 7.342196686788149e-11,
      "loss": 0.64908773,
      "num_input_tokens_seen": 179023665,
      "step": 8294,
      "time_per_iteration": 3.9600319862365723
    },
    {
      "auxiliary_loss_clip": 0.01136493,
      "auxiliary_loss_mlp": 0.01023347,
      "balance_loss_clip": 1.04620838,
      "balance_loss_mlp": 1.01611125,
      "epoch": 0.9974147778512595,
      "flos": 19678293515520.0,
      "grad_norm": 2.60315249095921,
      "language_loss": 0.68777084,
      "learning_rate": 6.689897725142834e-11,
      "loss": 0.7093693,
      "num_input_tokens_seen": 179043140,
      "step": 8295,
      "time_per_iteration": 2.603347063064575
    },
    {
      "auxiliary_loss_clip": 0.01138198,
      "auxiliary_loss_mlp": 0.01026665,
      "balance_loss_clip": 1.04162478,
      "balance_loss_mlp": 1.01961422,
      "epoch": 0.9975350207418986,
      "flos": 15960821391360.0,
      "grad_norm": 2.693479302201775,
      "language_loss": 0.88517284,
      "learning_rate": 6.067937589615545e-11,
      "loss": 0.90682155,
      "num_input_tokens_seen": 179061215,
      "step": 8296,
      "time_per_iteration": 3.5922701358795166
    },
    {
      "auxiliary_loss_clip": 0.01045999,
      "auxiliary_loss_mlp": 0.01000593,
      "balance_loss_clip": 1.00960279,
      "balance_loss_mlp": 0.99960977,
      "epoch": 0.9976552636325378,
      "flos": 59961879768960.0,
      "grad_norm": 0.7589687764242375,
      "language_loss": 0.57651407,
      "learning_rate": 5.476316374575241e-11,
      "loss": 0.59697998,
      "num_input_tokens_seen": 179124700,
      "step": 8297,
      "time_per_iteration": 3.214306116104126
    },
    {
      "auxiliary_loss_clip": 0.01167495,
      "auxiliary_loss_mlp": 0.01029731,
      "balance_loss_clip": 1.04759169,
      "balance_loss_mlp": 1.02249527,
      "epoch": 0.9977755065231768,
      "flos": 22487872452480.0,
      "grad_norm": 5.18924431652778,
      "language_loss": 0.72806275,
      "learning_rate": 4.9150341697723476e-11,
      "loss": 0.75003505,
      "num_input_tokens_seen": 179144590,
      "step": 8298,
      "time_per_iteration": 2.5853397846221924
    },
    {
      "auxiliary_loss_clip": 0.01132305,
      "auxiliary_loss_mlp": 0.01027638,
      "balance_loss_clip": 1.04225993,
      "balance_loss_mlp": 1.02072072,
      "epoch": 0.9978957494138159,
      "flos": 26030280666240.0,
      "grad_norm": 1.5466396544872563,
      "language_loss": 0.66642213,
      "learning_rate": 4.384091060338768e-11,
      "loss": 0.68802148,
      "num_input_tokens_seen": 179165060,
      "step": 8299,
      "time_per_iteration": 2.6678006649017334
    },
    {
      "auxiliary_loss_clip": 0.01150946,
      "auxiliary_loss_mlp": 0.01022862,
      "balance_loss_clip": 1.04481161,
      "balance_loss_mlp": 1.01542985,
      "epoch": 0.998015992304455,
      "flos": 22637835734400.0,
      "grad_norm": 2.206273950085483,
      "language_loss": 0.73731577,
      "learning_rate": 3.883487126810081e-11,
      "loss": 0.75905383,
      "num_input_tokens_seen": 179184320,
      "step": 8300,
      "time_per_iteration": 3.553802728652954
    },
    {
      "auxiliary_loss_clip": 0.01139989,
      "auxiliary_loss_mlp": 0.01024446,
      "balance_loss_clip": 1.03999925,
      "balance_loss_mlp": 1.01755309,
      "epoch": 0.9981362351950941,
      "flos": 18223444955520.0,
      "grad_norm": 2.294407814691813,
      "language_loss": 0.79331797,
      "learning_rate": 3.41322244516995e-11,
      "loss": 0.81496233,
      "num_input_tokens_seen": 179202265,
      "step": 8301,
      "time_per_iteration": 2.6677961349487305
    },
    {
      "auxiliary_loss_clip": 0.01097219,
      "auxiliary_loss_mlp": 0.01027008,
      "balance_loss_clip": 1.03966343,
      "balance_loss_mlp": 1.01999211,
      "epoch": 0.9982564780857331,
      "flos": 33474095573760.0,
      "grad_norm": 1.743485349939381,
      "language_loss": 0.62937844,
      "learning_rate": 2.9732970866946925e-11,
      "loss": 0.6506207,
      "num_input_tokens_seen": 179222145,
      "step": 8302,
      "time_per_iteration": 3.750641345977783
    },
    {
      "auxiliary_loss_clip": 0.01111337,
      "auxiliary_loss_mlp": 0.01026263,
      "balance_loss_clip": 1.03765273,
      "balance_loss_mlp": 1.01869941,
      "epoch": 0.9983767209763723,
      "flos": 15523465392000.0,
      "grad_norm": 5.227445713543878,
      "language_loss": 0.78405058,
      "learning_rate": 2.563711118175327e-11,
      "loss": 0.80542654,
      "num_input_tokens_seen": 179239030,
      "step": 8303,
      "time_per_iteration": 2.7135982513427734
    },
    {
      "auxiliary_loss_clip": 0.0111903,
      "auxiliary_loss_mlp": 0.01024815,
      "balance_loss_clip": 1.04182625,
      "balance_loss_mlp": 1.01765084,
      "epoch": 0.9984969638670114,
      "flos": 19974377324160.0,
      "grad_norm": 2.084235499593538,
      "language_loss": 0.83496726,
      "learning_rate": 2.184464601717728e-11,
      "loss": 0.85640568,
      "num_input_tokens_seen": 179257345,
      "step": 8304,
      "time_per_iteration": 2.7523605823516846
    },
    {
      "auxiliary_loss_clip": 0.01155408,
      "auxiliary_loss_mlp": 0.01023053,
      "balance_loss_clip": 1.04727757,
      "balance_loss_mlp": 1.01587677,
      "epoch": 0.9986172067576504,
      "flos": 20375750874240.0,
      "grad_norm": 2.087332515579905,
      "language_loss": 0.77609795,
      "learning_rate": 1.8355575948758585e-11,
      "loss": 0.79788256,
      "num_input_tokens_seen": 179275330,
      "step": 8305,
      "time_per_iteration": 2.6404216289520264
    },
    {
      "auxiliary_loss_clip": 0.01134873,
      "auxiliary_loss_mlp": 0.01029474,
      "balance_loss_clip": 1.03917444,
      "balance_loss_mlp": 1.02208281,
      "epoch": 0.9987374496482896,
      "flos": 23727903724800.0,
      "grad_norm": 2.057179884580709,
      "language_loss": 0.73321819,
      "learning_rate": 1.5169901505407424e-11,
      "loss": 0.75486159,
      "num_input_tokens_seen": 179292395,
      "step": 8306,
      "time_per_iteration": 2.709988594055176
    },
    {
      "auxiliary_loss_clip": 0.01134793,
      "auxiliary_loss_mlp": 0.01025063,
      "balance_loss_clip": 1.04351819,
      "balance_loss_mlp": 1.0186615,
      "epoch": 0.9988576925389286,
      "flos": 25044029959680.0,
      "grad_norm": 1.7783341340420615,
      "language_loss": 0.7435323,
      "learning_rate": 1.228762317073695e-11,
      "loss": 0.76513088,
      "num_input_tokens_seen": 179311225,
      "step": 8307,
      "time_per_iteration": 2.68908429145813
    },
    {
      "auxiliary_loss_clip": 0.01134799,
      "auxiliary_loss_mlp": 0.01024972,
      "balance_loss_clip": 1.04223454,
      "balance_loss_mlp": 1.01792383,
      "epoch": 0.9989779354295677,
      "flos": 31285627637760.0,
      "grad_norm": 2.4063371243067944,
      "language_loss": 0.78925514,
      "learning_rate": 9.70874138195299e-12,
      "loss": 0.81085289,
      "num_input_tokens_seen": 179333135,
      "step": 8308,
      "time_per_iteration": 2.731342077255249
    },
    {
      "auxiliary_loss_clip": 0.01165348,
      "auxiliary_loss_mlp": 0.01024565,
      "balance_loss_clip": 1.04413891,
      "balance_loss_mlp": 1.01745081,
      "epoch": 0.9990981783202069,
      "flos": 19573398823680.0,
      "grad_norm": 2.6571083618068063,
      "language_loss": 0.74686813,
      "learning_rate": 7.433256530076093e-12,
      "loss": 0.7687673,
      "num_input_tokens_seen": 179353090,
      "step": 8309,
      "time_per_iteration": 2.6491377353668213
    },
    {
      "auxiliary_loss_clip": 0.01112605,
      "auxiliary_loss_mlp": 0.01021324,
      "balance_loss_clip": 1.03746581,
      "balance_loss_mlp": 1.01456499,
      "epoch": 0.9992184212108459,
      "flos": 17199667514880.0,
      "grad_norm": 2.270668537134058,
      "language_loss": 0.75949287,
      "learning_rate": 5.46116896038562e-12,
      "loss": 0.78083217,
      "num_input_tokens_seen": 179367500,
      "step": 8310,
      "time_per_iteration": 2.757056474685669
    },
    {
      "auxiliary_loss_clip": 0.01132805,
      "auxiliary_loss_mlp": 0.01024665,
      "balance_loss_clip": 1.04198837,
      "balance_loss_mlp": 1.01758945,
      "epoch": 0.999338664101485,
      "flos": 46497853681920.0,
      "grad_norm": 2.1675303034542273,
      "language_loss": 0.61902583,
      "learning_rate": 3.792478972197699e-12,
      "loss": 0.64060056,
      "num_input_tokens_seen": 179388085,
      "step": 8311,
      "time_per_iteration": 2.8514411449432373
    },
    {
      "auxiliary_loss_clip": 0.01163891,
      "auxiliary_loss_mlp": 0.0102596,
      "balance_loss_clip": 1.04436779,
      "balance_loss_mlp": 1.0191381,
      "epoch": 0.9994589069921241,
      "flos": 15158253859200.0,
      "grad_norm": 3.038185337071505,
      "language_loss": 0.700418,
      "learning_rate": 2.4271868181990895e-12,
      "loss": 0.72231644,
      "num_input_tokens_seen": 179405250,
      "step": 8312,
      "time_per_iteration": 2.6360507011413574
    },
    {
      "auxiliary_loss_clip": 0.01150898,
      "auxiliary_loss_mlp": 0.01026268,
      "balance_loss_clip": 1.04265261,
      "balance_loss_mlp": 1.01912773,
      "epoch": 0.9995791498827632,
      "flos": 12531460256640.0,
      "grad_norm": 2.1595509165746822,
      "language_loss": 0.80993664,
      "learning_rate": 1.3652927060014973e-12,
      "loss": 0.83170831,
      "num_input_tokens_seen": 179420845,
      "step": 8313,
      "time_per_iteration": 2.5778400897979736
    },
    {
      "auxiliary_loss_clip": 0.01124644,
      "auxiliary_loss_mlp": 0.0102465,
      "balance_loss_clip": 1.04139256,
      "balance_loss_mlp": 1.01688373,
      "epoch": 0.9996993927734023,
      "flos": 19245175320960.0,
      "grad_norm": 2.460120528895731,
      "language_loss": 0.63693625,
      "learning_rate": 6.067967965872612e-13,
      "loss": 0.65842915,
      "num_input_tokens_seen": 179440455,
      "step": 8314,
      "time_per_iteration": 2.6499428749084473
    },
    {
      "auxiliary_loss_clip": 0.01124662,
      "auxiliary_loss_mlp": 0.0102848,
      "balance_loss_clip": 1.04374111,
      "balance_loss_mlp": 1.02146149,
      "epoch": 0.9998196356640414,
      "flos": 62952804518400.0,
      "grad_norm": 1.5221719591873728,
      "language_loss": 0.77016234,
      "learning_rate": 1.5169920497548615e-13,
      "loss": 0.79169381,
      "num_input_tokens_seen": 179465075,
      "step": 8315,
      "time_per_iteration": 3.0508453845977783
    },
    {
      "auxiliary_loss_clip": 0.01105398,
      "auxiliary_loss_mlp": 0.01012772,
      "balance_loss_clip": 1.02576685,
      "balance_loss_mlp": 1.00846672,
      "epoch": 0.9999398785546805,
      "flos": 50922375073920.0,
      "grad_norm": 1.1015050813165752,
      "language_loss": 0.55057263,
      "learning_rate": 0.0,
      "loss": 0.57175434,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "time_per_iteration": 3.2250914573669434
    },
    {
      "epoch": 0.9999398785546805,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "total_flos": 6.996749092776837e+17,
      "train_loss": 0.7897025029810648,
      "train_runtime": 25272.1878,
      "train_samples_per_second": 13.163,
      "train_steps_per_second": 0.329
    }
  ],
  "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
}